Cryptococcus neoformans Strains Preferentially Disseminate to the Central Nervous System during Coinfection
http://www.100md.com
感染与免疫杂志 2005年第8期
Departments of Molecular Genetics and Microbiology Cell Biology Medicine Pediatrics Pharmacology and Cancer Biology Howard Hughes Medical Institute, Duke University Medical Center, Durham, North Carolina 27710
Divisions of Infectious Diseases Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, New York 10461
Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts 02114
ABSTRACT
Cryptococcus neoformans is a fungal pathogen that has evolved over the past 40 million years into three distinct varieties or sibling species (gattii, grubii, and neoformans). Each variety manifests differences in epidemiology and disease, and var. grubii strains are responsible for the vast majority of human disease. In previous studies, strains were more virulent than congenic a strains in var. neoformans, whereas var. grubii congenic a and strains exhibited equivalent levels of virulence. Here the role of mating type in the virulence of var. grubii was further characterized in a panel of model systems. Congenic var. grubii a and strains had equivalent survival rates when cultured with amoebae, nematodes, and macrophages. No difference in virulence was observed between a and congenic strains in multiple inbred-mouse genetic backgrounds, and there was no difference in accumulations in the central nervous system (CNS) late in infection. In contrast, during coinfections, a and strains are equivalent in peripheral tissues but cells have an enhanced predilection to penetrate the CNS. These studies reveal the first virulence difference between congenic a and strains in the most common pathogenic variety and suggest an explanation for the prevalence of strains in clinical isolates.
INTRODUCTION
Over the past decade fungal diseases have increased due to an expanding population of immunocompromised individuals. Rates of human immunodeficiency virus (HIV) infections are rising in many areas of the world, and the regions with the highest AIDS burden often have the least medical resources, resulting in large immunocompromised populations. At the end of 2002, it was estimated that 29 million people in Africa were living with HIV and that 3 million people died of AIDS-related illnesses during 2002 (76). This phenomenon, combined with the development of aggressive surgeries, organ transplantation, and immunosuppressive therapies, has led to an increase in the prevalence of invasive fungal diseases. Fungal diseases, such as oropharyngeal candidiasis and cryptococcal meningitis, are often AIDS-defining illnesses; aspergillosis has emerged as an important cause of mortality following bone marrow transplants; and Candida species are now a leading cause of bloodstream infections in hospitalized patients (24). As fungal diseases become more prevalent, understanding the microbial characteristics responsible for virulence becomes increasingly important.
The basidiomycete Cryptococcus neoformans is both an important pathogen and an excellent model system for fungal virulence (37). C. neoformans infects the central nervous system (CNS) to cause meningoencephalitis that is uniformly fatal if untreated (1). Cryptococcal meningitis occurs in 6 to 10% of AIDS patients in the United States and is the first AIDS-defining illness in 40% of these patients (19, 31, 52, 62). In sub-Saharan Africa, where highly active antiretroviral therapy is less common, C. neoformans accounts for 45% of all cases of meningitis in adults; many individuals present with cryptococcosis as their first AIDS-defining illness, and the mean survival rate after diagnosis can be as short as 4 days (29, 32, 35, 50). Cryptococcus occurs in three varieties or species that can be distinguished based on genetic and phenotypic differences. While these varieties are pathogenic in humans, they have different disease epidemiologies. C. neoformans var. grubii (serotype A) strains cause 95% of all C. neoformans infections and >99% of infections in AIDS patients in the United States (10). However, C. neoformans var. neoformans (serotype D) strains can account for up to 20% of clinical isolates in Europe (69). Unlike the first two varieties of C. neoformans, which cause disease predominantly in individuals with impaired immunity, strains of the sibling species—C. gattii (serotypes B and C)—can infect immunocompetent individuals (63).
C. neoformans is an important human pathogen and has also proven to be an outstanding model system for studying the virulence of fungi. Cryptococcus is predominantly haploid, which facilitates genetic analyses, and has a defined sexual cycle that enables classical genetic studies. Genes can readily be disrupted by biolistic transformation and homologous recombination, and essential genes have been examined using stable diploid strains, conditional alleles, and RNA interference (20, 37, 47, 49, 72). C. neoformans is a facultative intracellular pathogen that can also infect the amoeba Acanthamoeba castellanii, the nematode Caenorhabditis elegans, the slime mold Dictyostelium discoideum, and the insect Drosophila melanogaster, allowing identification of genes involved in virulence in heterologous hosts (2, 53, 64-66). Finally, robust animal model systems that enable detailed studies of virulence characteristics, such as the polysaccharide capsule and the antioxidant pigment melanin, have been developed.
Mating type has been proposed to be a virulence factor in C. neoformans based on disease epidemiology and animal studies. The majority of clinical isolates are of the mating type (10, 41). Early experimental work on the role of mating type in virulence was conducted with serotype D var. neoformans because of the availability of congenic strains (42; reviewed in reference 34). In serotype D, a comparison of the levels of virulence of nonisogenic, unrelated a and strains revealed that strains are more virulent at lower inoculum levels than a strains (40, 78) and that the strain JEC21 was more virulent than the congenic a strain JEC20 in a murine tail vein infection model of cryptococcosis as well as in the C. elegans system (42, 54). In serotype A var. grubii, mating type a strains are exceptionally rare and have only recently been identified (39, 44, 46, 55, 73, 74). As with serotype D, when the virulence of nonisogenic a and strains was compared, the a strains were less virulent than the strains (39, 44, 55). However, when congenic serotype A strains in the highly virulent H99 background were compared in standard mouse inhalation and rabbit meningitis models of cryptococcosis, no difference in virulence was measured (55).
Here we characterize in detail the role of mating type in pathogenicity by examining the virulence of serotype A var. grubii a and congenic strains at various stages in the infective cycle—from their survival in potential environmental predators to their colonization of various organs in animal models—and found no difference between a and strains. Infection by multiple strains has been reported, and studies examining sequential isolates have shown that different strains can be isolated from the same patient (8, 9, 33, 38, 58, 68). These findings suggest that coinfection with strains of opposite mating types may occur. We therefore analyzed the virulence of the congenic serotype A var. grubii strains during coinfection with both mating types and discovered that a and congenic strains reached equivalent levels in the lungs and spleen but that a significantly higher proportion of cells infected the brain. Intracerebral coinfections revealed that the strains have equivalent levels of persistence in the CNS. Thus, the strain out-competes the a strain in entry into the CNS as a consequence of either fungal-fungal or fungal-host interactions.
MATERIALS AND METHODS
Strains and media. C. neoformans var. grubii strains H99 (70), KN99a (55), KN99 (55), KN99aNAT, and KN99NAT were maintained at –80°C. Cultures were initiated by inoculating yeast extract-peptone-dextrose (YPD) solid medium and incubating the plates for 2 days at 30°C, followed by growth in YPD liquid medium at 30°C overnight. Cells were harvested, washed three times with phosphate-buffered saline (PBS), and resuspended at the appropriate concentration in PBS. Acanthamoeba castellanii strain 30324 (ATCC) was maintained as previously described (51). Caenorhabditis elegans strain N2 Bristol was maintained at 15°C and propagated on Escherichia coli strain OP50 (53). The MH-S murine alveolar macrophage cell line (ATCC) was maintained in Dulbecco modified Eagle complete medium at 37°C with 5% CO2 and transferred to fresh medium every 3 to 5 days.
AFLP. Amplified fragment length polymorphisms (AFLPs) were generated and analyzed as previously described (46). Two different EcoRI primer combinations were used for the selective PCR. Both primer sequences were labeled at their 5' ends with 6-carboxyfluorescein (6FAM) but differed in the final two bases at the 3' ends of the primers. The primer sequences were as follows: primer 1, 5'-6FAM-GACTGCGTACCAATTCAC-3', and primer 2, 5'-6FAM-GACTGCGTACCAATTCTG-3'. Either primer 1 or primer 2 was combined with the MseI primer (5'-GATGAGTCCTGAGTAAG-3') in the selective PCR. The AFLP reaction and analysis were performed on at least two separate occasions for each strain. Only intense and reproducible bands were scored to identify differences between strains.
Phenotype microarrays. Phenotype microarrays were done as described previously (80) and according to instructions from the manufacturer (Biolog, Hayward, CA) with the following exceptions. Strain cultures were grown overnight in YPD liquid medium; the cells were then collected by centrifugation, washed once with sterile distilled H2O, resuspended in 12x yeast nitrogen base liquid medium supplemented with histidine, leucine, and uracil, and adjusted to 18% transmittance (T). IFY-0 was the inoculating fluid for all 20 96-well plate phenotype microarrays (PM1 to PM10 and PM21 to PM30). IFY-0 was supplemented with 5 mg/ml ammonium sulfate, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM1 and PM2; 20 mg/ml D-glucose, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM3 and PM6 to PM8; 20 mg/ml D-glucose and 1 mg/ml L-glutamate for PM4; or 20 mg/ml D-glucose, 5 mg/ml ammonium sulfate, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM9, PM10, and PM21 to PM30. Cells were combined with appropriately supplemented IFY-0 at a ratio of 1:11, and 100 μl of this mixture was added to each well. All phenotype microarrays were incubated at 30°C for 36 h in an OmniLog and monitored for color change in the wells every 15 min. Kinetic data were analyzed with OmniLog software using pairwise combinations.
Generation of KN99aNAT and KN99NAT strains. Strains KN99aNAT and KN99NAT were generated using overlap PCR by following the methods described in reference 27. The nourseothricin transgene (NAT) was inserted into the largest intergenic region in the corresponding MAT locus, where it would be anticipated to have the least effect on surrounding genes. For the serotype A MATa locus (43), the NAT transgene was inserted in a 12-kb region between the RPL39 and PRT1 genes. For the serotype A MAT locus, the transgene was inserted in a 4-kb region between the BSP1 and RPL39 genes. PCR was used to generate the flanking 5' (with KN99a, JOHE12926 and JOHE12927; with KN99, JOHE12921 and JOHE12922) and 3' (with KN99a, JOHE12928 and JOHE12929; with KN99, JOHE12923 and JOHE12924) regions containing linkers to a NATr cassette. The NATr cassette containing the nourseothricin transgene was amplified using oligonucleotides JOHE8733 and JOHE8738. The products were combined in a PCR overlap reaction (with KN99a, JOHE12926 and JOHE12929; with KN99, JOHE12921 and JOHE12924) to yield the NAT insertion allele. KN99a or KN99 was then transformed with the appropriate NAT insertion allele by biolistic transformation and selection for colonies resistant to nourseothricin (100 μg/ml). Transformants were identified by Southern blotting of KpnI-digested DNA using the 5' fragment as a probe. Primer sequences were as follows: for JOHE8733, GCTGCGAGGATGTGAGCTGGA; for JOHE8738, GGTTTATCTGTATTAACACGG; for JOHE12921, GGTGAGTTGAGTGCGGCTATT; for JOHE12922, AGCTCACATCCTCGCAGCGCATTTCAATTCTGTTAT (NATr linker is in bold); for JOHE12923, CCGTGTTAATACAGATAAACCCACATTGACTACTTATGACCA; for JOHE12924, GCTATAAACATTCCCGTCGTG; for JOHE12926, GACTCTGCCATCGTCCAAGAA; for JOHE12927, AGCTCACATCCTCGCAGCCAGTTTACCTGGATTTCGAGC; for JOHE12928, CCGTGTTAATACAGATAACCCACTTCTGTTATGCGAGATGC; and for JOHE12929, CCACCGCTCTTCCAGATGAAA.
Nematode killing assay. C. elegans killing assays were done as described previously (53, 54) with minor modifications. The C. neoformans strains were inoculated into 2 ml YPD and grown at 30°C for 24 h, 10 μl of the culture was spread on brain heart infusion agar (Difco), and the plates were incubated at 30°C for 48 h. C. elegans animals were transferred to the plates (divided across three plates of each strain) and incubated at 25°C, and then 120 to 150 animals were examined for viability at 24-h intervals. Replicate experiments showed similar results.
Amoeba assay. The survival of C. neoformans var. grubii in amoebae was as described previously (51). Briefly, A. castellanii organisms were collected, washed twice with PBS, suspended in PBS, counted with a hemocytometer, and diluted in PBS to the appropriate density. A. castellanii cells were added to a 96-well tissue culture plate at 1 x 104 cells per well and allowed to adhere for 1 h at 28°C. C. neoformans cells (1 x 104) were added to wells containing amoebae or control wells containing PBS alone, and the plates were incubated at 28°C. At 0, 24, and 48 h, the amoebae were lysed and the well contents were serially diluted and plated onto Sabouraud (SAB) dextrose medium. Numbers of CFU were determined after incubation overnight at 30°C. Replicate experiments showed similar results.
Macrophage assay. Macrophages were harvested by removal from monolayers, and cell viability was determined by trypan blue exclusion. Cell number was determined using a hemocytometer, and the macrophage density was adjusted to 5 x 105 cells/ml of culture medium. Macrophages were placed in 96-well plates and stimulated overnight with 50 units/ml gamma interferon (murine) and 0.3 mg/ml lipopolysaccharide (LPS). C. neoformans cells were added to the macrophage suspension at a density of 5 x 105 cells/ml of culture medium with 10 μg/ml of monoclonal antibody 1887 as an opsonin. Macrophage-yeast mixtures were incubated for 1 h and washed three times with culture medium. The macrophages were then either lysed with 0.05% sodium dodecyl sulfate (SDS) immediately to determine the phagocytic index or incubated with stimulation for 24 h at 37°C with 5% CO2. After incubation, the culture medium was aspirated from each well and the macrophages were lysed with two exchanges of 0.05% SDS. The aspirated medium and SDS solutions were combined, serially diluted, and plated onto YPD plates. Numbers of CFU were determined after incubation overnight at 30°C. Replicate experiments showed similar results.
Mouse virulence studies. Groups of 4- to 8-week-old female A/JCr, BALB/c, or C57BL/6 mice (five mice per strain) were anesthetized by intraperitoneal phenobarbital injection. Animals were infected intranasally with 5 x 104 fungal cells in 50 μl of normal saline pipetted slowly into the nares. Animals were infected intracerebrally with 1.5 x 103 fungal cells in 20 μl of normal saline injected slowly into the top of the cerebrum. In further experiments, animals were directly injected in their lateral tail veins with 2.5 x 104 cells. The concentration of yeast cells in the inoculum was confirmed by plating serial dilutions and enumerating CFU. Mice were monitored twice daily, and those that showed signs of severe morbidity (weight loss, extension of the cerebral portion of the cranium, abnormal gait, paralysis, seizures, convulsions, or coma) were sacrificed by CO2 inhalation.
