HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Clancy, C. J. Articles by Graybill, J. R. Search for Related Content PubMed PubMed Citation Articles by Clancy, C. J. Articles by Graybill, J. R. Agricola Articles by Clancy, C. J. Articles by Graybill, J. R.

Cryptococcus neoformans var. grubii isolates recovered from persons with AIDS demonstrate a wide range of virulence during murine meningoencephalitis that correlates with the expression of certain virulence factors

Cornelius J. Clancy^1,3,

M. Hong Nguyen^1,3,

¹ Department of Medicine, Division of Infectious Diseases, University of Florida College of Medicine, FL, USA
² Department of Pathology, University of Florida College of Medicine, FL, USA
³ VA Medical Center, Gainesville, FL, USA
⁴ University of Texas Health Science Center, and the Audie Murphy Memorial Hospital, San Antonio, TX, USA

Correspondence
Cornelius J. Clancy
clancyn{at}medicine.ufl.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Cryptococcus neoformans is a common cause of meningoencephalitis among AIDS patients. Several C. neoformans virulence factors have been identified, but the relative importance of particular factors is unknown. This study examined the corrrelation of the virulence of 18 C. neoformans var. grubii isolates from AIDS patients with the expression of several well-described virulence factors. The LD₅₀ at 15 days after intracranial inoculation of ICR mice was <100 c.f.u. for 22 % of isolates, 100–1000 for 28 %, 1000–10 000 for 11 % and >20 000 for 39 %. Higher cryptococcal concentrations in brains were noted for isolates with lower LD₅₀ (P=0.002). In survival studies, no immunocompetent BALB/c mice (nu/–) infected with 3xLD₅₀ of three virulent isolates (LD₅₀=62, 99, 1280) survived beyond 23 days, whereas 100 %, 90 % and 90 % of mice infected with 20 000 c.f.u. of three hypovirulent isolates (LD₅₀>20 000) survived for 60 days (P<0.0001). Even among BALB/c nude (nu/nu) mice, survival rates over 60 days were 100 %, 70 % and 50 %, respectively, for the hypovirulent isolates. Growth rate at 37 °C and capsule size within brains correlated with LD₅₀ by univariate (P=0.0001 and 0.028, respectively) and multivariate (P=0.017 and 0.016, respectively) analyses. There was no correlation between LD₅₀ and capsule size in vitro, phospholipase activity, melanin formation, proteinase activity and fluconazole MIC. In conclusion, AIDS patients are susceptible to infection by C. neoformans isolates of wide-ranging virulence, including isolates that are markedly hypovirulent. The virulence of a given isolate reflects a composite of factors rather than the contribution of a dominant factor. Growth at 37 °C and capsule size in vivo make particularly important contributions.

These authors contributed equally to this work.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Cryptococcus neoformans is a yeast-like fungus that is generally found in soil with a high content of avian guano or vegetative debris (Casadevall & Perfect, 1998; Mitchell & Perfect, 1995). Infections of humans, believed to be acquired by inhalation of small yeast cells or basidiospores, are often clinically silent. In fact, the most commonly recognized manifestation of cryptococcal disease (cryptococcosis) is not pneumonia but rather meningitis or meningoencephalitis. Cryptococcal central nervous system (CNS) infections and other forms of cryptococcosis are largely, but not exclusively, diseases of persons with AIDS or other immunosuppressed conditions. The pathogenesis of cryptococcosis reflects the interaction between host susceptibility, immune response to the infecting organism, and the virulence potential of the given C. neoformans strain. In North America, over 90 % of cases are caused by strains of serotype A, C. neoformans var. grubii, which are believed to be more pathogenic than strains of serotype D, C. neoformans var. neoformans (Mitchell & Perfect, 1995; Casadevall & Perfect, 1998).

Several factors have been identified that contribute to virulence of C. neoformans strains during animal models of cryptococcosis. Among the best characterized are the ability to grow at 37 °C, the presence of a protective polysaccharide capsule, being of the -mating type, the ability to synthesize the antioxidant pigment melanin, and extracellular phospholipase and proteinase activities. In general, investigators have implicated these factors in virulence by studying them individually, either by using isogenic gene disruption and reconstitution strains or by comparing strains that differ in expression of a phenotype of interest. As for other fungal pathogens, however, it is likely that the virulence of C. neoformans depends upon the coordinated expression of multiple factors (Kozel, 1995; Blackstock et al., 1999). Along these lines, it has been suggested that C. neoformans and other medically relevant fungi contain a set of genes that express a subset of traits constituting a composite virulence phenotype (Cutler, 1991; Casadevall & Perfect, 1998). Under given conditions, selected factors in the so-called virulence set might not be absolutely necessary to cause disease; rather, a critical number of factors might need to act in concert. The fact that some C. neoformans isolates demonstrate tissue tropism suggests that their composite virulence phenotypes are adapted to particular sites of infection (Grosse et al., 1975; Dixon & Polak, 1986; Van Cutsem et al., 1986; Fromtling et al., 1988; Barchiesi et al., 2005). It is likely that other isolates are more generally equipped to cause disease at diverse sites, although this hypothesis has not been widely tested.

