Cryptococcus neoformans is an encapsulated basdiomycete that is readily isolated from the environment, especially in areas contaminated with pigeon droppings. Although symptomatic cryptococcosis is widely recognized in immunocompromised individuals, unrecognized and subclinical infections occur in immunocompetent individuals, in a manner similar to other endemic mycoses.1 These infections may persist and later reactivate in the setting of acquired immunodeficiency.
Clinical symptoms cannot be used as a measure of the incidence of cryptococcal infection in a population. Consequently serologic studies are the mainstay of epidemiologic studies. Serologic studies support the idea of unrecognized or subclinical cryptococcosis among otherwise healthy individuals. In previous studies, we used an immunoblot assay to demonstrate a high prevalence of antibodies to whole cell extract proteins among children from the Bronx.2
In the current study, we analyzed the reactivity of sera of children from 3 distinct regions against cytoplasmic proteins of C. neoformans. These regions include: the Bronx, Dutchess County (a suburb approximately 60 miles north of New York City) and Manila, Philippines. Our findings suggest that the prevalence of subclinical and unrecognized C. neoformans infections are disproportionately high in the Bronx when compared with the other regions.
Sera were obtained from 3 sources: Bronx, Dutchess County (NY) and Manila, Philippines. The Bronx specimens consisted of a convenience sample of 57 specimens derived from our previous sera collection from a Bronx Emergency Department.2 For Dutchess County samples, leftover blood without identifying information was collected from 150 consecutive children evaluated in a private practice pediatric office. Dutchess County is a suburb approximately 60 miles north of New York City with a population of approximately 290,000. The county covers approximately 800 square miles. Thirty-nine samples from Philippine children were analyzed. Most of the Filipino children resided in Manila or in surrounding cities. These samples were obtained as part of an experimental protocol from children receiving directly observed therapy for a diagnosis of tuberculosis. This study was approved by the Committee on Clinical Investigations at the Albert Einstein College of Medicine and the University of the Philippines-Philippine General Hospital, Manila.
Strains and Growth Conditions.
C. neoformans strain 24067 (serotype D) and C. albicans (BSMY212) were obtained from American Type Culture Collection. Fungi were grown in Sabouraud's dextrose broth for 2 days at 30°C before protein isolation.
Fungal Protein Extracts.
Cytosolic protein extracts of C. neoformans and whole cell extract of C. albicans were prepared as described.3 Protein concentrations were determined by using a bicinchoninic acid based protein assay (Pierce).
Electrophoresis was done in the Bio-Rad Mini-Protean II system with a 10% resolving gel at 100 V. Gels were transferred to nitrocellulose membranes and incubated with diluted human (1:1600) sera. IgG was detected with horseradish peroxidase-conjugated goat antibody to human IgG. Blots were developed with luminol (Pierce).
ELISA plates (96 wells) were coated with C. neoformans polysaccharide (CNPS). Sera were applied and reactive antibodies were detected with alkaline phosphatase labeled goat-antihuman immunoglobulin G. Color was developed with p-nitrophenyl phosphate. The titer was defined as the dilution of sera resulting in an A405 2 times greater than the background.
Comparisons for reactivity against C. neoformans proteins and CNPS were done using a Kruskal–Wallis test and subsequent analysis with the Dunn's multiple comparison test. CNPS antibody titers were log transformed before analysis. Correlations between age and reactivity as well as reactivity against proteins and polysaccharide were tested by calculating a Spearman rank correlation coefficient. All statistics were done using In Stat software (San Diego, CA).
We collected 150 samples from Dutchess County and 122 were analyzed. Specimens were rejected if they were improperly labeled, or had inadequate specimen or background information. Of the 122 samples analyzed, 45% were from females and 55% were from males. The majority (93%) of children were seen during well child visits when the blood was drawn. The mean age was 9.6 ± 4.2 years and the median was 10 years (range 0–18 years). Approximately 22% children were younger than 5 years, 39% were between 5 and 10 years, and 39% were older than10 years. For the Bronx samples, 57 sera were analyzed, of which 53% were from females and 47% were males. The mean and median ages were 10.7 ± 6.8 and 8.5 years (range 0–21 years), respectively. Approximately 19% children were younger than 5 years, 33% were between 5 and 10 years and 47% were older than 10 years. For the Philippines, 39 samples were analyzed. Among Philippine samples, approximately 39% were from females and 61% were from males. The average and median ages were 12.8 ± 4.5 and 14 years (range 0–18 years), respectively. None of the samples were from immunocompromised children.
Antibodies to Cryptococcal Proteins.
Among the 3 cohorts, the greatest reactivity to cryptococcal proteins was present in the Bronx sera. Approximately 63% of Bronx sera reacted with 6 or more bands, whereas for Dutchess County and Philippines sera, only 22% and 18%, respectively reacted with 6 or more bands (P < 0.001). The median number of bands recognized by Bronx, Dutchess County and Philippines sera were: 7, 2 and 1, respectively (P < 0.001 for comparisons between Bronx and Dutchess County and Philippines) (Fig. 1). A direct correlation between age and reactivity to cryptococcal proteins was apparent for the Bronx cohort (r = 0.55, P < 0.0001). Similar age-related correlations were not present for either the Dutchess County or Philippines cohorts.