Quantitative analysis of tissue fungal burden. Groups of 4- to 6-week-old female A/JCr mice were infected either intranasally or intracerebrally. For intranasal infections, animals (five mice per strain per time interval) were sacrificed by CO2 inhalation at 7, 14, and 21 days postinfection, and lung, spleen, and brain were removed, weighed, and homogenized in 2 ml sterile PBS. Animals were sacrificed by CO2 inhalation at 4 days postinfection for intracerebral infections (10 mice per strain), and brain and lung were removed, weighed, and homogenized in 2 ml sterile PBS. Serial dilutions of the organ samples were plated on SAB agar plates containing 100 μg/ml chloramphenicol and incubated at 30°C overnight. Colony counts were performed and adjusted to reflect the total number of CFU/g tissue.
Coinfection experiments. Groups of 4- to 6-week-old female A/JCr mice were coinfected with a 1:1 ratio of a and cells either intranasally or intracerebrally. The inoculum was prepared immediately before animals were infected and then plated on SAB medium. Colonies were isolated and analyzed to determine the proportion of a cells present in the inoculum. Mice were sacrificed by CO2 inhalation; the lung, spleen, and brain were harvested and homogenized; and CFU were enumerated as in the previous assay. For the first coinfection experiments (five mice per time interval), animals were coinfected with KN99a and H99. Mice were sacrificed at 7, 14, and 21 days postinfection, and single colonies (64 from inoculum, 32 from lungs and spleen for each animal, and 64 from the brain of each animal) were randomly selected from SAB plates and analyzed for mating type based on mating with reference a and strains on V8 medium (55). Coinfections with KN99a/KN99NAT, KN99aNAT/KN99, KN99a/KN99aNAT, and KN99/KN99NAT used 10, 10, 5, and 5 mice, respectively, which were sacrificed at 21 days postinfection. More than 500 colonies from the lung, spleen, and brain from each animal were isolated on SAB plates and screened for NAT resistance on yeast peptone-dextrose plates containing 100 μg/ml nourseothricin to determine mating type. In control experiments, NAT resistance was found to be 100% stable during vegetative culture of KN99aNAT and KN99NAT. A subset of the colonies from coinfections was also analyzed for mating type based on mating with reference a and strains, and nourseothricin resistance cosegregated 100% with the appropriate a or allele of MAT. No instances of NAT loss or inactivation during infection were observed. The measured numbers of a and cells were compared to the expected numbers, assuming that both strains remained at the initial infection proportions (based on analysis of the initial inoculum). These raw data were then converted to percentages for presentation in Fig. 5 and 6.
Statistics. P values presented are two tailed. P values for the nematode and mouse survival data were derived from log rank tests. Wilcoxon rank sum analysis was used to determine significance for the amoeba, macrophage, and organ CFU data. For the coinfection experiments, the measured number of a and cells was compared to the expected number, assuming that both strains remained at the initial infection proportions (based on analysis of the initial inoculum), and Wilcoxon rank sum analysis of the raw data was used to determine P values. Thus, for the coinfections, the test of significance compared the ratios of the observed strains to the initial inoculum.
RESULTS
Further evidence that strains KN99a and KN99 are congenic with H99. We recently characterized mating in C. neoformans var. grubii using the serotype A mating type a strain 125.91 (55). A backcrossing scheme was designed using the mating characteristics of strain 125.91 to develop serotype A congenic strains in the H99 genetic background (Fig. 1A) (55). The resulting congenic strains KN99a and KN99 were compared to H99. The karyotypes of the KN99a and KN99 strains were examined and found to be identical to each other and to H99, indicating that the strains are congenic at the chromosome level (55). Restriction fragment length polymorphisms at two genetic loci were identified in the parental strains and the congenic strains and were found to have the same gene allele at both loci (55).
To further characterize the congenic strains at a molecular level, AFLP analysis with two independent primer pairs was used to create AFLP genotypes for each strain (Fig. 1B). Comparison of the AFLP genotypes of the parental strains 125.91, 8-1, KNA14, and H99 revealed polymorphic amplified DNA fragments between the strains. The congenic strains KN99a and KN99 had identical AFLP genotypes and were identical with H99 with both primer pairs tested, indicating a congenic relationship at a higher-resolution genomic level.
KN99a, KN99, and H99 showed no differences in several traits involved in virulence, such as the presence of melanin or capsule, prototrophy, or growth at 37°C (55). To examine the strains in more detail, we performed phenotype microarrays (4, 80; reviewed in reference 3). This analysis allows strain phenotypes to be quantitatively measured over time based on the physiological state of the cell. Cryptococcal cells are combined with a tetrazolium dye that absorbs electrons produced by yeast cell respiration. Reduction of the tetrazolium dye results in a color change that does not require growth of the cell. In the PM assay, >1,000 distinct culture conditions were selected to approximate a comprehensive scan of known cellular pathways. The effects of various carbon, nitrogen, phosphorus, and sulfur sources on the physiological state of the cell were examined. Cell respiration was also monitored in response to various osmotic conditions and in response to differing pHs. Finally, sensitivity to a wide range of chemicals, including antibiotics, respiratory inhibitors, and toxic metals, was examined. We have been able to observe distinct differences between mutant and wild-type C. neoformans strains using this method (X. Lin and J. Heitman, unpublished results). However, no consistent differences were observed with the serotype A var. grubii congenic strains, suggesting that KN99a, KN99, and H99 are also congenic at the phenotypic level.
Rates of survival of serotype A congenic strains are equivalent in nematodes and amoebae. The serotype A congenic strains KN99a and KN99 were previously shown to have virulence equivalent to that of strain H99 in two standard animal models of cryptococcosis (55). Because the vast majority of cases of cryptococcosis in humans are associated with mating type strains, we further characterized the potential role of mating type in the infectious process. C. neoformans is a soil-dwelling saprophyte that is not dependent on mammalian infection for reproduction or viability, yet many environmental isolates have the capacity to infect and cause disease in humans and experimental animals. Virulence is considered a complex phenotype that may be maintained as a result of selection. It has been hypothesized that factors that enable infection of mammals may have evolved to allow C. neoformans to survive predation by natural environmental predators, such as amoebae, nematodes, and insects, and the human virulence factors capsule and melanin have been shown to be important for the survival of C. neoformans in these organisms (2, 11, 53, 64, 66).
To determine whether mating type plays a role in the virulence of C. neoformans var. grubii strains in nematodes, we examined the survival of C. elegans exposed to KN99a, KN99, or H99. Previous studies with less-pathogenic serotype D var. neoformans congenic strains in C. elegans showed that the strain was more virulent than the congenic a strain (53). However, studies with the var. grubii congenic strains showed no significant difference in rates of survival of the nematodes (Fig. 2A) (P values were 0.20 for all comparisons). The var. grubii congenic strains were also examined for survival in amoebae. The congenic strains H99, KN99a, and KN99 were incubated with A. castellanii cells or a PBS control for 24 and 48 h. As shown in Fig. 2B, no significant difference in survival of the mating type strains H99 and KN99 was observed from that of the congenic mating type a strain KN99a in amoebae (P values were 0.49 for all comparisons).
Survival of serotype A congenic strains in macrophages. Humans are exposed to C. neoformans via inhalation of desiccated yeast cells or spores (67). Macrophages play a critical role in the early initial pulmonary immune response to cryptococcal infection and may also play a role in harboring the organism during latent infection as a facultative intracellular pathogen (5-7, 30, 45). Furthermore, the survival of Cryptococcus in macrophages has been linked to virulence (18, 22, 25, 26, 71). To study the role of mating type in infection and survival in macrophages, we examined the survival of the serotype A congenic strains KN99a and KN99 compared to that of H99 in the murine alveolar macrophage cell line MH-S and culture medium alone (Fig. 2C). Macrophages were stimulated with LPS and gamma interferon for 12 h and cocultured with the appropriate Cryptococcus strain for 1 hour, and afterwards, the macrophages were washed with culture medium to remove extracellular cryptococcal cells. After incubation for 24 h, the number of surviving cryptococcal cells, as measured by the number of CFU recovered, was examined for each strain. The serotype A congenic strains exhibited no difference in uptake by macrophages (data not shown) or survival in macrophages in vitro (P values were 0.4 for all comparisons).
Virulence of serotype A congenic strains in multiple murine models of cryptococcosis. To study the role of mating type in the virulence of the serotype A congenic strains, we compared the levels of virulence of the (H99) and a (KN99a) mating type strains in murine inbred lines, which can be divided into two groups: sensitive lines that are deficient in complement C5 secretion and resistant lines that have normal levels of complement C5 (59). Figure 3 shows three inbred lines commonly used for serotype A cryptococcal infections: A/JCr (sensitive), BALB/c (resistant), and C57BL/6 (resistant). A/JCr animals were most susceptible to intranasal infection, BALB/c animals exhibited intermediate susceptibility, and C57BL/6 animals were most resistant to infection by the serotype A congenic strains (P values were 0.009 for all comparisons). While there were significant differences in the susceptibilities of the strains to infection, all three inbred backgrounds showed no difference in levels of virulence for the mating type a and congenic strains (P values were 0.35 for all comparisons).
There are two predominant murine models of cryptococcosis based on inoculation via either inhalation or tail vein injection. Our previous studies showing that the serotype A congenic strains have equivalent levels of virulence utilized an inhalational model in which an initial pulmonary infection was followed by extrapulmonary dissemination, resulting in lethal CNS infection. Here we extended this comparison of the serotype A congenic strains H99 and KN99a using a mouse tail vein infection model. All animals succumbed to the infection in less than 10 days, and no significant difference in survival was observed between animals infected with the mating type strain H99 and those infected with the mating type a strain KN99a (data not shown).
Progression of C. neoformans var. grubii infection in a murine model. The previous animal models used survival (progression to severe morbidity) as the basis for the virulence of the serotype A congenic strains. While we found no difference in survival when mice were exposed to the serotype A congenic strains, we considered the possibility that the strains might cause disease in different ways. For example, one strain may result in a lethal pulmonary infection, while another strain may spread more quickly to the central nervous system to cause meningoencephalitis. Therefore, we examined the growth and dissemination of the strains in mice early in the infection (7 days), at an intermediate time point (14 days) and late in the infection (21 days). The number of CFU present in the lungs over time was used as a measure of pulmonary infection, while the number of CFU present in the spleen was used as a measure of the ability of the strain to spread hematogenously to other organs. Finally, numbers of CFU present in the brain were used to determine CNS infection (Fig. 4).
When the serotype A mating type strain H99 was compared to the congenic mating type a strain KN99a, H99 had slightly higher levels in the lung at the first time interval but no difference was observed in lung fungal burdens at the two later time intervals (P = 0.02 at 7 days, P = 0.15 at 14 days, P = 0.41 at 21 days). Large variability was observed between spleen fungal burdens over time, with KN99a having slightly higher levels at early and late time intervals (P = 0.01 at 7 days, P = 0.31 at 14 days, P = 0.02 at 21 days). Interestingly, higher levels of KN99a were present in the brain at the earliest time studied (7 days), but fungal burden in the brain increased at a higher rate, such that both strains had comparable numbers of brain CFU at the later times (P = 0.03 at 7 days, P = 0.31 at 14 days, P = 0.73 at 21 days). These data imply that a cells may initially disseminate to the brain faster than cells but that both cell types eventually produce equivalent levels of CNS infection. Based on symptoms, the animals appeared to have cryptococcal meningitis at the late time point and were without signs of a debilitating pulmonary infection. Because these observations correlate with equivalent CNS fungal burdens in the congenic a and strains, the animals appear to succumb to meningitis rather than pulmonary infection.
Coinfections with a and mating types. The serotype A congenic a and mating type strains have equivalent levels of virulence and similar modes of infection and disease progression when pure cultures are used as the infecting inoculum, yet strains cause the majority of cryptococcosis in humans. Interestingly, infection by multiple strains has been observed in humans (38, 58). The apparent paradox between animal studies and disease epidemiology could be explained if strains out-compete or inhibit a strains during simultaneous coinfection with both strains.
To examine this hypothesis, animals were infected with a mixture of the serotype A congenic a and strains and the proportions of each strain present in the lungs, spleen, and brain were monitored and compared to the proportions of each strain in the inoculum (Fig. 5 and 6). Infections with unmarked strains showed no difference in the proportions of the mating type a strain KN99a in the inoculum from the proportion of KN99a strains in the lungs and spleen in two independent coinfections at 7, 14, and 21 days postinfection (data not shown). Enumeration of total CFU from the lung and spleen were consistent with those observed in the individual infections (Fig. 4) at all time points. Interestingly, brain infection appeared to be initially delayed during coinfection with low or undetectable numbers of CFU at 7 days postinfection, but by days 14 and 21 postinfection, numbers of CFU increased and were consistent with those observed during pure culture infections (data not shown). At both 14 and 21 days postinfection, the proportions of KN99a strains in the brain were low in both assays. In 6 of the 10 mice examined, less than 5% of the strains analyzed were KN99a and >95% were H99, indicating a difference in CNS penetration or survival between the two mating types.
The strains analyzed in these coinfections had to be identified based on mating assay or PCR, which is laborious and limited the number of animals and strains analyzed. To extend these findings, large-scale analysis of coinfections was undertaken using marked strains and selective plating. A NAT resistance transgene was introduced into a noncoding region of the mating type locus of strains KN99a and KN99 to generate strains KN99aNAT and KN99NAT, respectively.
To verify that the transgenes were phenotypically neutral and did not alter the virulence of the NAT-marked strains, the NAT-marked strains were coinfected with their parent strain and the proportions of NAT-marked strains in the lung, spleen, and brain were analyzed at 21 days postinfection (Fig. 5B). The proportion of KN99aNAT cells present in the lungs, spleen, and brain in the aNAT/a mixture was equivalent to the starting inoculum (represented by the dotted line), demonstrating that KN99aNAT survives in vivo at levels equivalent to KN99a (lung, P = 0.75; spleen, P = 0.86; brain, P = 0.25). However, most animals had one strain type that predominated in the brain, with roughly half of the animals having a vast majority of marked aNAT cells and the other half having unmarked a cells in the CNS. This observation is consistent with a single cell or a small group of founder cells producing infection of the brain. Similar results were observed with NAT/ coinfections (data not shown). Thus, these data demonstrate that the dominant marker transgenes are phenotypically neutral in vivo.