It is clear from animal models of cryptococcosis that individual clinical and environmental C. neoformans strains differ in relative virulence (Vanbreuseghem, 1967; Hasenclever & Mitchell, 1960; Kwon-Chung et al., 1992; Blasi et al., 1992, 1995; Blackstock & Murphy, 1997; Curtis et al., 1994; Blackstock et al., 1999; Fries & Casadevall, 1998). Comparing two C. neoformans var. grubii strains of variable virulence in the lungs of mice, Blackstock et al. (1999) showed that the more virulent isolate had higher expression of several virulence factors. However, since the expression of multiple factors by a larger number of well-characterized strains recovered from unique patients has not been studied, the relative importance of particular virulence factors is unknown. The primary objective of the present study was to correlate the virulence of 18 C. neoformans var. grubii isolates recovered from the cerebrospinal fluid (CSF) of AIDS patients with the expression of several well-characterized virulence factors. We assessed virulence by measuring mortality rates during murine meningoencephalitis and tissue burdens within infected brains. In addition, we studied tissue tropism for two isolates by comparing their levels of virulence in the CNS with their abilities to cause disease in the lungs after intranasal inoculation.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Organisms.
Eighteen isolates of C. neoformans var. grubii were obtained from the Fungus Testing Laboratory at the University of Texas Health Sciences Center at San Antonio, Texas. These isolates were recovered from the CSF of HIV-infected patients. Upon arrival at the Fungus Testing Laboratory, the isolates were streaked onto Sabouraud-dextrose agar (SDA) plates and incubated at 35 °C for 24–48 h. They were then streaked onto SDA slants and grown overnight at 35 °C, prior to storage at –80 °C. For this study, all animal experiments were performed using isolates obtained from the –80 °C frozen stock. For each isolate, in vitro studies were conducted as the animal experiments proceeded, using isolates maintained on SDA slants at 4 °C. Five of the eighteen isolates were previously included in a study correlating fluconazole susceptibility in vitro to the response to fluconazole among mice with cryptococcal meningoencephalitis (Velez et al., 1993). These isolates are labelled V5, V15, H12, H19 and H11 in the present study, corresponding to isolates 6, 3, 17, 4 and 8, respectively, in the earlier study. In the present study, fresh cultures of the five isolates were obtained from the frozen stock and LD₅₀ values during murine meningoencephalitis were redetermined using the methods described below. As the data indicate, there were no significant differences in LD₅₀ between the earlier study and this.

Murine models of cryptococcosis.
Male ICR outbred mice were obtained from Harlan Sprague–Dawley. nu/– and nu/nu mice on a BALB/c background were obtained from a breeding colony at the University of Texas Health Sciences Center at San Antonio, Texas. Yeast cells grown overnight in brain heart infusion (BHI) broth were washed, counted and adjusted to the desired concentration in sterile saline. Viability of cells was confirmed by quantitative culture for each experiment.

Cryptococcal meningoencephalitis.
Prior to infection, the mice were anaesthetized with inhaled methoxyflurane. The cranium was swabbed with 70 % alcohol and the inoculum was injected in the midline of the cranium with a 26-gauge needle in a volume of 0.06 ml.

Assessment of LD₅₀.
For each isolate, eight ICR mice per group were initially inoculated with cryptococcal cells over a broad range of concentrations (10–400 c.f.u. per mouse). For those isolates for which an LD₅₀ could not be determined, inocula were increased to the range of 400–20 000 c.f.u. per mouse. The LD₅₀ was calculated by Probit analysis on day 15 of infection.

Assessment of tissue burdens.
Mice were sacrificed by cervical dislocation on day 9 of infection. The brains were removed, weighed and homogenized in 0.9 % saline containing amikacin and piperacillin (60 µg ml^–1 for both). The homogenates were serially diluted and plated onto SDA plates for burden quantification.

Assessment of capsule size within infected brains.
The capsule width was measured in cells obtained directly from homogenized brain tissue on day 9 using an eyepiece micrometer mounted on a phase-contrast microscope (resolution 0.12 µm). The width of the capsule was determined at a magnification of 1000x by measuring the distance between the outer edge of the yeast cell wall and the outer edge of the capsule. For each isolate, at least 10 cells in each of 20 fields (200 total cells) were examined in two mice.

Pulmonary cryptococcosis.
ICR mice received intraperitoneal injections of cortisone acetate (125 mg kg^–1 per mouse) 36 h prior to and 36 h after challenge with C. neoformans (Polak-Wyss, 1991). Prior to infection, mice (10 mice per group) were anaesthetized with inhaled methoxyflurane. The inocula were instilled slowly in equally divided aliquots (25 µl) to each nare using a micropipette and sterile tip, and mice were held upright until complete inhalation and resumption of normal breathing. The outcomes assessed were: (1) mortality over 21 days among mice infected with 1x10⁷ c.f.u. per mouse; (2) tissue burden in lungs, kidneys and brains of mice infected with a sublethal inoculum of 5x10⁶ c.f.u. per mouse; and (3) histopathology.

