Vol. 105, No. 6


Copyright © 1977 by The Johns Hopkins University School of Hygiene and Public Health

Printed in US A.

EPIDEMIOLOGIC DIFFERENCES AMONG SEROTYPES OF CRYPTOCOCCUS NEOFORMANS JOHN E. BENNETT,1 K. J. KWON-CHUNG,' AND DEXTER H. HOWARD2 Bennett, J.E. (Clinical Mycology Section, Lab. of Clinical Investigation, NIAID, NIH, Bethesda, MD 20014), K.J. Kwon-Chung and D.H. Howard. Epidemiologic differences among serotypes of Cryptococcus neoformans. Am J Epidemiol 105:582-586, 1977. In the USA, the most prevalent serotype of the fungus, Cryptococcus neoformans, was serotype A. This serotype constituted 203 of 272 isolates from infections and 85 of 89 isolates from the environment. Serotype B or C isolates were infrequent causes of infection, except in Southern California, and were not isolated at all from environmental sources. In Southern California, the absence of serotypes B and C in 67 soil and pigeon dropping isolates was striking, considering that 25 of 49 isolates from infections were serotypes B or C. The site in nature where serotypes B and C exist is currently unknown but differs from that of serotypes A and D. Serotype D may be unusually prevalent in both environmental and patient isolates from Denmark and Italy. Of 24 isolates from those countries, 21 were serotype D. Cryptococcus neoformans; serotyping; pigeons

Cryptococcus neoformans is an encapsulated yeast-like fungus which is pathogenic for animals and man. In the majority of infections, the source from which the infection was acquired cannot be discerned. Man-to-man or animal-to-man transmission has not been documented. Occasionally, there has been a known or Received for publication November 9, 1976, and in final form January 31, 1977. 1 Clinical Mycology Section, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, MD 20014. 2 Department of Microbiology and Immunology, School of Medicine, University of California, Los Angeles, CA 90024. The authors would like to thank all those who contributed isolates of C. neoformans for this study. Assistance from the following investigators was particularly valuable: Dr. Prasnad Thasnakorn, Thailand; Dr. A. Stenderup, Aarhus, Denmark; Dr. G. Gargani, Florence, Italy; Dr. L. Ajello, Atlanta, GA; Dr. H. Walch, San Diego, CA; and Dr. C. T. Dolan, formerly of Rochester, MN. Members of the Southwest Pigeon Fanciers Association cooperated in the project by supplying many of the specimens from Southern California. Their efforts and interest are appreciated. The technical assistance of Johnsie Bailey and Charlotte Davis is also gratefully acknowledged.

presumed exposure to weathered pigeon droppings, an environmental source which often contains the fungus. Soil or air likely to be contaminated with pigeon or other bird droppings have also yielded C. neoformans, but other environmental sources contain this fungus very infrequently (1). It is conceivable that avian droppings are the occult source of most or all cryptococcosis because birds, including pigeons, are so ubiquitous. If avian droppings were the major source, as is often assumed, then serotypes of C. neoformans from environmental sources would likely parallel the serotypes found in infections occurring in the same locality. Capsular polysaccharide of C. neoformans contains antigenic determinants that were divided initially in three (2) and, later, four serotypes (3). There has been prior information to suggest that all serotypes might not be equally represented in environmental isolates. Walter and Coffee (4) found only one type C and no type B among 276 environmental isolates of C. neoformans, whereas 8.2 per cent of 196


Downloaded from https://academic.oup.com/aje/article-abstract/105/6/582/166618 by University of Durham user on 17 February 2018