We used the marked strains to compare differences in disease progression between a and using larger numbers of animals and CFU. The KN99aNAT strain was coinfected with KN99, and the proportions of NAT-marked strains in the lung, spleen, and brain were analyzed at 21 days postinfection. More than 500 colonies from each organ were isolated and then individually screened for NAT resistance by selective plating. Figure 5A shows that the proportions of KN99aNAT cells present in the lungs and spleen were equivalent to the initial inoculum (represented by the dotted line) but that the level of KN99aNAT cells present in the brain was significantly lower than the inoculum level in the aNAT/ coinfection (lung, P = 0.55; spleen, P = 0.21; brain, P < 0.001). Similar results were observed with the NAT/a coinfection (data not shown). Thus, this larger data set corroborates the pattern observed in coinfection studies with unmarked a and strains. We conclude that there is a significant difference between congenic a and cells with respect to their ability to establish CNS infection during coinfection of mice.
The difference between a and cell accumulation in the CNS during coinfection may be due to either dissemination to the CNS or growth within the CNS. To distinguish between these two possibilities, animals were infected intracerebrally with either KN99a or KN99 or coinfected with KN99aNAT/KN99. Intracerebral infections showed no difference in rates of survival of animals (data not shown) or in brain fungal burdens between a and pure cultures or with aNAT/ coinfections of the CNS (Fig. 6A) (P values were 0.81, 0.24, and 0.20 for KN99a/KN99, KN99a/KN99a coinfection, and KN99/KN99 coinfection, respectively) and are consistent with the intranasal infections. In contrast to what occurs with intranasal coinfections where cells predominate in the CNS, Fig. 6B shows that when aNAT/ cells were coinfected intracerebrally, no difference in the levels of accumulation of the strains was observed (P = 0.99). Analogous results were observed with a rabbit meningitis coinfection model where the strains were inoculated directly into the cerebrospinal fluid (data not shown). These data show that a and cells grow equally well in the CNS and imply that cells predominate in the CNS during coinfection due to a difference in dissemination to the CNS.
DISCUSSION
Rates of AIDS infections are still on the rise in many areas of the world, including Africa and Asia (76, 77). This pandemic has led to an increase in the abundance and severity of infections by opportunistic fungal pathogens. Many virulence factors of one such pathogen, C. neoformans, have been identified. Some factors, such as the polysaccharide capsule and melanin, are clearly directly linked to the pathogenicity of C. neoformans because cells lacking these factors exhibit a significant decrease in virulence (48, 56, 79). Mating type has been proposed to be a virulence factor in C. neoformans. Epidemiological studies have shown that clinical isolates are severely biased towards the mating type (41) and that some genes associated with mating type, such as STE12a and STE12, have a role in the virulence of the serotype D var. neoformans lineage, whereas other genes, such as SXI1, STE20, and STE11, do not (12, 14, 17, 21, 36, 75). In heterogeneous populations of serotype D, no clear distinction between the levels of virulence of a and strains could be determined, although strains killed more mice at lower inoculums (40). To examine the role of mating type in virulence, congenic strains were generated by repeated backcrossing. Analysis of var. neoformans serotype D strains revealed that the congenic strain was more virulent than the a strain (42). However, clinical infections with var. neoformans strains account for a limited number of cases compared to cases caused by var. grubii serotype A strains. In contrast to serotype D, congenic a and var. grubii strains have equivalent levels of virulence in cellular (macrophage), heterologous (nematodes, amoebae), and mammalian (both mouse and rabbit) models of cryptococcosis (reference 55 and this study).
Examination of the virulence of var. grubii serotype A congenic strains in multiple inbred-mouse lines showed virulence consistent with the var. neoformans serotype D results of Rhodes et al. (59). A/JCr mice were more susceptible to the serotype A congenic strains than the relatively resistant BALB/c and C57BL/6 lines. The serotype A congenic strains were also more virulent in the BALB/c inbred-mouse line than in C57BL/6 mice. While we observed a difference in the overall rates of survival between the inbred-mouse lines, the congenic strains had equivalent levels of virulence in all of the inbred lines tested, showing that the similarity in the virulence potentials of the var. grubii congenic strains is not specific to the inbred-animal background.
While cryptococcal meningoencephalitis is the most common life-threatening manifestation of cryptococcosis, the lungs are the second-most-relevant site of cryptococcal infection (10). In animal models of cryptococcal meningitis, neurological symptoms, such as extension of the cerebral portion of the cranium, abnormal gait, paralysis, and lethargy, are observed, whereas animals with pulmonary cryptococcosis exhibit respiratory distress. However, in animal models based on lethal infection, the difference between cryptococcal meningitis and pulmonary cryptococcosis could be overlooked. To differentiate between these two forms of cryptococcosis, numbers of CFU from lung, spleen, and brain tissue (to represent pulmonary infection, hematogenous spread to other organs, and CNS infection, respectively) were compared from animals inoculated with the serotype A congenic a and mating type strains KN99a and H99, respectively. Mice infected with KN99a had higher numbers of CFU at initial time intervals in the brain, suggesting either a higher rate of dissemination from the initial pulmonary infection to the CNS or reduced clearance in the CNS. However, both KN99a and H99 had equivalent numbers of CFU in the brain by the intermediate and final time intervals studied, and all animals exhibited symptoms of CNS infection, indicating that the basis of mortality is the same in both cases.
The most notable virulence difference between the serotype A congenic a and strains was observed during coinfection. In these experiments the ability of one strain to out-compete or inhibit the growth of the other strain would manifest itself as an increase in the proportion of the dominant strain. During intranasal coinfections, the strain predominated in the brain even though the levels of a and strains were equivalent in the lung and spleen. Because intranasal coinfections with the unmarked strains and larger-scale coinfections with marked strains showed that a and strains disseminated equally well from the lung to the spleen, we conclude that the difference observed in the CNS is not simply due to an inability of a cells to disseminate from the lung but is specific to the CNS. Intracerebral infections showed that both strains had similar levels of persistence in the brain during coinfection. Thus, cell predominance in the CNS appears to be due to entry into and not persistence in the brain.
A number of hypotheses could explain the observation that cells had higher levels of entry into the CNS during coinfection. First, an interaction between and a cells could enhance the ability of the cells to invade the brain (Fig. 7A). For example, laccase is critical for the survival of Cryptococcus in the brain, and urease has been shown to enhance invasion of the central nervous system by Cryptococcus (57, 60). Increased production of these enzymes in the strain or their decreased production in the a strain may potentially result in a higher proportion of strains in the CNS.
A second hypothesis is that interactions between and a cells results in a physical alteration of the a cells that inhibits their ability to enter the CNS (Fig. 7B). In humans, C. neoformans replicates in the lung and then spreads via the bloodstream or lymphatic system to other organs as free cells or internalized in mononuclear cells, such as macrophages, that travel to microcapillary beds in the brain (16, 28, 57, 61). At the blood-brain barrier, C. neoformans can transverse microvascular endothelial cells either directly or via a Trojan horse mechanism within mononuclear phagocytes (13, 15, 61). Mating type a strains are known to generate large, bulbous cells in response to pheromone in serotype D var. neoformans, and the pheromone MF1 gene has been shown to be expressed in the brain during serotype A var. grubii infection as well as inside C. elegans (23, 53). While enlargement of a cells has not been observed in the laboratory for var. grubii serotype A, it is possible that a cells respond to pheromone produced by coinfection with cells in vivo. Pheromone-induced cellular changes in a cells may inhibit their entry into microcapillaries in the brain or their subsequent transversal of endothelial cells but not affect their ability to spread to other organs, such as the spleen.
Finally, as a third hypothesis, we cannot rule out the possibility that a and cells do not communicate with each other but interact differently with the host. But as the difference in virulence occurs only during coinfection, the data suggest a three-way dialog between the host and both a and cells of the fungal pathogen (Fig. 7C). For example, one strain could elicit an immune response from the host that affects the other strain disproportionately, which manifests itself as a virulence difference in the brain.
We have detected no evidence of mating or a- cell fusion in vivo to produce a/ diploid cells during coinfection (this work and Y.-S. Bahn and J. Heitman, unpublished results). Thus, these conditions may preclude mating or counterselect against a/ diploid or dikaryotic products of mating. This said, early events in a- cell signaling that transpire in vitro may occur in vivo, as several conditions conducive to mating (darkness, nutrient limitation, and pheromone expression) are present in vivo and could result in local a- cell signaling that differentially impacts fungal cell interactions with the host.
It is possible that strains are the predominant clinical isolates because they are more prevalent in the environment. If this is the case, then areas with high levels of environmental a strains would be expected to have roughly equal levels of a strains as clinical isolates. Examination of serotype A clinical strain populations from sub-Saharan countries show that up to 10% of the clinical isolates can be the a mating type, and a population of clinical isolates from Botswana showed evidence of sexual recombination (44, 46). It will be of interest to establish what proportion of environmental strains in sub-Saharan Africa is represented by a strains.
The predominance of cells in the CNS during mixed infections could also contribute to the predominance of strains in clinical isolates, which are derived predominantly from the cerebrospinal fluid. It is possible that strains appear to be predominant because coinfections from mixed a and spores, produced as a result of mating, result in CNS infections with a higher proportion of strains. Furthermore, most clinical isolates are single-colony isolates. Thus, if the cerebrospinal fluid has a higher proportion of cells, there is also a higher probability that the resulting single-colony isolate will be an strain. In the few cases where serial isolates have been collected from the same patient, a second strain has been isolated as often as 29% of the time (58). These patients could have been reinfected with a different strain or coinfected with multiple strains. Examination of multiple isolates from the same patient, preferably from lung or sputum cultures from patients in whom multiple strains might be expected to be present in roughly equal proportions, will provide more-accurate data on levels of coinfection and the genetic nature of the coinfecting strains with respect to the mating type locus and other genes linked to the virulence potential of Cryptococcus.
ACKNOWLEDGMENTS
We thank Cristl Arndt and A. Lisa De Obaldia for technical assistance, James Fraser for assistance with marked strain development, Alex Idnurm and Yong-Sun Bahn for animal assistance, and other members of the Duke University Mycology Research Unit (DUMRU) for their helpful comments and discussions.
This work was supported by NIAID R01 grant AI50113 to Joseph Heitman and NIAID R01 grant AI28388 to John Perfect. Daniel Benjamin received support from NICHD HD044799-01. Gary Cox was a Burroughs Wellcome New Investigator, and Joseph Heitman is a Burroughs Wellcome Scholar in Molecular Pathogenic Mycology and an investigator of the Howard Hughes Medical Institute.
REFERENCES
1. Aberg, J. A., and W. G. Powderly. 1997. Cryptococcal disease: implications of recent clinical trials on treatment and management. AIDS Clin. Rev. 1997-1998:229-248.
2. 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.
3. Bochner, B. R. 2003. New technologies to assess genotype-phenotype relationships. Nat. Rev. Genet. 4:309-314.
4. Bochner, B. R., P. Gadzinski, and E. Panomitros. 2001. Phenotype microarrays for high-throughput phenotypic testing and assay of gene function. Genome Res. 11:1246-1255.
5. Bolanos, B., and T. G. Mitchell. 1989. Killing of Cryptococcus neoformans by rat alveolar macrophages. J. Med. Vet. Mycol. 27:219-228.
6. Bolanos, B., and T. G. Mitchell. 1989. Phagocytosis and killing of Cryptococcus neoformans by rat alveolar macrophages in the absence of serum. J. Leukoc. Biol. 46:521-528.
7. Bolanos, B., and T. G. Mitchell. 1989. Phagocytosis of Cryptococcus neoformans by rat alveolar macrophages. J. Med. Vet. Mycol. 27:203-217.
8. Brandt, M. E., L. C. Hutwagner, L. A. Klug, W. S. Baughman, D. Rimland, E. A. Graviss, R. J. Hamill, C. Thomas, P. G. Pappas, A. L. Reingold, R. W. Pinner, and the Cryptococcal Disease Active Surveillance Group. 1996. Molecular subtype distribution of Cryptococcus neoformans in four areas of the United States. J. Clin. Microbiol. 34:912-917.
9. Casadevall, A., J. Mukherjee, R. Yuan, and J. Perfect. 1994. Management of injuries caused by Cryptococcus neoformans-contaminated needles. Clin. Infect. Dis. 19:951-953.
10. Casadevall, A., and J. R. Perfect. 1998. Cryptococcus neoformans. ASM Press, Washington, D.C.
11. Casadevall, A., J. N. Steenbergen, and J. D. Nosanchuk. 2003. ‘Ready made’ virulence and ‘dual use’ virulence factors in pathogenic environmental fungi—the Cryptococcus neoformans paradigm. Curr. Opin. Microbiol. 6:332-337.
12. Chang, Y. C., L. A. Penoyer, and K. J. Kwon-Chung. 2001. The second STE12 homologue of Cryptococcus neoformans is MATa-specific and plays an important role in virulence. Proc. Natl. Acad. Sci. USA 98:3258-3263.
13. Chang, Y. C., M. F. Stins, M. J. McCaffery, G. F. Miller, D. R. Pare, T. Dam, M. Paul-Satyaseela, K. S. Kim, and K. J. Kwon-Chung. 2004. Cryptococcal yeast cells invade the central nervous system via transcellular penetration of the blood-brain barrier. Infect. Immun. 72:4985-4995.
14. Chang, Y. C., B. L. Wickes, G. F. Miller, L. A. Penoyer, and K. J. Kwon-Chung. 2000. Cryptococcus neoformans STE12 regulates virulence but is not essential for mating. J. Exp. Med. 191:871-882.
15. Chen, S. H., M. F. Stins, S. H. Huang, Y. H. Chen, K. J. Kwon-Chung, Y. Chang, K. S. Kim, K. Suzuki, and A. Y. Jong. 2003. Cryptococcus neoformans induces alterations in the cytoskeleton of human brain microvascular endothelial cells. J. Med. Microbiol. 52:961-970.
16. Chretien, F., O. Lortholary, I. Kansau, S. Neuville, F. Gray, and F. Dromer. 2002. Pathogenesis of cerebral Cryptococcus neoformans infection after fungemia. J. Infect. Dis. 186:522-530.
17. Clarke, D. L., G. L. Woodlee, C. M. McClelland, T. S. Seymour, and B. L. Wickes. 2001. The Cryptococcus neoformans STE11 gene is similar to other fungal mitogen-activated protein kinase kinase kinase (MAPKKK) genes but is mating type specific. Mol. Microbiol. 40:200-213.
18. Cox, G. M., T. S. Harrison, H. C. McDade, C. P. Taborda, G. Heinrich, A. Casadevall, and J. R. Perfect. 2003. Superoxide dismutase influences the virulence of Cryptococcus neoformans by affecting growth within macrophages. Infect. Immun. 71:173-180.
19. Currie, B. P., and A. Casadevall. 1994. Estimation of the prevalence of cryptococcal infection among patients infected with the human immunodeficiency virus in New York City. Clin. Infect. Dis. 19:1029-1033.