Measurements of phenotypes in vitro.
Growth rate was determined in yeast-extract-glucose (YEG) broth at 37 °C in a shaker incubator (150 r.p.m.) (Fries & Casadevall, 1998). An aliquot of the culture was obtained hourly for colony determination by plate count as well as by optical density. Capsule size in vitro was measured for yeast cells incubated overnight at 30 °C in induction medium (1.7 g yeast nutrient base without amino acids and without ammonium sulfate supplemented with 1.5 g asparagine and 20 g glucose per litre of buffer containing 12 mM NaHCO₃ and 35 mM MOPS, pH 7.1; Garcia-Hermoso et al., 2004). Yeast cells were then suspended in India ink and capsule sizes of 200 cells were measured as within murine brains. Melanin production was measured indirectly using a standardized determination of phenoloxidase activity (Polacheck et al., 1982). One unit of enzyme activity was defined as the amount which caused a change in absorbance at 480 and 300 nm of 0.001. Protein concentration was measured by reference to a standard curve. Extracellular proteinase activity was measured as previously described (Andrade et al., 2002; Leighton et al., 1973). Units of proteolytic activity were expressed as mg azoalbumin hydrolysed per hour. Specific activity was calculated as the ratio between total protease activity and total protein concentration in 1 ml sample (U mg^–1). The extracellular protein content was determined using bovine serum albumin (Sigma) as the standard. Extracellular phospholipase activity was determined using a modified egg yolk agar method (Chen et al., 1997). Each isolate was tested in duplicate on at least six separate occasions. Cultures were incubated at 30 °C, and the diameter of the zone of precipitation around the isolate was measured at 48, 72 and 96 h. The index of phospholipase activity was the ratio of the diameter of the colony to the total diameter of the colony plus precipitation zone (P_z). In vitro susceptibility to fluconazole was determined using a macrodilution method approved by the National Committee for Clinical Laboratory Standards (1997).

Adherence to A549 cells.
Cells of the A549 human lung epithelial line (ATCC CAL-185) were grown to confluence in 24-well plates (Costar) in Dulbecco's modified Eagle medium (Gibco) supplemented with 0.37 % NaHCO₃, 200 U penicillin ml^–1, 200 µg streptomycin ml^–1 and 10 % heat-inactivated fetal bovine serum (FBS) (A549 medium). A549 monolayers were washed with Earle's balanced salts (EBS) (Sigma) containing 0.22 % NaHCO₃ and 5 % FBS (EBS-FBS). Yeast cells were grown overnight in YPD medium at 37 °C, washed with phosphate-buffered saline (PBS) and resuspended in EBS-FBS. After co-incubation at 37 °C in a humidified CO₂ incubator for 4.5 h (the time determined to be optimal in preliminary experiments), A549 cells were washed with EBS-FBS, and 500 µl warm SDA added to the well. At the same time, A549 cells in duplicated wells were lysed with 0.5 % sodium deoxycholate. Then 500 µl warm SDA was added to the wells. Colonies were counted from lysed and non-lysed wells after incubation for 48–72 h at 37 °C. There were no significant differences in the numbers of yeast cells for a given inoculum in the lysed and non-lysed wells. Adherence was calculated as percentage of the original inoculum recovered.

Damage to A549 cells.
Damage to A549 cells was measured using a standard lactose dehydrogenase (LDH)-cytotoxicity assay kit (BioVision). Confluent A549 cells were washed three times with PBS and resuspended in A549 medium. The following samples were prepared individually in triplicate wells of a 96-well plate: background control, 200 µl medium alone; low control, 1x10⁴ A549 cells in 200 µl; high control, 1x10⁴ A549 cells containing 1 % Triton X-1000 in 200 µl; test sample, 1x10⁴ A549 cells and 5x10⁵ yeast in 200 µl. Cells were incubated at 37 °C in a humidified CO₂ incubator for 5 h (the time determined to be optimal in preliminary experiments). Following centrifugation, 100 µl supernatant per well was transferred into corresponding wells of an optically clear 96-well plate and 100 µl kit reaction mixture added. The plate was incubated at room temperature in the dark for 30 min and absorbance read at 500 nm. The percentage cytotoxicity was calculated as: [(Test sample–Low control)x(High control–Low control)^–1]x100.

Statistical analysis.
The LD₅₀ was exponentially transformed to approach a normal distribution prior to statistical analysis. Non-parametric correlation using the Spearman rank correlation test was used to correlate LD₅₀ and specific virulence factors. A multivariate analysis to assess independent factors of virulence was performed using multiple linear regression.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
C. neoformans isolates recovered from the CSF of persons with AIDS exhibit a wide range of virulence during murine cryptococcal meningoencephalitis
The 18 isolates were demonstrated to be of serotype A (C. neoformans var. grubii) by a slide agglutination test (Crypto-Check kit, Iatron) and mating type by PCR with specific primers that amplify portions of the STE20a or STE20 allele (Nielsen et al., 2003). We determined the LD₅₀ for each of the isolates at 15 days following the intracranial (i.c.) inoculation of ICR mice (Table 1). The median LD₅₀ was 996 c.f.u. (range 51 to >20 000). The LD₅₀ was <100 c.f.u. for 22 % (4/18) of isolates, 100–1000 c.f.u. for 28 % (5/18), and 1000–10 000 c.f.u. for 11 % (2/18). For 39 % (7/18) of isolates, mortality rates following inoculation of up to 20 000 c.f.u. were insufficient for calculation of an LD₅₀ (i.e. LD₅₀>20 000 c.f.u.). Indeed, four isolates (H2, H6, H11, H19) failed to kill any mice during the 30 days following inoculation of 20 000 c.f.u.. Three additional isolates resulted in the deaths of 43 %, 29 % and 29 % of mice over 30 days (isolates H12, H16 and H17, respectively). The seven isolates for which LD₅₀ exceeded 20 000 c.f.u. were classified as hypovirulent. The others were considered virulent. We denoted the hypovirulent isolates with an ‘H’ preceding the isolate number, and the virulent isolates with a ‘V’ preceding their number.