cryptococcosis isolates were types B or C. These environmental and infection isolates were obtained from 22 states in the United States, Canada, Australia, Holland and Brazil. No attempt was made to compare patient and environmental isolates by geographic location. Felton and co-workers (5) found that all 25 isolates obtained from environmental sources in Oklahoma were serotype A. Apparently 21 of 22 patient isolates from that area were also type A. Denton and DiSalvo (6) found only one of 95 environmental isolates in Georgia was type C, the remainder being type A. Isolates from infections in that area were not studied. The importance of considering locality was underscored by a prior publication from this laboratory in which it was reported that serotypes B and C comprised 13 of 17 California isolates, whereas only five of 89 isolates from elsewhere in the United States were those serotypes (3). Comparison between environmental and patient isolates could not be made in that study because only seven environmental isolates were included, and none were from California. Although that publication referred to several B and C isolates being from "nature," those isolates were obtained from infected animals, not soil or avian droppings. The above studies suggested that currently recognized environmental sources of C. neoformans may rarely contain serotypes B or C. The only valid test of that hypothesis would be to study environmental isolates from an area in which B and C infections were reasonably frequent. The previously published study was therefore expanded to achieve this goal. MATERIALS AND METHODS

Source of isolates. Cultures of C. neoformans were obtained from our own collections, spanning the past 15 years, as well as from the generous gifts of colleagues (see footnotes). Isolation of C. neoformans from pigeon droppings in California and some other sites was made by the direct


plating method on niger seed agar (7). When fast-growing molds were encountered, mouse inoculation was also employed (8). However, isolates of C. neoformans were obtained by mouse inoculation only from sites that were also positive by direct plating. Only the direct plating isolates were serotyped. The most crucial environmental isolates used in this study were from Southern California. Three were soil isolates obtained from Los Angeles, Hot Springs and Westminster City, one isolate from each city. These isolates were sent to us by colleagues in California. Sixty-four isolates were obtained from pigeon droppings. These 64 represented six different localities, all at least 10 miles (16 km) apart from one another. The loci and number of isolates were as follows: San Fernando, 20; Torrance, 13; Redondo Beach, 10; Manhattan Beach, 10; Lomita, 9; Los Angeles, 2. Twelve other localities were sampled, but no isolates were obtained. Only extremely dry, weathered pigeon droppings from under the cages of pigeons kept by pigeon fanciers were examined. All six localities were sampled between April and June 1976. It is usually impossible to state the geographic locality in which a cryptococcal infection was acquired. For uniformity, the patient's place of residence immediately prior to hospitalization was taken as the site of infection. Identification. Cultures were identified by the criteria of Phaff and Fell (9). In addition, all isolates were found to form brown or tan colored colonies on creatinine-niger seed agar (10). Serotyping. Antisera were prepared in rabbits by intravenous injection of whole, heat-killed cryptococci. All antisera were absorbed with cells of the heterologous serotypes using methods previously published (3). As the number of isolates studied has increased, we have found that serotypes A and D form a continuous spectrum of antigenic variation. Although most iso-

Downloaded from https://academic.oup.com/aje/article-abstract/105/6/582/166618 by University of Durham user on 17 February 2018



lates were clearly A or D, some isolates were intermediate and would type as A, D or AD, depending on the antiserum and the number of absorptions. To control this variable, each A and D antiserum was checked against at least 10 A and 10 D

isolates chosen to include intermediate isolates. Another problem encountered was that primary isolates from avian sources would sometimes serotype as AD on the first subculture, then lose their ability to agglutinate with D antiserum after one or


Serotypes of man and animal* isolates ofC neoformans Origin United States Alabama Arkansas California Northernt Southernt Connecticut Delaware District of Columbia Florida Georgia Illinois Indiana Iowa Kentucky Louisiana Maryland Massachusetts Michigan Minnesota Missouri Nevada New Jersey New York North Carolina Ohio Oklahoma Pennsylvania South Carolina Tennessee Texas Virginia Washington West Virginia Wisconsin Total %



Total no. of isolates






10 1

2 0

1 0

0 0

1 1

1 0

15 2

11 21 5 1 4 4 2 3 1 4 1 3 24 0 3 4 15 0 5 5 20 5 0 9 6 3 2 23 3 3 2

0 18 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0

1 0 2 0 2 0 2 0 0 0 0 0 3 1 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0

1 1 1 0

1 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

15 49 9 1 6 4

1 0 0 1 0 1 0 0

1 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0

203 74.6

28 10.3

11 4.0

13 4.8

2 0 1

0 0

0 0 0 • o

0 0 0 0 0 0 0 o 1 0 1 1 1 0 0 2 0 0 0 0 0 0 0 11 4.0

0 0 0 0 0 0 1 0 0 6 2.2

5 3 1 4 1

3 27 1 3 5 16 1 9 6

23 5 1 12 6 3

3 23 5 3 2 272

0 0 7 Italy 0 0 0 7 0 1 0 0 0 0 1 Mexico 3 0 0 6 3 0 0 Thailand * Isolates from 1 cow (Italy), 5 cats (USA) and 3 monkeys (USA) are included. t Isolatesfromtowns in the San Francisco area were designated Northern California and those in the San Diego-Los Angeles area as Southern California.