20. Davidson, R. C., J. R. Blankenship, P. R. Kraus, M. D. Berrios, C. M. Hull, C. D'Souza, P. Wang, and J. Heitman. 2002. A PCR-based strategy to generate integrative targeting alleles with large regions of homology. Microbiology 148:2607-2615.
21. Davidson, R. C., C. B. Nichols, G. M. Cox, J. R. Perfect, and J. Heitman. 2003. A MAP kinase cascade composed of cell type specific and non-specific elements controls mating and differentiation of the fungal pathogen Cryptococcus neoformans. Mol. Microbiol. 49:469-485.
22. DeJesus-Berrios, M., L. Liu, J. Nussbaum, G. M. Cox, J. S. Stamler, and J. Heitman. 2003. Enzymes that counteract nitrosative challenge promote fungal virulence. Curr. Biol. 13:1963-1968.
23. Del Poeta, M., D. L. Toffaletti, T. H. Rude, S. D. Sparks, J. Heitman, and J. R. Perfect. 1999. Cryptococcus neoformans differential gene expression detected in vitro and in vivo with green fluorescent protein. Infect. Immun. 67:1812-1820.
24. Edmond, M. B., S. E. Wallace, D. K. McClish, M. A. Pfaller, R. N. Jones, and R. P. Wenzel. 1999. Nosocomial bloodstream infections in United States hospitals: a three-year analysis. Clin. Infect. Dis. 29:239-244.
25. Feldmesser, M., Y. Kress, P. Novikoff, and A. Casadevall. 2000. Cryptococcus neoformans is a facultative intracellular pathogen in murine pulmonary infection. Infect. Immun. 68:4225-4237.
26. Feldmesser, M., S. Tucker, and A. Casadevall. 2001. Intracellular parasitism of macrophages by Cryptococcus neoformans. Trends Microbiol. 9:273-278.
27. Fraser, J. A., R. L. Subaran, C. B. Nichols, and J. Heitman. 2003. Recapitulation of the sexual cycle of the primary fungal pathogen Cryptococcus neoformans variety gattii: implications for an outbreak on Vancouver Island. Eukaryot. Cell 2:1036-1045.
28. Garcia-Hermoso, D., G. Janbon, and F. Dromer. 1999. Epidemiological evidence for dormant Cryptococcus neoformans infection. J. Clin. Microbiol. 37:3204-3209.
29. Gordon, S. B., A. L. Walsh, M. Chaponda, M. A. Gordon, D. Soko, M. Mbwvinji, M. E. Molyneux, and R. C. Read. 2000. Bacterial meningitis in Malawian adults: pneumococcal disease is common, severe, and seasonal. Clin. Infect. Dis. 31:53-57.
30. Granger, D. L., J. B. Hibbs, Jr., J. R. Perfect, and D. T. Durack. 1988. Specific amino acid (L-arginine) requirement for the microbiostatic activity of murine macrophages. J. Clin. Investig. 81:1129-1136.
31. Hajjeh, R. A., L. A. Conn, D. S. Stephens, W. Baughman, R. Hamill, E. Graviss, P. G. Pappas, C. Thomas, A. Reingold, G. Rothrock, L. C. Hutwagner, A. Schuchat, M. E. Brandt, R. W. Pinner, et al. 1999. Cryptococcosis: population-based multistate active surveillance and risk factors in human immunodeficiency virus-infected persons. J. Infect. Dis. 179:449-454.
32. Hakim, J. G., I. T. Gangaidzo, R. S. Heyderman, J. Mielke, E. Mushangi, A. Taziwa, V. J. Robertson, P. Musvaire, and P. R. Mason. 2000. Impact of HIV infection on meningitis in Harare, Zimbabwe: a prospective study of 406 predominantly adult patients. AIDS 14:1401-1407.
33. Haynes, K. A., D. J. Sullivan, D. C. Coleman, J. C. Clarke, R. Emilianus, C. Atkinson, and K. J. Cann. 1995. Involvement of multiple Cryptococcus neoformans strains in a single episode of cryptococcosis and reinfection with novel strains in recurrent infection demonstrated by random amplification of polymorphic DNA and DNA fingerprinting. J. Clin. Microbiol. 33:99-102.
34. Heitman, J., B. Allen, J. A. Alspaugh, and K. J. Kwon-Chung. 1999. On the origins of the congenic MAT and MATa strains of the pathogenic yeast Cryptococcus neoformans. Fungal Genet. Biol. 28:1-5.
35. Heyderman, R. S., I. T. Gangaidzo, J. G. Hakim, J. Mielke, A. Taziwa, P. Musvaire, V. J. Robertson, and P. R. Mason. 1998. Cryptococcal meningitis in human immunodeficiency virus-infected patients in Harare, Zimbabwe. Clin. Infect. Dis. 26:284-289.
36. Hull, C. M., G. M. Cox, and J. Heitman. 2004. The -specific cell identity factor Sxi1 is not required for virulence of Cryptococcus neoformans. Infect. Immun. 72:3643-3645.
37. Hull, C. M., R. C. Davidson, and J. Heitman. 2002. Cell identity and sexual development in Cryptococcus neoformans are controlled by the mating-type-specific homeodomain protein Sxi1. Genes Dev. 16:3046-3060.
38. Igreja, R. P., S. Lazera Mdos, B. Wanke, M. C. Galhardo, S. E. Kidd, and W. Meyer. 2004. Molecular epidemiology of Cryptococcus neoformans isolates from AIDS patients of the Brazilian city, Rio de Janeiro. Med. Mycol. 42:229-238.
39. Keller, S. M., M. A. Viviani, M. C. Esposto, M. Cogliati, and B. L. Wickes. 2003. Molecular and genetic characterization of a serotype A MATa Cryptococcus neoformans isolate. Microbiology 149:131-142.
40. Kwon-Chung, J. K., and W. B. Hill. 1981. Sexuality and pathogenicity of Filobasidiella neoformans (Cryptococcus neoformans), p. 243-250. In C. Vanbreuseghem (ed.), Sexuality and pathogenicity of fungi. Masson, Paris, France.
41. Kwon-Chung, K. J., and J. E. Bennett. 1978. Distribution of alpha and a mating types of Cryptococcus neoformans among natural and clinical isolates. Am. J. Epidemiol. 108:337-340.
42. Kwon-Chung, K. J., J. C. Edman, and B. L. Wickes. 1992. Genetic association of mating types and virulence in Cryptococcus neoformans. Infect. Immun. 60:602-605.
43. Lengeler, K. B., D. S. Fox, J. A. Fraser, A. Allen, K. Forrester, F. S. Dietrich, and J. Heitman. 2002. Mating-type locus of Cryptococcus neoformans: a step in the evolution of sex chromosomes. Eukaryot. Cell 1:704-718.
44. Lengeler, K. B., P. Wang, G. M. Cox, J. R. Perfect, and J. Heitman. 2000. Identification of the MATa mating-type locus of Cryptococcus neoformans reveals a serotype A MATa strain thought to have been extinct. Proc. Natl. Acad. Sci. USA 97:14455-14460.
45. Li, R. K., and T. G. Mitchell. 1997. Induction of interleukin-6 mRNA in rat alveolar macrophages by in vitro exposure to both Cryptococcus neoformans and anti-C. neoformans antiserum. J. Med. Vet. Mycol. 35:327-334.
46. Litvintseva, A. P., R. E. Marra, K. Nielsen, J. Heitman, R. Vilgalys, and T. G. Mitchell. 2003. Evidence of sexual recombination among Cryptococcus neoformans serotype A isolates in sub-Saharan Africa. Eukaryot. Cell 2:1162-1168.
47. Liu, H., T. R. Cottrell, L. M. Pierini, W. E. Goldman, and T. L. Doering. 2002. RNA interference in the pathogenic fungus Cryptococcus neoformans. Genetics 160:463-470.
48. Liu, L., R. P. Tewari, and P. R. Williamson. 1999. Laccase protects Cryptococcus neoformans from antifungal activity of alveolar macrophages. Infect. Immun. 67:6034-6039.
49. Lodge, J. K., E. Jackson-Machelski, D. L. Toffaletti, J. R. Perfect, and J. I. Gordon. 1994. Targeted gene replacement demonstrates that myristoyl-CoA:protein N-myristoyltransferase is essential for viability of Cryptococcus neoformans. Proc. Natl. Acad. Sci. USA 91:12008-12012.
50. Maher, D., and H. Mwandumba. 1994. Cryptococcal meningitis in Lilongwe and Blentyre, Malawi. J. Infect. 28:59-64.
51. Malliaris, S. D., J. N. Steenbergen, and A. Casadevall. 2004. Cryptococcus neoformans var. gattii can exploit Acanthamoeba castellanii for growth. Med. Mycol. 42:149-158.
52. Mitchell, T. G., and J. R. Perfect. 1995. Cryptococcosis in the era of AIDS—100 years after the discovery of Cryptococcus neoformans. Clin. Microbiol. Rev. 8:515-548.
53. 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.
54. 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.
55. 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.
56. Nosanchuk, J. D., P. Valadon, M. Feldmesser, and A. Casadevall. 1999. Melanization of Cryptococcus neoformans in murine infection. Mol. Cell. Biol. 19:745-750.
57. Olszewski, M. A., M. C. Noverr, G. H. Chen, G. B. Toews, G. M. Cox, J. R. Perfect, and G. B. Huffnagle. 2004. Urease expression by Cryptococcus neoformans promotes microvascular sequestration, thereby enhancing central nervous system invasion. Am. J. Pathol. 164:1761-1771.
58. Pfaller, M., J. Zhang, S. Messer, M. Tumberland, E. Mbidde, C. Jessup, and M. Ghannoum. 1998. Molecular epidemiology and antifungal susceptibility of Cryptococcus neoformans isolates from Ugandan AIDS patients. Diagn. Microbiol. Infect. Dis. 32:191-199.
59. Rhodes, J. C., L. S. Wicker, and W. J. Urba. 1980. Genetic control of susceptibility to Cryptococcus neoformans in mice. Infect. Immun. 29:494-499.
60. 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.
61. Santangelo, R., H. Zoellner, T. Sorrell, C. Wilson, C. Donald, J. Djordjevic, Y. Shounan, and L. Wright. 2004. Role of extracellular phospholipases and mononuclear phagocytes in dissemination of cryptococcosis in a murine model. Infect. Immun. 72:2229-2239.
62. Sorvillo, F., G. Beall, P. A. Turner, V. L. Beer, A. A. Kovacs, and P. R. Kerndt. 1997. Incidence and factors associated with extrapulmonary cryptococcosis among persons with HIV infection in Los Angeles County. AIDS 11:673-679.
63. Speed, B., and D. Dunt. 1995. Clinical and host differences between infections with the two varieties of Cryptococcus neoformans. Clin. Infect. Dis. 21:28-34.
64. Steenbergen, J. N., and A. Casadevall. 2003. The origin and maintenance of virulence for the human pathogenic fungus Cryptococcus neoformans. Microbes Infect. 5:667-675.
65. Steenbergen, J. N., J. D. Nosanchuk, S. D. Malliaris, and A. Casadevall. 2003. Cryptococcus neoformans virulence is enhanced after growth in the genetically malleable host Dictyostelium discoideum. Infect. Immun. 71:4862-4872.
66. 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.
67. Sukroongreung, S., K. Kitiniyom, C. Nilakul, and S. Tantimavanich. 1998. Pathogenicity of basidiospores of Filobasidiella neoformans var. neoformans. Med. Mycol. 36:419-424.
68. Sullivan, D., K. Haynes, G. Moran, D. Shanley, and D. Coleman. 1996. Persistence, replacement, and microevolution of Cryptococcus neoformans strains in recurrent meningitis in AIDS patients. J. Clin. Microbiol. 34:1739-1744.
69. Tintelnot, K., K. Lemmer, H. Losert, G. Schar, and A. Polak. 2004. Follow-up of epidemiological data of cryptococcosis in Austria, Germany and Switzerland with special focus on the characterization of clinical isolates. Mycoses 47:455-464.
70. Toffaletti, D. L., T. H. Rude, S. A. Johnston, D. T. Durack, and J. R. Perfect. 1993. Gene transfer in Cryptococcus neoformans by use of biolistic delivery of DNA. J. Bacteriol. 175:1405-1411.
71. Tucker, S. C., and A. Casadevall. 2002. Replication of Cryptococcus neoformans in macrophages is accompanied by phagosomal permeabilization and accumulation of vesicles containing polysaccharide in the cytoplasm. Proc. Natl. Acad. Sci. USA 99:3165-3170.
72. Vallim, M. A., L. Fernandes, and J. A. Alspaugh. 2004. The RAM1 gene encoding a protein-farnesyltransferase beta-subunit homologue is essential in Cryptococcus neoformans. Microbiology 150:1925-1935.
73. Viviani, M. A., M. C. Esposto, M. Cogliati, M. T. Montagna, and B. L. Wickes. 2001. Isolation of a Cryptococcus neoformans serotype A MATa strain from the Italian environment. Med. Mycol. 39:383-386.
74. Viviani, M. A., R. Nikolova, M. C. Esposto, G. Prinz, and M. Cogliati. 2003. First European case of serotype A MATa Cryptococcus neoformans infection. Emerg. Infect. Dis. 9:1179-1180.
75. Wang, P., C. B. Nichols, K. B. Lengeler, M. E. Cardenas, G. M. Cox, J. R. Perfect, and J. Heitman. 2002. Mating-type-specific and nonspecific PAK kinases play shared and divergent roles in Cryptococcus neoformans. Eukaryot. Cell 1:257-272.
76. World Heath Organization. 2003. HIV/AIDS Epidemiological surveillance update for the W.H.O. African region. World Health Organization, Geneva, Switzerland.
77. World Health Organization. 2001. HIV/AIDS in Asia and the Pacific Region. World Health Organization, Geneva, Switzerland.
78. Wickes, B. L., and J. K. Kwon-Chung. 2002. Genetic basis of pathogenicity in Cryptococcus neoformans, p. 25-49. In R. L. Calderone (ed.), Fungal pathogenesis: principles and clinical applications. Marcell Dekker, New York, N.Y.
79. Wilder, J. A., G. K. Olson, Y. C. Chang, K. J. Kwon-Chung, and M. F. Lipscomb. 2002. Complementation of a capsule deficient Cryptococcus neoformans with Cap64 restores virulence in a murine lung infection. Am. J. Respir. Cell. Mol. Biol. 26:306-314.