As a final assessment of each isolate's virulence in the CNS, we measured cryptococcal tissue burdens within the brains of ICR mice 9 days after i.c. inoculation (3xLD₅₀ for all virulent isolates except isolate V14, and 20 000 c.f.u. for isolate V14 and all hypovirulent isolates). There was a remarkable range of tissue burdens (250–1.01x10¹¹ c.f.u. g^–1) (Table 1); the median burden was 1.26x10⁸ c.f.u. g^–1. Overall, we demonstrated that there was an inverse correlation between tissue burden and LD₅₀ (i.e. higher tissue burden correlated with lower LD₅₀) (P=0.038; Spearman rank correlation).

The expression of certain virulence factors correlates with the virulence demonstrated during murine cryptococcal meningoencephalitis
Next, we characterized the expression of several well-described virulence factors by each of the isolates to determine if they correlated with the virulence observed in mice with cryptococcal meningoencephalitis (Table 1). Since reports have suggested that fluconazole resistance might influence the virulence of individual strains (Iwata et al., 1990), we also determined fluconazole MICs.

The growth rate at 37 °C in vitro was measured as doubling time in YEG liquid medium. The median doubling time for the 18 isolates was 4.61 h (range 2.13–8.36 h). The median capsule size after overnight growth in vitro was 2.48 µm (range 2.13–3.32 µm). The median capsule size was similar within homogenized brain tissues (3.29 µm), but the range of sizes was greater (range 1.09–6.84 µm). Extracellular phospholipase activity, measured as a zone of precipitation using an egg yolk agar assay (Chen et al., 1997), was detected for all isolates; the median P_z value (i.e. ratio of the diameter of the colony to the total diameter of the colony plus the precipitation zone) was 0.66 (range 0.59–0.73). Melanin production, assessed by measuring phenoloxidase activity, was also detected for each of the isolates. The median activity was 4930 U mg^–1 (range 796–29 500 U mg^–1). We did not detect extracellular proteinase activity for four isolates (V18, V8, V15 and V17). Among the remaining isolates, the median activity was 17.2 U (mg protein)^–1 [range 4.9–206.7 U (mg protein)^–1]. The median fluconazole MIC for the isolates was 4 µg ml^–1 (range 0.5–16 µg ml^–1) and the MIC₉₀ was 8 µg ml^–1. No isolates were resistant in vitro to fluconazole (i.e. MIC64 µg ml^–1).

There was a positive association between LD₅₀ and doubling time (i.e. lower LD₅₀ was associated with shorter doubling time) (P=0.0001; non-parametric Spearman rank correlation test), and an inverse association between LD₅₀ and capsule size in vivo (P=0.028; non-parametric Spearman rank correlation test). There was no correlation between LD₅₀ and phospholipase, phenoloxidase and proteinase activities, capsule size measured in vitro, and fluconazole MIC.

Multivariate analysis including the factors found to be associated with LD₅₀ by univariate analyses identified both the doubling time and capsule size in vivo as significantly associated with LD₅₀ (P=0.017 and 0.016, respectively; multiple regression analysis).

The relative virulence of two isolates within murine lungs following inhalation is similar to that observed during meningoencephalitis
We compared the abilities of a virulent and a hypovirulent isolate (V7 and H16, respectively) to cause disease following intranasal inoculations (1x10⁷ c.f.u.) of mice immunosuppressed with cortisone acetate. Mice infected with isolate V7 suffered significantly greater mortality during pulmonary cryptococcosis than mice infected with strain H16 (P<0.001; log rank test) (Fig. 1).

View larger version (10K): [in this window] [in a new window]   Fig. 1. Kaplan–Meier survival curve of mice infected intranasally with C. neoformans isolates V7 () and H16 ().

The more virulent isolate exhibits greater adherence to human pneumocytes in vitro and causes greater cellular damage
Next, we studied the interactions between the two isolates and A549 human lung carcinoma epithelial cells (ATCC CAL-185) in vitro. We found that isolate V7 exhibited significantly greater adherence to A549 cells than isolate H16 (12.7±6.5 % versus 4.1±2.7 %; P=0.0003). Finally, the more virulent isolate also caused greater damage to the A549 cells, as measured by an LDH release assay of cytotoxicity at 5 h (20±6.2 % cytotoxicity versus 6.6±2.9 %; P=0.02).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
To our knowledge, this is the largest study to correlate the CNS virulence of C. neoformans var. grubii isolates from different patients with the expression of several well-characterized virulence factors. The most notable finding of the study was the wide range of virulence exhibited by the isolates, as measured by both LD₅₀ (range 51 to >20 000 c.f.u.) and the recovery of viable organisms from infected brains (range 250–1.01x10¹¹ c.f.u. g^–1). In particular, we found that 39 % (7/18) of the isolates were markedly hypovirulent, causing significantly lower tissue burdens and unable to kill enough immunocompetent mice to permit the calculation of an LD₅₀. Moreover, one of three hypovirulent isolates tested in T-cell-deficient nude (nu/nu) mice did not cause any mortality over 60 days (isolate H2), and the other two isolates were associated with mortality rates of only 30 and 50 % (isolates H6 and H11, respectively). At the same time, the overall survival times among nude mice infected with isolate H11 were significantly shorter than among immunocompetent littermates, as would be predicted for an opportunistic pathogen that causes disease principally among patients with impaired cell-mediated immunity. Taken together, our data highlight that an organism's disease-causing potential in humans is shaped, in large part, by immune status and other host factors such as genetic characteristics. Clearly, significant numbers of AIDS patients and others with profound suppression of cell-mediated immunity are unable to defend themselves from CNS infection by organisms of even minimal virulence.