Downloaded from https://academic.oup.com/aje/article-abstract/105/6/582/166618 by University of Durham user on 17 February 2018



from Torrance, which were D, and one from Redondo Beach, which was untypable. The difference in prevalence of serotypes B and C in environmental sources in Southern California, compared to infections in that area, was highly significant (0/67 vs. 25/49; p < 0.005). The pattern in Southern California was mirrored by the isolates from Thailand in that three of six patient isolates but none of 11 environRESULTS mental isolates were serotype B or C. The Serotypes of isolates from 277 patients geographic location of these particular paand nine infected animals are given in tients and environmental isolates within table 1. Isolates from animals were not Thailand have been published (11). presented separately because in each geoPrevalence of serotype D in environmengraphic area the serotype paralleled the tal isolates from Denmark and Italy was serotypes from human cases. Southern quite different from the United States (14/ California had an unusually high preva- 17 vs.3/89; (p < 0.005)) and was in keeping lence of types B and C (25 of 49 isolates). with the prevalence of serotype D in paThis distribution differed markedly from tient isolates (7/7) from Italy. The Danish the remainder of the United States where isolates were from pigeon houses in Aaronly 14 of 223 isolates were types B or C (p hus and had been considered serotype A or < 0.005). Serotypes B and C accounted for G by others (12). three of nine New Jersey isolates, but too DISCUSSION few isolates were studied from this area to draw conclusions. An unexpected finding In order to make unassailable concluwas that all seven isolates from Italy were sions about the effects of geography on serotype D, a significant difference com- serotypes of C. neoformans, it would be pared to only 4.8 per cent of isolates from preferable to have random unbiased samthe United States (p < 0.005). ples from areas of interest. Instead, the Serotypes of C. neoformans isolated isolates available for this and similar prior from environmental sources are given in studies were subjected to unknown biases table 2. Among the 67 isolates from South- by the collection process. With this limitaern California, all were type A except two tion in mind, several conclusions seem two further subcultures. Patient isolates were not observed to do this. As will be shown subsequently, a few isolates consistently serotyped as AD. All AD isolates were cloned repeatedly to be sure a mixture was not present. Statistical analysis. For comparison of frequencies, chi square with Yates correction for continuity was used.


Serotypes of C. neoformans isolated from bird droppings, soil and air Origin

United States Southern California Elsewhere* Denmark Italy Thailand Total %

Serotype A



64 21 2 1 11

0 0 0 0 0

0 0 0 0

99 84.6








2 1 10 4 0

0 0 0

17 14.5


0 0


• Untypable

1 0 0 0 0 1 0.8

Total no. of isolates

67 22 12 5 11 117

* Georgia, 8; Kansas, 2; Maryland, 3; Hawaii, 1; Iowa, 1; Virginia, 3; District of Columbia, 1; Oklahoma, 1; Alabama, 1; Arizona, 1.

Downloaded from https://academic.oup.com/aje/article-abstract/105/6/582/166618 by University of Durham user on 17 February 2018