80. Zhou, L., X. H. Lei, B. R. Bochner, and B. L. Wanner. 2003. Phenotype microarray analysis of Escherichia coli K-12 mutants with deletions of all two-component systems. J. Bacteriol. 185:4956-4972.(Kirsten Nielsen, Gary M. )
Divisions of Infectious Diseases Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, New York 10461
Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts 02114
ABSTRACT
Cryptococcus neoformans is a fungal pathogen that has evolved over the past 40 million years into three distinct varieties or sibling species (gattii, grubii, and neoformans). Each variety manifests differences in epidemiology and disease, and var. grubii strains are responsible for the vast majority of human disease. In previous studies, strains were more virulent than congenic a strains in var. neoformans, whereas var. grubii congenic a and strains exhibited equivalent levels of virulence. Here the role of mating type in the virulence of var. grubii was further characterized in a panel of model systems. Congenic var. grubii a and strains had equivalent survival rates when cultured with amoebae, nematodes, and macrophages. No difference in virulence was observed between a and congenic strains in multiple inbred-mouse genetic backgrounds, and there was no difference in accumulations in the central nervous system (CNS) late in infection. In contrast, during coinfections, a and strains are equivalent in peripheral tissues but cells have an enhanced predilection to penetrate the CNS. These studies reveal the first virulence difference between congenic a and strains in the most common pathogenic variety and suggest an explanation for the prevalence of strains in clinical isolates.
INTRODUCTION
Over the past decade fungal diseases have increased due to an expanding population of immunocompromised individuals. Rates of human immunodeficiency virus (HIV) infections are rising in many areas of the world, and the regions with the highest AIDS burden often have the least medical resources, resulting in large immunocompromised populations. At the end of 2002, it was estimated that 29 million people in Africa were living with HIV and that 3 million people died of AIDS-related illnesses during 2002 (76). This phenomenon, combined with the development of aggressive surgeries, organ transplantation, and immunosuppressive therapies, has led to an increase in the prevalence of invasive fungal diseases. Fungal diseases, such as oropharyngeal candidiasis and cryptococcal meningitis, are often AIDS-defining illnesses; aspergillosis has emerged as an important cause of mortality following bone marrow transplants; and Candida species are now a leading cause of bloodstream infections in hospitalized patients (24). As fungal diseases become more prevalent, understanding the microbial characteristics responsible for virulence becomes increasingly important.
The basidiomycete Cryptococcus neoformans is both an important pathogen and an excellent model system for fungal virulence (37). C. neoformans infects the central nervous system (CNS) to cause meningoencephalitis that is uniformly fatal if untreated (1). Cryptococcal meningitis occurs in 6 to 10% of AIDS patients in the United States and is the first AIDS-defining illness in 40% of these patients (19, 31, 52, 62). In sub-Saharan Africa, where highly active antiretroviral therapy is less common, C. neoformans accounts for 45% of all cases of meningitis in adults; many individuals present with cryptococcosis as their first AIDS-defining illness, and the mean survival rate after diagnosis can be as short as 4 days (29, 32, 35, 50). Cryptococcus occurs in three varieties or species that can be distinguished based on genetic and phenotypic differences. While these varieties are pathogenic in humans, they have different disease epidemiologies. C. neoformans var. grubii (serotype A) strains cause 95% of all C. neoformans infections and >99% of infections in AIDS patients in the United States (10). However, C. neoformans var. neoformans (serotype D) strains can account for up to 20% of clinical isolates in Europe (69). Unlike the first two varieties of C. neoformans, which cause disease predominantly in individuals with impaired immunity, strains of the sibling species—C. gattii (serotypes B and C)—can infect immunocompetent individuals (63).
C. neoformans is an important human pathogen and has also proven to be an outstanding model system for studying the virulence of fungi. Cryptococcus is predominantly haploid, which facilitates genetic analyses, and has a defined sexual cycle that enables classical genetic studies. Genes can readily be disrupted by biolistic transformation and homologous recombination, and essential genes have been examined using stable diploid strains, conditional alleles, and RNA interference (20, 37, 47, 49, 72). C. neoformans is a facultative intracellular pathogen that can also infect the amoeba Acanthamoeba castellanii, the nematode Caenorhabditis elegans, the slime mold Dictyostelium discoideum, and the insect Drosophila melanogaster, allowing identification of genes involved in virulence in heterologous hosts (2, 53, 64-66). Finally, robust animal model systems that enable detailed studies of virulence characteristics, such as the polysaccharide capsule and the antioxidant pigment melanin, have been developed.
Mating type has been proposed to be a virulence factor in C. neoformans based on disease epidemiology and animal studies. The majority of clinical isolates are of the mating type (10, 41). Early experimental work on the role of mating type in virulence was conducted with serotype D var. neoformans because of the availability of congenic strains (42; reviewed in reference 34). In serotype D, a comparison of the levels of virulence of nonisogenic, unrelated a and strains revealed that strains are more virulent at lower inoculum levels than a strains (40, 78) and that the strain JEC21 was more virulent than the congenic a strain JEC20 in a murine tail vein infection model of cryptococcosis as well as in the C. elegans system (42, 54). In serotype A var. grubii, mating type a strains are exceptionally rare and have only recently been identified (39, 44, 46, 55, 73, 74). As with serotype D, when the virulence of nonisogenic a and strains was compared, the a strains were less virulent than the strains (39, 44, 55). However, when congenic serotype A strains in the highly virulent H99 background were compared in standard mouse inhalation and rabbit meningitis models of cryptococcosis, no difference in virulence was measured (55).
Here we characterize in detail the role of mating type in pathogenicity by examining the virulence of serotype A var. grubii a and congenic strains at various stages in the infective cycle—from their survival in potential environmental predators to their colonization of various organs in animal models—and found no difference between a and strains. Infection by multiple strains has been reported, and studies examining sequential isolates have shown that different strains can be isolated from the same patient (8, 9, 33, 38, 58, 68). These findings suggest that coinfection with strains of opposite mating types may occur. We therefore analyzed the virulence of the congenic serotype A var. grubii strains during coinfection with both mating types and discovered that a and congenic strains reached equivalent levels in the lungs and spleen but that a significantly higher proportion of cells infected the brain. Intracerebral coinfections revealed that the strains have equivalent levels of persistence in the CNS. Thus, the strain out-competes the a strain in entry into the CNS as a consequence of either fungal-fungal or fungal-host interactions.
MATERIALS AND METHODS
Strains and media. C. neoformans var. grubii strains H99 (70), KN99a (55), KN99 (55), KN99aNAT, and KN99NAT were maintained at –80°C. Cultures were initiated by inoculating yeast extract-peptone-dextrose (YPD) solid medium and incubating the plates for 2 days at 30°C, followed by growth in YPD liquid medium at 30°C overnight. Cells were harvested, washed three times with phosphate-buffered saline (PBS), and resuspended at the appropriate concentration in PBS. Acanthamoeba castellanii strain 30324 (ATCC) was maintained as previously described (51). Caenorhabditis elegans strain N2 Bristol was maintained at 15°C and propagated on Escherichia coli strain OP50 (53). The MH-S murine alveolar macrophage cell line (ATCC) was maintained in Dulbecco modified Eagle complete medium at 37°C with 5% CO2 and transferred to fresh medium every 3 to 5 days.
AFLP. Amplified fragment length polymorphisms (AFLPs) were generated and analyzed as previously described (46). Two different EcoRI primer combinations were used for the selective PCR. Both primer sequences were labeled at their 5' ends with 6-carboxyfluorescein (6FAM) but differed in the final two bases at the 3' ends of the primers. The primer sequences were as follows: primer 1, 5'-6FAM-GACTGCGTACCAATTCAC-3', and primer 2, 5'-6FAM-GACTGCGTACCAATTCTG-3'. Either primer 1 or primer 2 was combined with the MseI primer (5'-GATGAGTCCTGAGTAAG-3') in the selective PCR. The AFLP reaction and analysis were performed on at least two separate occasions for each strain. Only intense and reproducible bands were scored to identify differences between strains.
Phenotype microarrays. Phenotype microarrays were done as described previously (80) and according to instructions from the manufacturer (Biolog, Hayward, CA) with the following exceptions. Strain cultures were grown overnight in YPD liquid medium; the cells were then collected by centrifugation, washed once with sterile distilled H2O, resuspended in 12x yeast nitrogen base liquid medium supplemented with histidine, leucine, and uracil, and adjusted to 18% transmittance (T). IFY-0 was the inoculating fluid for all 20 96-well plate phenotype microarrays (PM1 to PM10 and PM21 to PM30). IFY-0 was supplemented with 5 mg/ml ammonium sulfate, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM1 and PM2; 20 mg/ml D-glucose, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM3 and PM6 to PM8; 20 mg/ml D-glucose and 1 mg/ml L-glutamate for PM4; or 20 mg/ml D-glucose, 5 mg/ml ammonium sulfate, 0.85 mg/ml potassium phosphate monobasic, 0.15 mg/ml potassium phosphate dibasic, and 0.5 mg/ml magnesium sulfate for PM9, PM10, and PM21 to PM30. Cells were combined with appropriately supplemented IFY-0 at a ratio of 1:11, and 100 μl of this mixture was added to each well. All phenotype microarrays were incubated at 30°C for 36 h in an OmniLog and monitored for color change in the wells every 15 min. Kinetic data were analyzed with OmniLog software using pairwise combinations.
Generation of KN99aNAT and KN99NAT strains. Strains KN99aNAT and KN99NAT were generated using overlap PCR by following the methods described in reference 27. The nourseothricin transgene (NAT) was inserted into the largest intergenic region in the corresponding MAT locus, where it would be anticipated to have the least effect on surrounding genes. For the serotype A MATa locus (43), the NAT transgene was inserted in a 12-kb region between the RPL39 and PRT1 genes. For the serotype A MAT locus, the transgene was inserted in a 4-kb region between the BSP1 and RPL39 genes. PCR was used to generate the flanking 5' (with KN99a, JOHE12926 and JOHE12927; with KN99, JOHE12921 and JOHE12922) and 3' (with KN99a, JOHE12928 and JOHE12929; with KN99, JOHE12923 and JOHE12924) regions containing linkers to a NATr cassette. The NATr cassette containing the nourseothricin transgene was amplified using oligonucleotides JOHE8733 and JOHE8738. The products were combined in a PCR overlap reaction (with KN99a, JOHE12926 and JOHE12929; with KN99, JOHE12921 and JOHE12924) to yield the NAT insertion allele. KN99a or KN99 was then transformed with the appropriate NAT insertion allele by biolistic transformation and selection for colonies resistant to nourseothricin (100 μg/ml). Transformants were identified by Southern blotting of KpnI-digested DNA using the 5' fragment as a probe. Primer sequences were as follows: for JOHE8733, GCTGCGAGGATGTGAGCTGGA; for JOHE8738, GGTTTATCTGTATTAACACGG; for JOHE12921, GGTGAGTTGAGTGCGGCTATT; for JOHE12922, AGCTCACATCCTCGCAGCGCATTTCAATTCTGTTAT (NATr linker is in bold); for JOHE12923, CCGTGTTAATACAGATAAACCCACATTGACTACTTATGACCA; for JOHE12924, GCTATAAACATTCCCGTCGTG; for JOHE12926, GACTCTGCCATCGTCCAAGAA; for JOHE12927, AGCTCACATCCTCGCAGCCAGTTTACCTGGATTTCGAGC; for JOHE12928, CCGTGTTAATACAGATAACCCACTTCTGTTATGCGAGATGC; and for JOHE12929, CCACCGCTCTTCCAGATGAAA.
Nematode killing assay. C. elegans killing assays were done as described previously (53, 54) with minor modifications. The C. neoformans strains were inoculated into 2 ml YPD and grown at 30°C for 24 h, 10 μl of the culture was spread on brain heart infusion agar (Difco), and the plates were incubated at 30°C for 48 h. C. elegans animals were transferred to the plates (divided across three plates of each strain) and incubated at 25°C, and then 120 to 150 animals were examined for viability at 24-h intervals. Replicate experiments showed similar results.
Amoeba assay. The survival of C. neoformans var. grubii in amoebae was as described previously (51). Briefly, A. castellanii organisms were collected, washed twice with PBS, suspended in PBS, counted with a hemocytometer, and diluted in PBS to the appropriate density. A. castellanii cells were added to a 96-well tissue culture plate at 1 x 104 cells per well and allowed to adhere for 1 h at 28°C. C. neoformans cells (1 x 104) were added to wells containing amoebae or control wells containing PBS alone, and the plates were incubated at 28°C. At 0, 24, and 48 h, the amoebae were lysed and the well contents were serially diluted and plated onto Sabouraud (SAB) dextrose medium. Numbers of CFU were determined after incubation overnight at 30°C. Replicate experiments showed similar results.
Macrophage assay. Macrophages were harvested by removal from monolayers, and cell viability was determined by trypan blue exclusion. Cell number was determined using a hemocytometer, and the macrophage density was adjusted to 5 x 105 cells/ml of culture medium. Macrophages were placed in 96-well plates and stimulated overnight with 50 units/ml gamma interferon (murine) and 0.3 mg/ml lipopolysaccharide (LPS). C. neoformans cells were added to the macrophage suspension at a density of 5 x 105 cells/ml of culture medium with 10 μg/ml of monoclonal antibody 1887 as an opsonin. Macrophage-yeast mixtures were incubated for 1 h and washed three times with culture medium. The macrophages were then either lysed with 0.05% sodium dodecyl sulfate (SDS) immediately to determine the phagocytic index or incubated with stimulation for 24 h at 37°C with 5% CO2. After incubation, the culture medium was aspirated from each well and the macrophages were lysed with two exchanges of 0.05% SDS. The aspirated medium and SDS solutions were combined, serially diluted, and plated onto YPD plates. Numbers of CFU were determined after incubation overnight at 30°C. Replicate experiments showed similar results.
Mouse virulence studies. Groups of 4- to 8-week-old female A/JCr, BALB/c, or C57BL/6 mice (five mice per strain) were anesthetized by intraperitoneal phenobarbital injection. Animals were infected intranasally with 5 x 104 fungal cells in 50 μl of normal saline pipetted slowly into the nares. Animals were infected intracerebrally with 1.5 x 103 fungal cells in 20 μl of normal saline injected slowly into the top of the cerebrum. In further experiments, animals were directly injected in their lateral tail veins with 2.5 x 104 cells. The concentration of yeast cells in the inoculum was confirmed by plating serial dilutions and enumerating CFU. Mice were monitored twice daily, and those that showed signs of severe morbidity (weight loss, extension of the cerebral portion of the cranium, abnormal gait, paralysis, seizures, convulsions, or coma) were sacrificed by CO2 inhalation.