The ranges in the expression of selected virulence factors by C. neoformans isolates were also remarkable. Doubling time, for example, ranged from 2.13 to 8.36 h (isolates V18 and H17, respectively), and phenoloxidase activity from 796 to 29 500 U (mg protein)^–1 (isolates V3 and H12, respectively). The inclusion of a large number of isolates that demonstrated wide ranges of LD₅₀ and virulence factor expression enabled us to evaluate the relationships between specific factors and virulence. There was no association between virulence and capsule size in vitro, extracellular phospholipase activity, the amount of melanin formation (measured as phenoloxidase activity), proteinase activity and fluconazole MIC. However, we found that growth rate at 37 °C (measured as doubling time) and capsule size in vivo correlated with LD₅₀ by univariate and multivariate analyses. Despite the significance of these factors, neither was indispensable or dominant. For example, one of the most virulent isolates (V3; LD₅₀ 67 c.f.u.) exhibited the smallest capsule within the brain. As such, our findings support a composite virulence phenotype (Cutler, 1991; Casadevall & Perfect, 1998), in which a variety of factors contribute to the overall pathogenicity of a C. neoformans isolate and a limited contribution by one factor can be compensated by the expression of others.

The finding that growth rate at 37 °C correlated with LD₅₀ is consistent with previous studies. The ability to grow at human body temperatures of 37 °C and above is essential for pathogenic fungi to cause disease. In a previous study of seven C. neoformans strains of various serotypes, the three fastest growing strains at 37 °C were the most virulent following the intravenous inoculation of mice (Kagaya et al., 1985). The finding of an association between capsule size within infected brains and virulence, on the other hand, is more notable. Generally taken to be the most important cryptococcal virulence factor (Janbon, 2004), the capsule inhibits (but does not prevent) phagocytosis in vivo (Feldmesser et al., 2001), facilitates survival and replication within macrophages (Feldmesser et al., 2001), inhibits migration of leukocytes (Dong & Murphy, 1995), alters antigen presentation (Collins & Bancroft, 1991), dysregulates cytokine secretion (Retini et al., 1996; Vecchiarelli et al., 1995, 1996), and can deplete complement (Macher et al., 1978). Disruption of multiple genes that are individually necessary for capsule synthesis results in acapsular strains that are avirulent in murine cryptococcosis (Chang & Kwon-Chung, 1994, 1998, 1999; Chang et al., 1996). Although it has long been recognized that the capsule thickness of C. neoformans increases during infection (Littman & Tsubura, 1959), a correlation between capsule size in vivo and relative virulence has not to our knowledge been reported previously.

Our findings are particularly noteworthy since capsule size in vitro did not correlate with virulence, an observation consistent with previous studies (Littman & Tsubura, 1959; Dykstra et al., 1977). Taken together, the data have important implications. Clearly, the regulation of capsule formation varies depending on the local environment, as previously suggested by studies showing organ-dependent variation of capsule size (Rivera et al., 1998). More generally, results obtained under in vitro conditions cannot necessarily be extrapolated to settings in vivo. As such, we must acknowledge that factors like melanin formation and proteinase activity that did not correlate with virulence in this study might nevertheless be important within infected brains.

It is not surprising that we did not find correlations between phospholipase, phenoloxidase and proteinase activities in vitro and LD₅₀. Although phospholipase and phenoloxidase activity are clearly linked to virulence, and melanin can be detected within human brains and other tissues infected with C. neoformans (Casadevall et al., 2000; Chen et al., 1997; Cox et al., 2001; Jacobson & Emery, 1991; Kwon-Chung et al., 1982; Kwon-Chung & Rhodes, 1986; Nosanchuk & Casadevall, 2003; Noverr et al., 2004; Rhodes et al., 1982; Salas et al., 1996; Williamson, 1997), the levels of phospholipase and phenoloxidase activity have not been shown to be associated with virulence. Like the overwhelming majority of clinical isolates, all our isolates exhibited phospholipase and phenoloxidase activity (Kwon-Chung et al., 1982; Chen et al., 1997). The data, therefore, suggest that some minimum levels of phospholipase and melanin formation are likely to play roles in pathogenesis, but there are no gains in virulence beyond these levels. Along similar lines, previous studies have failed to find a correlation between the amount of proteinase activity and virulence (Chen et al., 1996; Franzot et al., 1998; Fries & Casadevall, 1998). In fact, we found that even strains lacking detectable proteinase activity were associated with low LD₅₀ (e.g. isolate V18; LD₅₀ 51 c.f.u.), suggesting that this property might be dispensable for the pathogenesis of disease (Chen et al., 1996; Franzot et al., 1998; Fries & Casadevall, 1998). Since we did not detect any resistance to fluconazole, it is not possible to assess a previous report that azole-resistant C. neoformans isolates demonstrate attenuated virulence (Iwata et al., 1990). Nevertheless, we did not find any correlation between LD₅₀ and fluconazole MIC.