reasonable. It is now clear that serotypes B and C of C. neoformans are not isolated from pigeon droppings and are isolated extremely rarely from other environmental sources currently being cultured. Considering that all serotypes grow equally well on the primary isolation medium, niger seed agar, it is unlikely that this isolation technique selectively retrieves A and D serotypes in preference to B and C. It is unknown whether mouse inoculation is more sensitive for certain serotypes. To date, no serotype B isolates have been cultured from environmental sources; and only two serotype C isolates have been recovered, both from debris within houses (4, 6). This situation is particularly of note in Southern California where about half the infections were found due to serotypes B and C, but no environmental isolates were those serotypes. It seems that serotypes B and C have a reservoir in nature that differs from serotypes A and D. The reservoir may be the same for serotypes B and C, considering that they share biochemical properties that distinguish them from serotypes A and D, as will be reported subsequently. Also, the perfect state of serotypes B and C is the same and differs substantially from that of serotypes A and D. The perfect state of serotypes B and C is Filobasidiella bacillispora (13), while that of A and D is F. neoformans (14). There is a suggestion from both environmental and infection isolates that serotype D may be more prevalent in Europe than in the United States. Seeliger (15) cited as an unpublished observation that 95 per cent of 120 isolates studied in West Germany were serotype A, but it was not stated whether he absorbed his A typing serum with D cells. Otherwise, antiserum to type A invariably will agglutinate D as well as A cells. Unpublished results from Dr. Hans Scholer have revealed D to be

the most frequent serotype in isolates from Switzerland (personal communication, 1976). Further studies from Europe will be necessary to resolve this point. REFERENCES

1. Gordon MA: The epidemiology of cryptococcosis. In: The Epidemiology of Human Mycotic Diseases. Edited by Y Al-Doory. Springfield, Charles C Thomas, 1975, pp 142-151 2. Evans EE: An immunologic comparison of twelve strains of Cryptococcus neoformans (Torula histolytica). Proc Soc Exp Biol Med 71:644646, 1949 3. Wilson DE, Bennett JE, Bailey JW: Serologic grouping of Cryptococcus neoformans. Proc Soc Exp Biol Med 127:820-823, 1968 4. Walter JE, Coffee EG: Distribution and epidemiologic significance of the serotypes of Cryptococcus neoformans. Am J Epidemiol 87:167-172, 1968 5. Felton FG, Muchmore HG, McCarty MA: Epidemiology of cryptococcosis: 1. Environmental distribution of cryptococci in Oklahoma. Health Lab Sci 11:201-204, 1974 6. Denton JF, DiSalvo AF: The prevalence of Cryptococcus neoformans in various natural habitats. Sabouraudia 6:213-217, 1968 7. Staib F, Seeliger HPR: A new selective medium for the isolation of C. neoformans from soil and fecal matter. Ann Inst Pasteur, Paris 110:792793, 1966 8. Emmons CW: Saprophytic sources of Cryptococcus neoformans associated with the pigeon (Columba livia). Am J Hyg 62:227-232, 1955 9. Phaff HF, Fell JW: Cryptococcus. In: The Yeasts. Edited by J Lodder. Amsterdam, North Holland Publishing Co, 1970, pp 1088-1145 10. Jennings AE, Bennett JE, Young VM: Identification of Cryptococcus neoformans in a routine clinical laboratory. Mycopathologia 35:256-264, 1964 11. Mondana G, Thasnakorn P: In vitro sensitivity to amphotericin B of Cryptococcus neoformans isolated in Thailand. J Med Assoc Thailand 54:405-410, 1971 12. Bach A, Fabricant C, Stenderup A: Cryptococcus neoformans: ecology and immunology. Antonie van Leeuwenhock 35 (Suppl): E35, 1969 13. Kwon-Chung KJ: A new species of Filobasidiella, the sexual state of Cryptococcus neoformans B and C serotypes. Mycologia 68:942-946, 1976 14.. Kwon-Chung KJ: A new genus, Filobasidiella, the perfect state of Cryptococcus neoformans. Mycologia 67:1197-1200, 1975 15. Seeliger HPR: Immunologie der Kryptokokkose. Mykosen 12:49-60, 1969

Downloaded from https://academic.oup.com/aje/article-abstract/105/6/582/166618 by University of Durham user on 17 February 2018

Epidemiologic differences among serotypes of Cryptococcus neoformans.

Vol. 105, No. 6 AMERICAN JOURNAL OF EPIDEMIOLOGY Copyright © 1977 by The Johns Hopkins University School of Hygiene and Public Health Printed in US...
355KB Sizes 0 Downloads 0 Views