Quantitative analysis of tissue fungal burden. Groups of 4- to 6-week-old female A/JCr mice were infected either intranasally or intracerebrally. For intranasal infections, animals (five mice per strain per time interval) were sacrificed by CO2 inhalation at 7, 14, and 21 days postinfection, and lung, spleen, and brain were removed, weighed, and homogenized in 2 ml sterile PBS. Animals were sacrificed by CO2 inhalation at 4 days postinfection for intracerebral infections (10 mice per strain), and brain and lung were removed, weighed, and homogenized in 2 ml sterile PBS. Serial dilutions of the organ samples were plated on SAB agar plates containing 100 μg/ml chloramphenicol and incubated at 30°C overnight. Colony counts were performed and adjusted to reflect the total number of CFU/g tissue.
Coinfection experiments. Groups of 4- to 6-week-old female A/JCr mice were coinfected with a 1:1 ratio of a and cells either intranasally or intracerebrally. The inoculum was prepared immediately before animals were infected and then plated on SAB medium. Colonies were isolated and analyzed to determine the proportion of a cells present in the inoculum. Mice were sacrificed by CO2 inhalation; the lung, spleen, and brain were harvested and homogenized; and CFU were enumerated as in the previous assay. For the first coinfection experiments (five mice per time interval), animals were coinfected with KN99a and H99. Mice were sacrificed at 7, 14, and 21 days postinfection, and single colonies (64 from inoculum, 32 from lungs and spleen for each animal, and 64 from the brain of each animal) were randomly selected from SAB plates and analyzed for mating type based on mating with reference a and strains on V8 medium (55). Coinfections with KN99a/KN99NAT, KN99aNAT/KN99, KN99a/KN99aNAT, and KN99/KN99NAT used 10, 10, 5, and 5 mice, respectively, which were sacrificed at 21 days postinfection. More than 500 colonies from the lung, spleen, and brain from each animal were isolated on SAB plates and screened for NAT resistance on yeast peptone-dextrose plates containing 100 μg/ml nourseothricin to determine mating type. In control experiments, NAT resistance was found to be 100% stable during vegetative culture of KN99aNAT and KN99NAT. A subset of the colonies from coinfections was also analyzed for mating type based on mating with reference a and strains, and nourseothricin resistance cosegregated 100% with the appropriate a or allele of MAT. No instances of NAT loss or inactivation during infection were observed. The measured numbers of a and cells were compared to the expected numbers, assuming that both strains remained at the initial infection proportions (based on analysis of the initial inoculum). These raw data were then converted to percentages for presentation in Fig. 5 and 6.
Statistics. P values presented are two tailed. P values for the nematode and mouse survival data were derived from log rank tests. Wilcoxon rank sum analysis was used to determine significance for the amoeba, macrophage, and organ CFU data. For the coinfection experiments, the measured number of a and cells was compared to the expected number, assuming that both strains remained at the initial infection proportions (based on analysis of the initial inoculum), and Wilcoxon rank sum analysis of the raw data was used to determine P values. Thus, for the coinfections, the test of significance compared the ratios of the observed strains to the initial inoculum.
RESULTS
Further evidence that strains KN99a and KN99 are congenic with H99. We recently characterized mating in C. neoformans var. grubii using the serotype A mating type a strain 125.91 (55). A backcrossing scheme was designed using the mating characteristics of strain 125.91 to develop serotype A congenic strains in the H99 genetic background (Fig. 1A) (55). The resulting congenic strains KN99a and KN99 were compared to H99. The karyotypes of the KN99a and KN99 strains were examined and found to be identical to each other and to H99, indicating that the strains are congenic at the chromosome level (55). Restriction fragment length polymorphisms at two genetic loci were identified in the parental strains and the congenic strains and were found to have the same gene allele at both loci (55).
To further characterize the congenic strains at a molecular level, AFLP analysis with two independent primer pairs was used to create AFLP genotypes for each strain (Fig. 1B). Comparison of the AFLP genotypes of the parental strains 125.91, 8-1, KNA14, and H99 revealed polymorphic amplified DNA fragments between the strains. The congenic strains KN99a and KN99 had identical AFLP genotypes and were identical with H99 with both primer pairs tested, indicating a congenic relationship at a higher-resolution genomic level.
KN99a, KN99, and H99 showed no differences in several traits involved in virulence, such as the presence of melanin or capsule, prototrophy, or growth at 37°C (55). To examine the strains in more detail, we performed phenotype microarrays (4, 80; reviewed in reference 3). This analysis allows strain phenotypes to be quantitatively measured over time based on the physiological state of the cell. Cryptococcal cells are combined with a tetrazolium dye that absorbs electrons produced by yeast cell respiration. Reduction of the tetrazolium dye results in a color change that does not require growth of the cell. In the PM assay, >1,000 distinct culture conditions were selected to approximate a comprehensive scan of known cellular pathways. The effects of various carbon, nitrogen, phosphorus, and sulfur sources on the physiological state of the cell were examined. Cell respiration was also monitored in response to various osmotic conditions and in response to differing pHs. Finally, sensitivity to a wide range of chemicals, including antibiotics, respiratory inhibitors, and toxic metals, was examined. We have been able to observe distinct differences between mutant and wild-type C. neoformans strains using this method (X. Lin and J. Heitman, unpublished results). However, no consistent differences were observed with the serotype A var. grubii congenic strains, suggesting that KN99a, KN99, and H99 are also congenic at the phenotypic level.
Rates of survival of serotype A congenic strains are equivalent in nematodes and amoebae. The serotype A congenic strains KN99a and KN99 were previously shown to have virulence equivalent to that of strain H99 in two standard animal models of cryptococcosis (55). Because the vast majority of cases of cryptococcosis in humans are associated with mating type strains, we further characterized the potential role of mating type in the infectious process. C. neoformans is a soil-dwelling saprophyte that is not dependent on mammalian infection for reproduction or viability, yet many environmental isolates have the capacity to infect and cause disease in humans and experimental animals. Virulence is considered a complex phenotype that may be maintained as a result of selection. It has been hypothesized that factors that enable infection of mammals may have evolved to allow C. neoformans to survive predation by natural environmental predators, such as amoebae, nematodes, and insects, and the human virulence factors capsule and melanin have been shown to be important for the survival of C. neoformans in these organisms (2, 11, 53, 64, 66).
To determine whether mating type plays a role in the virulence of C. neoformans var. grubii strains in nematodes, we examined the survival of C. elegans exposed to KN99a, KN99, or H99. Previous studies with less-pathogenic serotype D var. neoformans congenic strains in C. elegans showed that the strain was more virulent than the congenic a strain (53). However, studies with the var. grubii congenic strains showed no significant difference in rates of survival of the nematodes (Fig. 2A) (P values were 0.20 for all comparisons). The var. grubii congenic strains were also examined for survival in amoebae. The congenic strains H99, KN99a, and KN99 were incubated with A. castellanii cells or a PBS control for 24 and 48 h. As shown in Fig. 2B, no significant difference in survival of the mating type strains H99 and KN99 was observed from that of the congenic mating type a strain KN99a in amoebae (P values were 0.49 for all comparisons).
Survival of serotype A congenic strains in macrophages. Humans are exposed to C. neoformans via inhalation of desiccated yeast cells or spores (67). Macrophages play a critical role in the early initial pulmonary immune response to cryptococcal infection and may also play a role in harboring the organism during latent infection as a facultative intracellular pathogen (5-7, 30, 45). Furthermore, the survival of Cryptococcus in macrophages has been linked to virulence (18, 22, 25, 26, 71). To study the role of mating type in infection and survival in macrophages, we examined the survival of the serotype A congenic strains KN99a and KN99 compared to that of H99 in the murine alveolar macrophage cell line MH-S and culture medium alone (Fig. 2C). Macrophages were stimulated with LPS and gamma interferon for 12 h and cocultured with the appropriate Cryptococcus strain for 1 hour, and afterwards, the macrophages were washed with culture medium to remove extracellular cryptococcal cells. After incubation for 24 h, the number of surviving cryptococcal cells, as measured by the number of CFU recovered, was examined for each strain. The serotype A congenic strains exhibited no difference in uptake by macrophages (data not shown) or survival in macrophages in vitro (P values were 0.4 for all comparisons).
Virulence of serotype A congenic strains in multiple murine models of cryptococcosis. To study the role of mating type in the virulence of the serotype A congenic strains, we compared the levels of virulence of the (H99) and a (KN99a) mating type strains in murine inbred lines, which can be divided into two groups: sensitive lines that are deficient in complement C5 secretion and resistant lines that have normal levels of complement C5 (59). Figure 3 shows three inbred lines commonly used for serotype A cryptococcal infections: A/JCr (sensitive), BALB/c (resistant), and C57BL/6 (resistant). A/JCr animals were most susceptible to intranasal infection, BALB/c animals exhibited intermediate susceptibility, and C57BL/6 animals were most resistant to infection by the serotype A congenic strains (P values were 0.009 for all comparisons). While there were significant differences in the susceptibilities of the strains to infection, all three inbred backgrounds showed no difference in levels of virulence for the mating type a and congenic strains (P values were 0.35 for all comparisons).
There are two predominant murine models of cryptococcosis based on inoculation via either inhalation or tail vein injection. Our previous studies showing that the serotype A congenic strains have equivalent levels of virulence utilized an inhalational model in which an initial pulmonary infection was followed by extrapulmonary dissemination, resulting in lethal CNS infection. Here we extended this comparison of the serotype A congenic strains H99 and KN99a using a mouse tail vein infection model. All animals succumbed to the infection in less than 10 days, and no significant difference in survival was observed between animals infected with the mating type strain H99 and those infected with the mating type a strain KN99a (data not shown).
Progression of C. neoformans var. grubii infection in a murine model. The previous animal models used survival (progression to severe morbidity) as the basis for the virulence of the serotype A congenic strains. While we found no difference in survival when mice were exposed to the serotype A congenic strains, we considered the possibility that the strains might cause disease in different ways. For example, one strain may result in a lethal pulmonary infection, while another strain may spread more quickly to the central nervous system to cause meningoencephalitis. Therefore, we examined the growth and dissemination of the strains in mice early in the infection (7 days), at an intermediate time point (14 days) and late in the infection (21 days). The number of CFU present in the lungs over time was used as a measure of pulmonary infection, while the number of CFU present in the spleen was used as a measure of the ability of the strain to spread hematogenously to other organs. Finally, numbers of CFU present in the brain were used to determine CNS infection (Fig. 4).
When the serotype A mating type strain H99 was compared to the congenic mating type a strain KN99a, H99 had slightly higher levels in the lung at the first time interval but no difference was observed in lung fungal burdens at the two later time intervals (P = 0.02 at 7 days, P = 0.15 at 14 days, P = 0.41 at 21 days). Large variability was observed between spleen fungal burdens over time, with KN99a having slightly higher levels at early and late time intervals (P = 0.01 at 7 days, P = 0.31 at 14 days, P = 0.02 at 21 days). Interestingly, higher levels of KN99a were present in the brain at the earliest time studied (7 days), but fungal burden in the brain increased at a higher rate, such that both strains had comparable numbers of brain CFU at the later times (P = 0.03 at 7 days, P = 0.31 at 14 days, P = 0.73 at 21 days). These data imply that a cells may initially disseminate to the brain faster than cells but that both cell types eventually produce equivalent levels of CNS infection. Based on symptoms, the animals appeared to have cryptococcal meningitis at the late time point and were without signs of a debilitating pulmonary infection. Because these observations correlate with equivalent CNS fungal burdens in the congenic a and strains, the animals appear to succumb to meningitis rather than pulmonary infection.
Coinfections with a and mating types. The serotype A congenic a and mating type strains have equivalent levels of virulence and similar modes of infection and disease progression when pure cultures are used as the infecting inoculum, yet strains cause the majority of cryptococcosis in humans. Interestingly, infection by multiple strains has been observed in humans (38, 58). The apparent paradox between animal studies and disease epidemiology could be explained if strains out-compete or inhibit a strains during simultaneous coinfection with both strains.
To examine this hypothesis, animals were infected with a mixture of the serotype A congenic a and strains and the proportions of each strain present in the lungs, spleen, and brain were monitored and compared to the proportions of each strain in the inoculum (Fig. 5 and 6). Infections with unmarked strains showed no difference in the proportions of the mating type a strain KN99a in the inoculum from the proportion of KN99a strains in the lungs and spleen in two independent coinfections at 7, 14, and 21 days postinfection (data not shown). Enumeration of total CFU from the lung and spleen were consistent with those observed in the individual infections (Fig. 4) at all time points. Interestingly, brain infection appeared to be initially delayed during coinfection with low or undetectable numbers of CFU at 7 days postinfection, but by days 14 and 21 postinfection, numbers of CFU increased and were consistent with those observed during pure culture infections (data not shown). At both 14 and 21 days postinfection, the proportions of KN99a strains in the brain were low in both assays. In 6 of the 10 mice examined, less than 5% of the strains analyzed were KN99a and >95% were H99, indicating a difference in CNS penetration or survival between the two mating types.
The strains analyzed in these coinfections had to be identified based on mating assay or PCR, which is laborious and limited the number of animals and strains analyzed. To extend these findings, large-scale analysis of coinfections was undertaken using marked strains and selective plating. A NAT resistance transgene was introduced into a noncoding region of the mating type locus of strains KN99a and KN99 to generate strains KN99aNAT and KN99NAT, respectively.
To verify that the transgenes were phenotypically neutral and did not alter the virulence of the NAT-marked strains, the NAT-marked strains were coinfected with their parent strain and the proportions of NAT-marked strains in the lung, spleen, and brain were analyzed at 21 days postinfection (Fig. 5B). The proportion of KN99aNAT cells present in the lungs, spleen, and brain in the aNAT/a mixture was equivalent to the starting inoculum (represented by the dotted line), demonstrating that KN99aNAT survives in vivo at levels equivalent to KN99a (lung, P = 0.75; spleen, P = 0.86; brain, P = 0.25). However, most animals had one strain type that predominated in the brain, with roughly half of the animals having a vast majority of marked aNAT cells and the other half having unmarked a cells in the CNS. This observation is consistent with a single cell or a small group of founder cells producing infection of the brain. Similar results were observed with NAT/ coinfections (data not shown). Thus, these data demonstrate that the dominant marker transgenes are phenotypically neutral in vivo.