We demonstrated that the level of virulence for two isolates in the CNS also corresponded with virulence in the lungs following intranasal inoculation. Isolate V7, which was virulent during meningoencephalitis, caused greater mortality, higher tissue burdens within homogenized lungs, and a more profound inflammatory response than the hypovirulent isolate H16. These findings are consistent with our observations that isolate V7 demonstrated greater adherence to pulmonary epithelilal cells in vitro and caused greater damage to these cells. Clearly, the relative virulence of at least some C. neoformans isolates is conserved within diverse organs (Barchiesi et al., 2005). We also noted that both isolates were able to disseminate from the lungs to brains and kidneys. Consistent with the findings in the lungs, the tissue burdens of isolate V7 were greater in the brains and kidneys after 21 days than those of isolate H16. We cannot conclude from these experiments, however, whether this stems from differences among the isolates in tropism for these organs or merely reflects increased numbers of isolate V7 gaining access to the bloodstream from the lungs.

In conclusion, we have characterized the CNS virulence of 18 C. neoformans var. grubii isolates recovered from the CSF of AIDS patients, and correlated this with the expression of several virulence factors in vitro. In addition to the insights into the pathogenesis of cryptococcosis provided by our data, this study is important because our collection of isolates now represents a valuable resource for future research. For example, we are planning to characterize a number of the isolates for virulence within murine lungs, as well as more thoroughly investigate interactions with pneumocytes. Other future investigations will include comparisons of transcriptional profiles under conditions relevant to pathogenesis, and studies of the interactions between isolates and host cells important to the pathogenic process such as alveolar macrophages.

	ACKNOWLEDGEMENTS

 
Drs Clancy and Nguyen's research is conducted in laboratories at the North Florida/South Georgia VA Medical Center, Gainesville, FL, USA. Their participation in the project was supported by the Medical Research Service of the Department of Veterans Affairs. Dr Clancy was supported by a research grant from the American Lung Association of Florida. The project was conducted as part of the University of Florida Mycology Research Unit.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Andrade, V. S., Sarubbo, L. A., Fukushima, K., Miyaji, M., Nishimura, K. & de Campos-Takaki, G. M. (2002). Production of extracellular proteases by Mucor circinelloides using D-glucose as carbon source/substrate. Braz J Microbiol 33, 106–110.

Barchiesi, F., Cogliati, M., Esposto, M. C., Spreghini, E., Schimizzi, A. M., Wickes, B. L., Scalise, G. & Viviani, M. A. (2005). Comparative analysis of pathogenicity of Cryptococcus neoformans serotypes A, D and AD in murine cryptococcosis. J Infect 51, 10–16.[CrossRef][Medline]

Blackstock, R. & Murphy, J. W. (1997). Secretion of the C3 component of complement by peritoneal cells cultured with encapsulated Cryptococcus neoformans. Infect Immun 65, 4114–4121.[Abstract]

Blackstock, R., Buchanan, K. L., Cherniak, R., Mitchell, T. G., Wong, B., Bartiss, A., Jackson, L. & Murphy, J. W. (1999). Pathogenesis of Cryptococcus neoformans is associated with quantitative differences in multiple virulence factors. Mycopathologia 147, 1–11.[CrossRef][Medline]

Blasi, E., Barluzzi, R., Mazzolla, R., Mosci, P. & Bistoni, F. (1992). Experimental model of intracerebral infection with Cryptococcus neoformans: roles of phagocytes and opsonization. Infect Immun 60, 3682–3688.[Abstract/Free Full Text]

Blasi, E., Barluzzi, R., Mazzolla, R., Tancini, B., Saleppico, S., Puliti, L., Pitzurra, M. & Bistoni, F. (1995). Role of nitric oxide and melanogenesis in the accomplishment of anticryptococcal activity by the BV-2 microglial cell line. J Neuroimmunol 58, 111–116.[CrossRef][Medline]

Casadevall, A. & Perfect, J. R. (1998). Cryptococcus neoformans. Washington, DC: American Society for Microbiology.

Casadevall, A., Rosas, A. L. & Nosanchuk, J. D. (2000). Melanin and virulence in Cryptococcus neoformans. Curr Opin Microbiol 3, 354–358.[CrossRef][Medline]

Chang, Y. C. & Kwon-Chung, J. K. (1994). Complementation of a capsule-deficiency mutation of Cryptococcus neoformans restores its virulence. Mol Cell Biol 14, 4912–4919.[Abstract/Free Full Text]

Chang, Y. C. & Kwon-Chung, J. K. (1998). Isolation of the third capsule-associated gene, CAP60, required for virulence in Cryptococcus neoformans. Infect Immun 66, 2230–2236.[Abstract/Free Full Text]

Chang, Y. C. & Kwon-Chung, J. K. (1999). Isolation, characterization, and localization of a capsule-associated gene, CAP10, of Cryptococcus neoformans. J Bacteriol 181, 5636–43.[Abstract/Free Full Text]

Chang, Y. C., Penoyer, L. A. & Kwon-Chung, K. J. (1996). The second capsule gene of Cryptococcus neoformans, Cap64, is essential for virulence. Infect Immun 64, 1977–1983.[Abstract]

Chen, L. C., Blank, E. S. & Casadevall, A. (1996). Extracellular proteinase activity of Cryptococcus neoformans. Clin Diagn Lab Immunol 3, 570–574.[Medline]