Antibodies to Whole Cell Extracts of C. albicans.
To determine whether sera from Dutchess County children were selectively nonreactive to cytoplasmic C. neoformans proteins, as opposed to generally nonreactive, these sera were also blotted against whole cell extracts of C. albicans. Reactivity against C. albicans proteins was extensive among this group. The median number of proteins recognized was 11 (range 4–18).
Antibodies to Cryptococcal Polysaccharide.
Median antibody titers (log10) to CNPS for children from the Bronx, Dutchess County and Philippine children were: 2.30, 2.90 and 2.60, respectively (P < 0.001 Bronx vs. Dutchess). Within individual groups, no difference in CNPS titers were present for children <5 years when compared with children 5–21 years old.
Correlation Between Polysaccharide Titers and Immunoblot Reactivity.
We tested for the correlations between reactivity against polysaccharide and cryptococcal proteins among the various sera. Overall, there was no correlation between antibody titers to CNPS and number of bands recognized in immunoblot studies (r = 0.09, P = 0.16).
Our immunoblot studies demonstrate a high frequency of seroreactivity to C. neoformans among children from the Bronx, but not for children from Dutchess County, New York or Manila, Philippines. These findings suggest that unrecognized or subclinical cryptococcosis is common among Bronx children. Unrecognized cryptococcosis may manifest with respiratory signs and symptoms that are confused with a viral illness. The lack of reactivity observed in children from other areas highlights the specificity of immunoblot studies and suggests that the prevalence of subclinical cryptococcosis in children is location-dependent.
The high prevalence of subclinical cryptococcosis in Bronx children may be related to the high density of pigeons in this population. Pigeons are well suited for living in metropolitan areas. It is estimated that pigeon populations are roughly 2% the size of the human population in large cities.4 The acidic nature and high nitrogen content of pigeon excreta provide excellent growth conditions for C. neoformans. C. neoformans can attain densities of greater than 10 million organisms per gram of excreta and survive for prolonged intervals (almost 2 years in one study).1 Children from Manila, a very densely populated urban environment, demonstrated a low prevalence of antibodies to C. neoformans proteins. It is likely that other environmental factors (ie, temperature and humidity) are important in the acquisition of cryptococcal infection and account for the discrepancy in seroreactivity between these populations.
In contrast to our immunoblot findings, we observed significant reactivity against cryptococcal polysaccharide among children from Dutchess County. One explanation for this discrepancy is that this high level of reactivity is related to the presence of cross-reactive antibodies derived to exposures to other polysaccharides (ie, Pneumococcus 5 and Trichosporon).6 These findings suggest that antibody titers to cryptococcal polysaccharide are less specific indicators of cryptococcal infection when compared with immunoblot studies.
We interpret our studies to indicate a high prevalence of unrecognized or asymptomatic cryptococcosis in immunocompetent Bronx children, but not children from Dutchess County or the Philippines. Animal studies indicate that subclinical cryptococcal infection enhances pulmonary allergic responses and airway reactivity disease.7 Additional studies are needed to determine the health impact of childhood cryptococcosis in the Bronx, including its contribution to the high prevalence of asthma in this area.
The authors thank Drs. Hnin Khine and Tsoline Kojaoghlanian for their critical reading of this manuscript.
AC is supported by the following grants: AI 3774-11, HL59842-07, AI33142-02, GM07142-01.
1. Littman ML, Schneierson SS. Cryptococcus neoformans
in pigeon excreta in New York City. Am J Hyg
2. Goldman DL, Khine H, Abadi J, et al. Serologic evidence for Cryptococcus neoformans
infection in early childhood. Pediatrics
3. Chen L-C, Goldman DL, Doering TL, Pirofski L-A, Casadevall A. The antibody response to Cryptococcus neoformans
proteins in rodents and humans. Infect Immun
4. Merkus PJ, van Toorenenbergen AW. Birds or zebras? Allergy
5. Rebers PA, Barker SA, Heidelberger M, Dische Z, Evans EE. Precipitation of the specific polysaccharide of Cryptcococcus neoformans
A by types II and XIV antipneumococcal antisera. J Am Chem Soc
6. McManus EJ, Jones JM. Detection of a Trichosporon beigelii
antigen cross-reactive with Cryptococcus neoformans
capsular polysaccharide in serum from a patient with disseminated Trichosporon
infection. J Clin Microbiol
7. Goldman DL, Davis J, Bommarito F, Shao X, Casadevall A. Enhanced allergic inflammation and airway responsiveness in rats with chronic Cryptococcus neoformans
infection: potential role for fungal pulmonary infection in the pathogenesis of asthma. J Infect Dis