We used the marked strains to compare differences in disease progression between a and using larger numbers of animals and CFU. The KN99aNAT strain was coinfected with KN99, and the proportions of NAT-marked strains in the lung, spleen, and brain were analyzed at 21 days postinfection. More than 500 colonies from each organ were isolated and then individually screened for NAT resistance by selective plating. Figure 5A shows that the proportions of KN99aNAT cells present in the lungs and spleen were equivalent to the initial inoculum (represented by the dotted line) but that the level of KN99aNAT cells present in the brain was significantly lower than the inoculum level in the aNAT/ coinfection (lung, P = 0.55; spleen, P = 0.21; brain, P < 0.001). Similar results were observed with the NAT/a coinfection (data not shown). Thus, this larger data set corroborates the pattern observed in coinfection studies with unmarked a and strains. We conclude that there is a significant difference between congenic a and cells with respect to their ability to establish CNS infection during coinfection of mice.
The difference between a and cell accumulation in the CNS during coinfection may be due to either dissemination to the CNS or growth within the CNS. To distinguish between these two possibilities, animals were infected intracerebrally with either KN99a or KN99 or coinfected with KN99aNAT/KN99. Intracerebral infections showed no difference in rates of survival of animals (data not shown) or in brain fungal burdens between a and pure cultures or with aNAT/ coinfections of the CNS (Fig. 6A) (P values were 0.81, 0.24, and 0.20 for KN99a/KN99, KN99a/KN99a coinfection, and KN99/KN99 coinfection, respectively) and are consistent with the intranasal infections. In contrast to what occurs with intranasal coinfections where cells predominate in the CNS, Fig. 6B shows that when aNAT/ cells were coinfected intracerebrally, no difference in the levels of accumulation of the strains was observed (P = 0.99). Analogous results were observed with a rabbit meningitis coinfection model where the strains were inoculated directly into the cerebrospinal fluid (data not shown). These data show that a and cells grow equally well in the CNS and imply that cells predominate in the CNS during coinfection due to a difference in dissemination to the CNS.
DISCUSSION
Rates of AIDS infections are still on the rise in many areas of the world, including Africa and Asia (76, 77). This pandemic has led to an increase in the abundance and severity of infections by opportunistic fungal pathogens. Many virulence factors of one such pathogen, C. neoformans, have been identified. Some factors, such as the polysaccharide capsule and melanin, are clearly directly linked to the pathogenicity of C. neoformans because cells lacking these factors exhibit a significant decrease in virulence (48, 56, 79). Mating type has been proposed to be a virulence factor in C. neoformans. Epidemiological studies have shown that clinical isolates are severely biased towards the mating type (41) and that some genes associated with mating type, such as STE12a and STE12, have a role in the virulence of the serotype D var. neoformans lineage, whereas other genes, such as SXI1, STE20, and STE11, do not (12, 14, 17, 21, 36, 75). In heterogeneous populations of serotype D, no clear distinction between the levels of virulence of a and strains could be determined, although strains killed more mice at lower inoculums (40). To examine the role of mating type in virulence, congenic strains were generated by repeated backcrossing. Analysis of var. neoformans serotype D strains revealed that the congenic strain was more virulent than the a strain (42). However, clinical infections with var. neoformans strains account for a limited number of cases compared to cases caused by var. grubii serotype A strains. In contrast to serotype D, congenic a and var. grubii strains have equivalent levels of virulence in cellular (macrophage), heterologous (nematodes, amoebae), and mammalian (both mouse and rabbit) models of cryptococcosis (reference 55 and this study).
Examination of the virulence of var. grubii serotype A congenic strains in multiple inbred-mouse lines showed virulence consistent with the var. neoformans serotype D results of Rhodes et al. (59). A/JCr mice were more susceptible to the serotype A congenic strains than the relatively resistant BALB/c and C57BL/6 lines. The serotype A congenic strains were also more virulent in the BALB/c inbred-mouse line than in C57BL/6 mice. While we observed a difference in the overall rates of survival between the inbred-mouse lines, the congenic strains had equivalent levels of virulence in all of the inbred lines tested, showing that the similarity in the virulence potentials of the var. grubii congenic strains is not specific to the inbred-animal background.
While cryptococcal meningoencephalitis is the most common life-threatening manifestation of cryptococcosis, the lungs are the second-most-relevant site of cryptococcal infection (10). In animal models of cryptococcal meningitis, neurological symptoms, such as extension of the cerebral portion of the cranium, abnormal gait, paralysis, and lethargy, are observed, whereas animals with pulmonary cryptococcosis exhibit respiratory distress. However, in animal models based on lethal infection, the difference between cryptococcal meningitis and pulmonary cryptococcosis could be overlooked. To differentiate between these two forms of cryptococcosis, numbers of CFU from lung, spleen, and brain tissue (to represent pulmonary infection, hematogenous spread to other organs, and CNS infection, respectively) were compared from animals inoculated with the serotype A congenic a and mating type strains KN99a and H99, respectively. Mice infected with KN99a had higher numbers of CFU at initial time intervals in the brain, suggesting either a higher rate of dissemination from the initial pulmonary infection to the CNS or reduced clearance in the CNS. However, both KN99a and H99 had equivalent numbers of CFU in the brain by the intermediate and final time intervals studied, and all animals exhibited symptoms of CNS infection, indicating that the basis of mortality is the same in both cases.
The most notable virulence difference between the serotype A congenic a and strains was observed during coinfection. In these experiments the ability of one strain to out-compete or inhibit the growth of the other strain would manifest itself as an increase in the proportion of the dominant strain. During intranasal coinfections, the strain predominated in the brain even though the levels of a and strains were equivalent in the lung and spleen. Because intranasal coinfections with the unmarked strains and larger-scale coinfections with marked strains showed that a and strains disseminated equally well from the lung to the spleen, we conclude that the difference observed in the CNS is not simply due to an inability of a cells to disseminate from the lung but is specific to the CNS. Intracerebral infections showed that both strains had similar levels of persistence in the brain during coinfection. Thus, cell predominance in the CNS appears to be due to entry into and not persistence in the brain.
A number of hypotheses could explain the observation that cells had higher levels of entry into the CNS during coinfection. First, an interaction between and a cells could enhance the ability of the cells to invade the brain (Fig. 7A). For example, laccase is critical for the survival of Cryptococcus in the brain, and urease has been shown to enhance invasion of the central nervous system by Cryptococcus (57, 60). Increased production of these enzymes in the strain or their decreased production in the a strain may potentially result in a higher proportion of strains in the CNS.
A second hypothesis is that interactions between and a cells results in a physical alteration of the a cells that inhibits their ability to enter the CNS (Fig. 7B). In humans, C. neoformans replicates in the lung and then spreads via the bloodstream or lymphatic system to other organs as free cells or internalized in mononuclear cells, such as macrophages, that travel to microcapillary beds in the brain (16, 28, 57, 61). At the blood-brain barrier, C. neoformans can transverse microvascular endothelial cells either directly or via a Trojan horse mechanism within mononuclear phagocytes (13, 15, 61). Mating type a strains are known to generate large, bulbous cells in response to pheromone in serotype D var. neoformans, and the pheromone MF1 gene has been shown to be expressed in the brain during serotype A var. grubii infection as well as inside C. elegans (23, 53). While enlargement of a cells has not been observed in the laboratory for var. grubii serotype A, it is possible that a cells respond to pheromone produced by coinfection with cells in vivo. Pheromone-induced cellular changes in a cells may inhibit their entry into microcapillaries in the brain or their subsequent transversal of endothelial cells but not affect their ability to spread to other organs, such as the spleen.
Finally, as a third hypothesis, we cannot rule out the possibility that a and cells do not communicate with each other but interact differently with the host. But as the difference in virulence occurs only during coinfection, the data suggest a three-way dialog between the host and both a and cells of the fungal pathogen (Fig. 7C). For example, one strain could elicit an immune response from the host that affects the other strain disproportionately, which manifests itself as a virulence difference in the brain.
We have detected no evidence of mating or a- cell fusion in vivo to produce a/ diploid cells during coinfection (this work and Y.-S. Bahn and J. Heitman, unpublished results). Thus, these conditions may preclude mating or counterselect against a/ diploid or dikaryotic products of mating. This said, early events in a- cell signaling that transpire in vitro may occur in vivo, as several conditions conducive to mating (darkness, nutrient limitation, and pheromone expression) are present in vivo and could result in local a- cell signaling that differentially impacts fungal cell interactions with the host.
It is possible that strains are the predominant clinical isolates because they are more prevalent in the environment. If this is the case, then areas with high levels of environmental a strains would be expected to have roughly equal levels of a strains as clinical isolates. Examination of serotype A clinical strain populations from sub-Saharan countries show that up to 10% of the clinical isolates can be the a mating type, and a population of clinical isolates from Botswana showed evidence of sexual recombination (44, 46). It will be of interest to establish what proportion of environmental strains in sub-Saharan Africa is represented by a strains.
The predominance of cells in the CNS during mixed infections could also contribute to the predominance of strains in clinical isolates, which are derived predominantly from the cerebrospinal fluid. It is possible that strains appear to be predominant because coinfections from mixed a and spores, produced as a result of mating, result in CNS infections with a higher proportion of strains. Furthermore, most clinical isolates are single-colony isolates. Thus, if the cerebrospinal fluid has a higher proportion of cells, there is also a higher probability that the resulting single-colony isolate will be an strain. In the few cases where serial isolates have been collected from the same patient, a second strain has been isolated as often as 29% of the time (58). These patients could have been reinfected with a different strain or coinfected with multiple strains. Examination of multiple isolates from the same patient, preferably from lung or sputum cultures from patients in whom multiple strains might be expected to be present in roughly equal proportions, will provide more-accurate data on levels of coinfection and the genetic nature of the coinfecting strains with respect to the mating type locus and other genes linked to the virulence potential of Cryptococcus.
ACKNOWLEDGMENTS
We thank Cristl Arndt and A. Lisa De Obaldia for technical assistance, James Fraser for assistance with marked strain development, Alex Idnurm and Yong-Sun Bahn for animal assistance, and other members of the Duke University Mycology Research Unit (DUMRU) for their helpful comments and discussions.
This work was supported by NIAID R01 grant AI50113 to Joseph Heitman and NIAID R01 grant AI28388 to John Perfect. Daniel Benjamin received support from NICHD HD044799-01. Gary Cox was a Burroughs Wellcome New Investigator, and Joseph Heitman is a Burroughs Wellcome Scholar in Molecular Pathogenic Mycology and an investigator of the Howard Hughes Medical Institute.
REFERENCES
1. Aberg, J. A., and W. G. Powderly. 1997. Cryptococcal disease: implications of recent clinical trials on treatment and management. AIDS Clin. Rev. 1997-1998:229-248.
2. 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.
3. Bochner, B. R. 2003. New technologies to assess genotype-phenotype relationships. Nat. Rev. Genet. 4:309-314.
4. Bochner, B. R., P. Gadzinski, and E. Panomitros. 2001. Phenotype microarrays for high-throughput phenotypic testing and assay of gene function. Genome Res. 11:1246-1255.
5. Bolanos, B., and T. G. Mitchell. 1989. Killing of Cryptococcus neoformans by rat alveolar macrophages. J. Med. Vet. Mycol. 27:219-228.
6. Bolanos, B., and T. G. Mitchell. 1989. Phagocytosis and killing of Cryptococcus neoformans by rat alveolar macrophages in the absence of serum. J. Leukoc. Biol. 46:521-528.
7. Bolanos, B., and T. G. Mitchell. 1989. Phagocytosis of Cryptococcus neoformans by rat alveolar macrophages. J. Med. Vet. Mycol. 27:203-217.
8. Brandt, M. E., L. C. Hutwagner, L. A. Klug, W. S. Baughman, D. Rimland, E. A. Graviss, R. J. Hamill, C. Thomas, P. G. Pappas, A. L. Reingold, R. W. Pinner, and the Cryptococcal Disease Active Surveillance Group. 1996. Molecular subtype distribution of Cryptococcus neoformans in four areas of the United States. J. Clin. Microbiol. 34:912-917.
9. Casadevall, A., J. Mukherjee, R. Yuan, and J. Perfect. 1994. Management of injuries caused by Cryptococcus neoformans-contaminated needles. Clin. Infect. Dis. 19:951-953.
10. Casadevall, A., and J. R. Perfect. 1998. Cryptococcus neoformans. ASM Press, Washington, D.C.
11. Casadevall, A., J. N. Steenbergen, and J. D. Nosanchuk. 2003. ‘Ready made’ virulence and ‘dual use’ virulence factors in pathogenic environmental fungi—the Cryptococcus neoformans paradigm. Curr. Opin. Microbiol. 6:332-337.
12. Chang, Y. C., L. A. Penoyer, and K. J. Kwon-Chung. 2001. The second STE12 homologue of Cryptococcus neoformans is MATa-specific and plays an important role in virulence. Proc. Natl. Acad. Sci. USA 98:3258-3263.
13. Chang, Y. C., M. F. Stins, M. J. McCaffery, G. F. Miller, D. R. Pare, T. Dam, M. Paul-Satyaseela, K. S. Kim, and K. J. Kwon-Chung. 2004. Cryptococcal yeast cells invade the central nervous system via transcellular penetration of the blood-brain barrier. Infect. Immun. 72:4985-4995.
14. Chang, Y. C., B. L. Wickes, G. F. Miller, L. A. Penoyer, and K. J. Kwon-Chung. 2000. Cryptococcus neoformans STE12 regulates virulence but is not essential for mating. J. Exp. Med. 191:871-882.
15. Chen, S. H., M. F. Stins, S. H. Huang, Y. H. Chen, K. J. Kwon-Chung, Y. Chang, K. S. Kim, K. Suzuki, and A. Y. Jong. 2003. Cryptococcus neoformans induces alterations in the cytoskeleton of human brain microvascular endothelial cells. J. Med. Microbiol. 52:961-970.
16. Chretien, F., O. Lortholary, I. Kansau, S. Neuville, F. Gray, and F. Dromer. 2002. Pathogenesis of cerebral Cryptococcus neoformans infection after fungemia. J. Infect. Dis. 186:522-530.
17. Clarke, D. L., G. L. Woodlee, C. M. McClelland, T. S. Seymour, and B. L. Wickes. 2001. The Cryptococcus neoformans STE11 gene is similar to other fungal mitogen-activated protein kinase kinase kinase (MAPKKK) genes but is mating type specific. Mol. Microbiol. 40:200-213.
18. Cox, G. M., T. S. Harrison, H. C. McDade, C. P. Taborda, G. Heinrich, A. Casadevall, and J. R. Perfect. 2003. Superoxide dismutase influences the virulence of Cryptococcus neoformans by affecting growth within macrophages. Infect. Immun. 71:173-180.
19. Currie, B. P., and A. Casadevall. 1994. Estimation of the prevalence of cryptococcal infection among patients infected with the human immunodeficiency virus in New York City. Clin. Infect. Dis. 19:1029-1033.