Chen, S. C. A., Muller, M., Zhou, J. Z., Wright, L. C. & Sorrell, T. C. (1997). Phospholipase activity in Cryptococcus neoformans: a new virulence factor? J Infect Dis 175, 414–420.[Medline]

Collins, H. L. & Bancroft, G. J. (1991). Encapsulation of Cryptococcus neoformans impairs antigen-specific T-cell responses. Infect Immun 59, 3883–3888.[Abstract/Free Full Text]

Cox, G. M., McDade, H. C., Chen, S. C. A. & 8 other authors (2001). Extracellular phospholipase activity is a virulence factor for Cryptococcus neoformans. Mol Microbiol 39, 166–175.[CrossRef][Medline]

Curtis, J. L., Huffnagle, G. B., Chen, G. H., Warnock, M. L., Gyetko, M. R., McDonald, R. A., Scott, P. J. & Toews, G. B. (1994). Experimental murine pulmonary cryptococcosis: differences in pulmonary inflammation and lymphocyte recruitment induced by two encapsulated strains of Cryptococcus neoformans. Lab Invest 71, 113–126.[Medline]

Cutler, J. E. (1991). Putative virulence factors of Candida albicans. Annu Rev Microbiol 45, 187–218.[CrossRef][Medline]

Dixon, D. M. & Polak, A. (1986). In vivo and in vitro studies with an atypical, rhinotrophic isolate of Cryptococcus neoformans. Mycopathologia 96, 33–96, 40.[CrossRef][Medline]

Dong, Z. M. & Murphy, J. W. (1995). Effects of the two varieties of Cryptococcus neoformans cells and culture filtrate antigens on neutrophil locomotion. Infect Immun 63, 2632–2644.[Abstract]

Dykstra, M. A., Friedman, L. & Seed, B. (1977). Capsule size of Cryptococcus neoformans: control and relationship to virulence. Infect Immun 16, 129–135.[Abstract/Free Full Text]

Feldmesser, M., Kress, Y. & Casadevall, A. (2001). Dynamic changes in the morphology of Cryptococcus neoformans during murine pulmonary infection. Microbiology 147, 2355–65.[Abstract/Free Full Text]

Franzot, S. P., Mukherjee, J., Cherniak, R., Chen, L., Hamdan, J. S. & Casadevall, A. (1998). Microevolution of a standard strain of Cryptococcus neoformans resulting in differences in virulence and other phenotypes. Infect Immun 66, 89–97.[Abstract/Free Full Text]

Fries, B. C. & Casadevall, A. (1998). Serial isolates of Cryptococcus neoformans from patients with AIDS differ in virulence for mice. J Infect Dis 178, 1761–1766.[CrossRef][Medline]

Fromtling, R. A., Abruzzo, G. K. & Ruiz, A. (1988). Cryptococcus neoformans: a central nervous system isolate from an AIDS patient that is rhinotropic in a normal mouse model. Mycopathologia 102, 72–86.

Garcia-Hermoso, D., Dromer, F. & Janbon, G. (2004). Cryptococcus neoformans capsule structure evolution in vitro and during murine infection. Infect Immun 72, 3359–3365.[Abstract/Free Full Text]

Grosse, G., Mishra, S. K. & Staib, F. (1975). Selective involvement of the brain in experimental murine cryptococcosis. II. Histopathological observations. Zentralbl Bakteriol 230, 518–533.

Hasenclever, H. F. & Mitchell, W. O. (1960). Virulence and growth rates of Cryptococcus neoformans in mice. Ann N Y Acad Sci 89, 156–62.[Medline]

Iwata, K., Yamashita, T., Ohsumi, M., Baba, M., Naito, N., Taki, A. & Yamada, N. (1990). Comparative morphological and biological studies on the itraconazole- and ketoconazole-resistant mutants of Cryptococcus neoformans. J Med Vet Mycol 28, 77–90.[Medline]

Jacobson, E. S. & Emery, H. S. (1991). Catecholamine uptake, melanization, and oxygen toxicity in Cryptococcus neoformans. J Bacteriol 173, 401–403.[Abstract/Free Full Text]

Janbon, G. (2004). Cryptococcus neoformans capsule biosynthesis and regulation. FEMS Yeast Res 4, 765–771.[CrossRef][Medline]

Kagaya, K., Yamada, T., Miyakawa, Y., Fukazawa, Y. & Saito, S. (1985). Characterization of pathogenic constituents of Cryptococcus neoformans strains. Microbiol Immunol 29, 517–532.[Medline]

Kozel, T. R. (1995). Virulence factors of Cryptococcus neoformans. Trends Microbiol 3, 295–299.[CrossRef][Medline]

Kwon-Chung, K. J. & Rhodes, J. C. (1986). Encapsulation and melanin formation as indicators of virulence in Cryptococcus neoformans. Infect Immun 51, 218–223.[Abstract/Free Full Text]

Kwon-Chung, K. J., Polacheck, I. & Popkin, T. J. (1982). Melanin-lacking mutants of Cryptococcus neoformans and their virulence for mice. J Bacteriol 150, 1414–1421.[Abstract/Free Full Text]

Kwon-Chung, K. J., Wickes, B. L., Stockman, L., Roberts, G. D., Ellis, D. & Howard, D. H. (1992). Virulence, serotype, and molecular characteristics of environmental strains of Cryptococcus neoformans var. gattii. Infect Immun 60, 1869–1874.[Abstract/Free Full Text]

Leighton, T. J., Doi, R. H., Warren, R. A. J. & Kelln, R. A. (1973). The relationship of serine protease activity to RNA polymerase modification and sporulation in Bacillus subtilis. J Mol Biol 76, 103–122.[CrossRef][Medline]

Littman, M. L. & Tsubura, E. (1959). Effect of degree of encapsulation upon virulence of Cryptococcus neoformans. Proc Soc Exp Biol Med 101, 773–776.[Medline]

Macher, A. M., Bennett, E., Gadek, J. E. & Frank, M. M. (1978). Complement depletion in cryptococcal sepsis. J Immunol 120, 1686–1690.[Abstract/Free Full Text]

Mitchell, T. G. & Perfect, J. R. (1995). Cryptococcosis in the era of AIDS – 100 years after the discovery of Cryptococcus neoformans. Clin Microbiol Rev 8, 515–548.[Abstract]

National Committee for Clinical Laboratory Standards (1997). Reference Method for Broth Dilution Antifungal Susceptibility Testing of Yeasts. Approved standard. Document M-27A. Wayne, PA: National Committee for Clinical Laboratory Standards.

Nielsen, K., Cox, G., Wang, P., Toffaletti, D., Perfect, J. R. & Heitman, J. (2003). Sexual cycle of Cryptococcus neoformans var. grubii and virulence of congenic a and isolates. Infect Immun 71, 4831–4841.[Abstract/Free Full Text]

Nosanchuk, J. D. & Casadevall, A. (2003). Budding of melanized Cryptococcus neoformans in the presence or absence of L-dopa. Microbiology 149, 1945–1951.[Abstract/Free Full Text]

Noverr, M. C., Williamson, P. R., Fajardo, R. S. & Huffnagle, G. B. (2004). CNLAC1 is required for extrapulmonary dissemination of Cryptococcus neoformans but not pulmonary persistence. Infect Immun 72, 1693–1699.[Abstract/Free Full Text]

Polacheck, I., Hearing, V. J. & Kwon-Chung, K. J. (1982). Biochemical studies of phenoloxidase and utilization of catechalamines in Cryptococcus neoformans. J Bacteriol 150, 1212–1220.[Abstract/Free Full Text]

Polak-Wyss, A. (1991). Protective effect of human granulocyte colony-stimulating factor (hG-CSF) on Cryptococcus and Aspergillus infections in normal and immunosuppressed mice. Mycoses 34, 205–215.[Medline]

Retini, C., Vecchiatelli, A., Monari, C., Tascini, C., Bistoni, F. & Kozel, T. R. (1996). Capsular polysaccharide of Cryptococcus neoformans induces proinflammatory cytokine release by human neutrophils. Infect Immun 64, 2897–2903.[Abstract]

Rhodes, J. C., Polacheck, I. & Kwon-Chung, K. J. (1982). Phenoloxidase activity and virulence in isogenic strains of Cryptococcus neoformans. Infect Immun 36, 1175–1184.[Abstract/Free Full Text]

Rivera, J., Feldmesser, M., Cammer, M. & Casadevall, A. (1998). Organ-dependent variation of capsule thickness in Cryptococcus neoformans during experimental murine infection. Infect Immun 66, 5027–5030.[Abstract/Free Full Text]

Salas, S. D., Bennett, J. E., Kwon-Chung, K. L., Perfect, J. R. & Williamson, P. R. (1996). Effect of the laccase gene CNLAC1, on virulence of Cryptococcus neoformans. J Exp Med 184, 377–86.[Abstract/Free Full Text]

Vanbreuseghem, R. (1967). White mice sensitivity to Cryptococcus neoformans. Production of resistance against a cerebral inoculation. Ann Soc Belg Med Trop 47, 281–294.[Medline]

Van Cutsem, J., Fransen, J., Van Gerven, F. & Janssen, P. A. J. (1986). Experimental cryptococcosis: dissemination of Cryptococcus neoformans and dermatropism in guinea pigs. Mykosen 29, 561–575.[Medline]

Vecchiarelli, A., Retini, C., Pietrella, D., Monari, C., Tascini, C., Beccari, T. & Kozel, T. T. (1995). Down-regulation by cryptococcal polysaccharide of tumor necrosis factor alpha and interleukin-1 beta secretion from human monocytes. Infect Immun 63, 2919–2923.[Abstract]

Vecchiarelli, A., Retini, C., Monari, C., Tascini, C., Bistoni, F. & Kozel, T. R. (1996). Purified capsular polysaccharide of Cryptococcus neoformans induces interleukin-10 secretion by human monocytes. Infect Immun 64, 2846–2849.[Abstract]

Velez, J. D., Allendoerfer, R. R., Luther, M., Rinaldi, M. G. & Graybill, J. R. (1993). Correlation of in vitro azole susceptibility with in vivo response in a murine model of cryptococcal meningitis. J Infect Dis 168, 508–510.[Medline]

Williamson, P. R. (1997). Laccase and melanin in the pathogenesis of Cryptococcus neoformans. Front Biosci 2, e99–e107.[Medline]

Received 22 December 2005; revised 22 March 2006; accepted 29 March 2006.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Clancy, C. J. Articles by Graybill, J. R. Search for Related Content PubMed PubMed Citation Articles by Clancy, C. J. Articles by Graybill, J. R. Agricola Articles by Clancy, C. J. Articles by Graybill, J. R.