20. Davidson, R. C., J. R. Blankenship, P. R. Kraus, M. D. Berrios, C. M. Hull, C. D'Souza, P. Wang, and J. Heitman. 2002. A PCR-based strategy to generate integrative targeting alleles with large regions of homology. Microbiology 148:2607-2615.
21. Davidson, R. C., C. B. Nichols, G. M. Cox, J. R. Perfect, and J. Heitman. 2003. A MAP kinase cascade composed of cell type specific and non-specific elements controls mating and differentiation of the fungal pathogen Cryptococcus neoformans. Mol. Microbiol. 49:469-485.
22. DeJesus-Berrios, M., L. Liu, J. Nussbaum, G. M. Cox, J. S. Stamler, and J. Heitman. 2003. Enzymes that counteract nitrosative challenge promote fungal virulence. Curr. Biol. 13:1963-1968.
23. Del Poeta, M., D. L. Toffaletti, T. H. Rude, S. D. Sparks, J. Heitman, and J. R. Perfect. 1999. Cryptococcus neoformans differential gene expression detected in vitro and in vivo with green fluorescent protein. Infect. Immun. 67:1812-1820.
24. Edmond, M. B., S. E. Wallace, D. K. McClish, M. A. Pfaller, R. N. Jones, and R. P. Wenzel. 1999. Nosocomial bloodstream infections in United States hospitals: a three-year analysis. Clin. Infect. Dis. 29:239-244.
25. Feldmesser, M., Y. Kress, P. Novikoff, and A. Casadevall. 2000. Cryptococcus neoformans is a facultative intracellular pathogen in murine pulmonary infection. Infect. Immun. 68:4225-4237.
26. Feldmesser, M., S. Tucker, and A. Casadevall. 2001. Intracellular parasitism of macrophages by Cryptococcus neoformans. Trends Microbiol. 9:273-278.
27. Fraser, J. A., R. L. Subaran, C. B. Nichols, and J. Heitman. 2003. Recapitulation of the sexual cycle of the primary fungal pathogen Cryptococcus neoformans variety gattii: implications for an outbreak on Vancouver Island. Eukaryot. Cell 2:1036-1045.
28. Garcia-Hermoso, D., G. Janbon, and F. Dromer. 1999. Epidemiological evidence for dormant Cryptococcus neoformans infection. J. Clin. Microbiol. 37:3204-3209.
29. Gordon, S. B., A. L. Walsh, M. Chaponda, M. A. Gordon, D. Soko, M. Mbwvinji, M. E. Molyneux, and R. C. Read. 2000. Bacterial meningitis in Malawian adults: pneumococcal disease is common, severe, and seasonal. Clin. Infect. Dis. 31:53-57.
30. Granger, D. L., J. B. Hibbs, Jr., J. R. Perfect, and D. T. Durack. 1988. Specific amino acid (L-arginine) requirement for the microbiostatic activity of murine macrophages. J. Clin. Investig. 81:1129-1136.
31. Hajjeh, R. A., L. A. Conn, D. S. Stephens, W. Baughman, R. Hamill, E. Graviss, P. G. Pappas, C. Thomas, A. Reingold, G. Rothrock, L. C. Hutwagner, A. Schuchat, M. E. Brandt, R. W. Pinner, et al. 1999. Cryptococcosis: population-based multistate active surveillance and risk factors in human immunodeficiency virus-infected persons. J. Infect. Dis. 179:449-454.
32. Hakim, J. G., I. T. Gangaidzo, R. S. Heyderman, J. Mielke, E. Mushangi, A. Taziwa, V. J. Robertson, P. Musvaire, and P. R. Mason. 2000. Impact of HIV infection on meningitis in Harare, Zimbabwe: a prospective study of 406 predominantly adult patients. AIDS 14:1401-1407.
33. Haynes, K. A., D. J. Sullivan, D. C. Coleman, J. C. Clarke, R. Emilianus, C. Atkinson, and K. J. Cann. 1995. Involvement of multiple Cryptococcus neoformans strains in a single episode of cryptococcosis and reinfection with novel strains in recurrent infection demonstrated by random amplification of polymorphic DNA and DNA fingerprinting. J. Clin. Microbiol. 33:99-102.
34. Heitman, J., B. Allen, J. A. Alspaugh, and K. J. Kwon-Chung. 1999. On the origins of the congenic MAT and MATa strains of the pathogenic yeast Cryptococcus neoformans. Fungal Genet. Biol. 28:1-5.
35. Heyderman, R. S., I. T. Gangaidzo, J. G. Hakim, J. Mielke, A. Taziwa, P. Musvaire, V. J. Robertson, and P. R. Mason. 1998. Cryptococcal meningitis in human immunodeficiency virus-infected patients in Harare, Zimbabwe. Clin. Infect. Dis. 26:284-289.
36. Hull, C. M., G. M. Cox, and J. Heitman. 2004. The -specific cell identity factor Sxi1 is not required for virulence of Cryptococcus neoformans. Infect. Immun. 72:3643-3645.
37. Hull, C. M., R. C. Davidson, and J. Heitman. 2002. Cell identity and sexual development in Cryptococcus neoformans are controlled by the mating-type-specific homeodomain protein Sxi1. Genes Dev. 16:3046-3060.
38. Igreja, R. P., S. Lazera Mdos, B. Wanke, M. C. Galhardo, S. E. Kidd, and W. Meyer. 2004. Molecular epidemiology of Cryptococcus neoformans isolates from AIDS patients of the Brazilian city, Rio de Janeiro. Med. Mycol. 42:229-238.
39. Keller, S. M., M. A. Viviani, M. C. Esposto, M. Cogliati, and B. L. Wickes. 2003. Molecular and genetic characterization of a serotype A MATa Cryptococcus neoformans isolate. Microbiology 149:131-142.
40. Kwon-Chung, J. K., and W. B. Hill. 1981. Sexuality and pathogenicity of Filobasidiella neoformans (Cryptococcus neoformans), p. 243-250. In C. Vanbreuseghem (ed.), Sexuality and pathogenicity of fungi. Masson, Paris, France.
41. Kwon-Chung, K. J., and J. E. Bennett. 1978. Distribution of alpha and a mating types of Cryptococcus neoformans among natural and clinical isolates. Am. J. Epidemiol. 108:337-340.
42. Kwon-Chung, K. J., J. C. Edman, and B. L. Wickes. 1992. Genetic association of mating types and virulence in Cryptococcus neoformans. Infect. Immun. 60:602-605.
43. Lengeler, K. B., D. S. Fox, J. A. Fraser, A. Allen, K. Forrester, F. S. Dietrich, and J. Heitman. 2002. Mating-type locus of Cryptococcus neoformans: a step in the evolution of sex chromosomes. Eukaryot. Cell 1:704-718.
44. Lengeler, K. B., P. Wang, G. M. Cox, J. R. Perfect, and J. Heitman. 2000. Identification of the MATa mating-type locus of Cryptococcus neoformans reveals a serotype A MATa strain thought to have been extinct. Proc. Natl. Acad. Sci. USA 97:14455-14460.
45. Li, R. K., and T. G. Mitchell. 1997. Induction of interleukin-6 mRNA in rat alveolar macrophages by in vitro exposure to both Cryptococcus neoformans and anti-C. neoformans antiserum. J. Med. Vet. Mycol. 35:327-334.
46. Litvintseva, A. P., R. E. Marra, K. Nielsen, J. Heitman, R. Vilgalys, and T. G. Mitchell. 2003. Evidence of sexual recombination among Cryptococcus neoformans serotype A isolates in sub-Saharan Africa. Eukaryot. Cell 2:1162-1168.
47. Liu, H., T. R. Cottrell, L. M. Pierini, W. E. Goldman, and T. L. Doering. 2002. RNA interference in the pathogenic fungus Cryptococcus neoformans. Genetics 160:463-470.
48. Liu, L., R. P. Tewari, and P. R. Williamson. 1999. Laccase protects Cryptococcus neoformans from antifungal activity of alveolar macrophages. Infect. Immun. 67:6034-6039.
49. Lodge, J. K., E. Jackson-Machelski, D. L. Toffaletti, J. R. Perfect, and J. I. Gordon. 1994. Targeted gene replacement demonstrates that myristoyl-CoA:protein N-myristoyltransferase is essential for viability of Cryptococcus neoformans. Proc. Natl. Acad. Sci. USA 91:12008-12012.
50. Maher, D., and H. Mwandumba. 1994. Cryptococcal meningitis in Lilongwe and Blentyre, Malawi. J. Infect. 28:59-64.
51. Malliaris, S. D., J. N. Steenbergen, and A. Casadevall. 2004. Cryptococcus neoformans var. gattii can exploit Acanthamoeba castellanii for growth. Med. Mycol. 42:149-158.
52. Mitchell, T. G., and J. R. Perfect. 1995. Cryptococcosis in the era of AIDS—100 years after the discovery of Cryptococcus neoformans. Clin. Microbiol. Rev. 8:515-548.
53. 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.
54. 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.
55. 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.
56. Nosanchuk, J. D., P. Valadon, M. Feldmesser, and A. Casadevall. 1999. Melanization of Cryptococcus neoformans in murine infection. Mol. Cell. Biol. 19:745-750.
57. Olszewski, M. A., M. C. Noverr, G. H. Chen, G. B. Toews, G. M. Cox, J. R. Perfect, and G. B. Huffnagle. 2004. Urease expression by Cryptococcus neoformans promotes microvascular sequestration, thereby enhancing central nervous system invasion. Am. J. Pathol. 164:1761-1771.
58. Pfaller, M., J. Zhang, S. Messer, M. Tumberland, E. Mbidde, C. Jessup, and M. Ghannoum. 1998. Molecular epidemiology and antifungal susceptibility of Cryptococcus neoformans isolates from Ugandan AIDS patients. Diagn. Microbiol. Infect. Dis. 32:191-199.
59. Rhodes, J. C., L. S. Wicker, and W. J. Urba. 1980. Genetic control of susceptibility to Cryptococcus neoformans in mice. Infect. Immun. 29:494-499.
60. 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.
61. Santangelo, R., H. Zoellner, T. Sorrell, C. Wilson, C. Donald, J. Djordjevic, Y. Shounan, and L. Wright. 2004. Role of extracellular phospholipases and mononuclear phagocytes in dissemination of cryptococcosis in a murine model. Infect. Immun. 72:2229-2239.
62. Sorvillo, F., G. Beall, P. A. Turner, V. L. Beer, A. A. Kovacs, and P. R. Kerndt. 1997. Incidence and factors associated with extrapulmonary cryptococcosis among persons with HIV infection in Los Angeles County. AIDS 11:673-679.
63. Speed, B., and D. Dunt. 1995. Clinical and host differences between infections with the two varieties of Cryptococcus neoformans. Clin. Infect. Dis. 21:28-34.
64. Steenbergen, J. N., and A. Casadevall. 2003. The origin and maintenance of virulence for the human pathogenic fungus Cryptococcus neoformans. Microbes Infect. 5:667-675.
65. Steenbergen, J. N., J. D. Nosanchuk, S. D. Malliaris, and A. Casadevall. 2003. Cryptococcus neoformans virulence is enhanced after growth in the genetically malleable host Dictyostelium discoideum. Infect. Immun. 71:4862-4872.
66. 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.
67. Sukroongreung, S., K. Kitiniyom, C. Nilakul, and S. Tantimavanich. 1998. Pathogenicity of basidiospores of Filobasidiella neoformans var. neoformans. Med. Mycol. 36:419-424.
68. Sullivan, D., K. Haynes, G. Moran, D. Shanley, and D. Coleman. 1996. Persistence, replacement, and microevolution of Cryptococcus neoformans strains in recurrent meningitis in AIDS patients. J. Clin. Microbiol. 34:1739-1744.
69. Tintelnot, K., K. Lemmer, H. Losert, G. Schar, and A. Polak. 2004. Follow-up of epidemiological data of cryptococcosis in Austria, Germany and Switzerland with special focus on the characterization of clinical isolates. Mycoses 47:455-464.
70. Toffaletti, D. L., T. H. Rude, S. A. Johnston, D. T. Durack, and J. R. Perfect. 1993. Gene transfer in Cryptococcus neoformans by use of biolistic delivery of DNA. J. Bacteriol. 175:1405-1411.
71. Tucker, S. C., and A. Casadevall. 2002. Replication of Cryptococcus neoformans in macrophages is accompanied by phagosomal permeabilization and accumulation of vesicles containing polysaccharide in the cytoplasm. Proc. Natl. Acad. Sci. USA 99:3165-3170.
72. Vallim, M. A., L. Fernandes, and J. A. Alspaugh. 2004. The RAM1 gene encoding a protein-farnesyltransferase beta-subunit homologue is essential in Cryptococcus neoformans. Microbiology 150:1925-1935.
73. Viviani, M. A., M. C. Esposto, M. Cogliati, M. T. Montagna, and B. L. Wickes. 2001. Isolation of a Cryptococcus neoformans serotype A MATa strain from the Italian environment. Med. Mycol. 39:383-386.
74. Viviani, M. A., R. Nikolova, M. C. Esposto, G. Prinz, and M. Cogliati. 2003. First European case of serotype A MATa Cryptococcus neoformans infection. Emerg. Infect. Dis. 9:1179-1180.
75. Wang, P., C. B. Nichols, K. B. Lengeler, M. E. Cardenas, G. M. Cox, J. R. Perfect, and J. Heitman. 2002. Mating-type-specific and nonspecific PAK kinases play shared and divergent roles in Cryptococcus neoformans. Eukaryot. Cell 1:257-272.
76. World Heath Organization. 2003. HIV/AIDS Epidemiological surveillance update for the W.H.O. African region. World Health Organization, Geneva, Switzerland.
77. World Health Organization. 2001. HIV/AIDS in Asia and the Pacific Region. World Health Organization, Geneva, Switzerland.
78. Wickes, B. L., and J. K. Kwon-Chung. 2002. Genetic basis of pathogenicity in Cryptococcus neoformans, p. 25-49. In R. L. Calderone (ed.), Fungal pathogenesis: principles and clinical applications. Marcell Dekker, New York, N.Y.
79. Wilder, J. A., G. K. Olson, Y. C. Chang, K. J. Kwon-Chung, and M. F. Lipscomb. 2002. Complementation of a capsule deficient Cryptococcus neoformans with Cap64 restores virulence in a murine lung infection. Am. J. Respir. Cell. Mol. Biol. 26:306-314.
80. Zhou, L., X. H. Lei, B. R. Bochner, and B. L. Wanner. 2003. Phenotype microarray analysis of Escherichia coli K-12 mutants with deletions of all two-component systems. J. Bacteriol. 185:4956-4972.(Kirsten Nielsen, Gary M. )