JOURNAL OF BACTERIOLOGY, Nov. 1967, p. 1475-1479 Vol. 94, No. 5Copyright © 1967 American Society for Microbiology Printed in U.S.A.

Cryptococcus neoformansI. Nonencapsulated Mutants

G. S. BULMER, M. D. SANS, AND C. M. GUNNDepartment of Microbiology, University of Oklahoma School of Medicine,

Oklahoma City, Oklahoma 73104

Received for publication 1 September 1967

Seven nonencapsulated mutants of Cryptococcus neoformans were isolated from anencapsulated strain of human origin. Initially, the mutants were avirulent for mice.After several months of subculturing, six of the seven isolates reverted LO the encap-sulated state and possessed varying degrees of virulence. The results of these experi-ments suggest that a strong correlation exists between the presence of a capsule andthe virulence of C. neoformans.

Studies on the relationship of the capsule ofCryptococcus neoformans to the virulence of theorganism are limited and the results have beenconflicting. Drouhet, Segretain, and Aubert (2)stated that the virulence of the fungus is directlyrelated to the size of the capsule. However, Kaoand Schwarz (4), Littman and Tsubura (6), andHasenclever and Mitchell (3) failed to observesuch a correlation. A small-capsule variantisolated by Kase and Metzgar (5) induced centralnervous system symptoms in mice in 7 to 12 days,whereas the parent large-capsule strain required14 to 120 days to produce the same symptoms.

It was the purpose of these studies to isolatenonencapsulated strains of C. neoformans andcontrast the virulence of such strains to the encap-sulated parent.

MATERIALS AND METHODS

An encapsulated strain of C. neoformans (serotypeA; J. E. Walter, personal communication; strain CP-5,Veterans Administration Hospital, Oklahoma City,Okla.), hereafter referred to as CIA, was obtainedfrom a human case of cryptococcal meningitis. Thisorganism was subcultured weekly, as were all otherstrains, on Sabouraud Dextrose Agar (Difco) andmaintained under mineral oil.

Ncnencapsulated mutants of the parent strain (CIA)were obtained by irradiating an aqueous suspension ofcells (106) with an ultraviolet (UV) lamp (Miner-alight, model R51, Ultra Violet Products, San Gabriel,Calif.) at a distance of 25 cm. To insure maximalexposure to UV light, the cell suspension was placedin the bottom of a petri dish and kept under constantagitation with a magnetic stirrer. The time requiredto kill 99% of the cells was determined in preliminaryexperiments. Samples of cells irradiated to that pointwere plated onto Sabouraud Dextrose Agar and in-cubated for S to 7 days at 26 C. India ink prepara-tions of cells from dry-appearing colonies wereexamined microscopically for the presence of a cap-sule. When cells from a given colony were observed

to be nonencapsulated, subcultures were prepared onSabouraud Dextrose Agar in an attempt to obtainhomogeneously nonencapsulated clones. This processwas repeated several times.

Assimilation studies were performed in collabora-tion with Frances Felton, Veterans AdministrationHospital, Oklahoma City, Okla., according to theprocedures of Wickerham (8).

Pathogenicity of all of the strains was examined byinjecting 106 viable cells suspended in 0.5 ml of salineintraperitoneally (ip) and 0.05 ml of saline intracere-brally (ic) into waite mice obtained from the mousecolony at the University of Oklahoma MedicalSchool. Four to six animals were used for each test.India ink preparations of brain tissues from animalswhich died were examined microscopically for yeastcells. Tissue samples were also streaked onto Sa-bouraud Dextrose Agar. The cultures were incubatedfor 1 to 2 weeks at 26 C and examined for growth ofC. neoformans. Animals surviving 140 days (ic injec-tion) or 250 days (ip injection) were sacrificed, andthe organisms with which they were inoculated wereconsidered to be avirulent.

Cells from the various strains of C. neoformanswere examined for capsular constitutents by paperchromatography. For this purpose, 1.5 ml of a thickaqueous suspension of whole cells was hydrolyzed byadding 1.5 ml of 3 N HCI and heating at 100 C for 4hr. The hydrolysate was evaporated to dryness invacuo to remove HCI. The residue was then taken upin a minimal amount of 80% ethyl alcohol or distilledwater and assayed for carbohydrates by means ofdescending paper chromatography with SS filterpaper and solvent systems consisting of n-butanol-pyridine-0.1 N HCI (5:3:2, v/v, 16 hr) and n-butanol-ethyl alcohol-water (5:1:4, v/v, 20 hr). Chromato-grams were developed with 2-amino biphenyl.Standards consisted of D-glucose, D-xylose,D-galactose, D-mannose, D-ribose, D-glucuro-nolactone, and D-glucuronic acid.

RESULTSIn several experiments, the time required for

UV light to kill 99% of the cells of the1475

on June 25, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

BULMER, SANS, AND GUNN

encapsulated parent strain, CIA, was found to beapproximately 4 min. India ink preparations of35 isolates revealed that the organisms were, forthe most part, devoid of a visible capsule. In manyinstances, the cells were peculiar in shape, i.e.,long and slender, ovoid, twisted, or banana-shaped. Many of the isolated strains were com-posed of cells which were similar microscopicallyto other Cryptococcus spp. For a period of 5 to 6weeks, samples of these colonies were subculturedweekly onto Sabouraud Dextrose Agar inattempts to isolate dry-appearing colonies which,microscopically, were composed of cells that werehomogeneously devoid of a capsule and had amorphology similar to that of C. neoformans.From the original 35 isolates, 7 were selectedwhich met these criteria. These mutants will bereferred to hereafter as MI, M2, M3, M4, M5,

M6, and M7. India ink preparations of the encap-sulated parent strain, CIA (A), and one of thenonencapsulated mutants (B) are shown in Fig. 1.The mucoid appearance of the colonies of CIA(C), as compared with the dry, wrinkled coloniesof the nonencapsulated mutant (D), is also shown.

All seven of the nonencapsulated mutants grewat 37 C, were urease-positive, and turned brownon Niger seed medium (7). The results of assimila-tion studies indicated that, with the exception ofM3, all of the mutants had an assimilation patternsimilar to the parent strain (Table 1).

In a further attempt to determine more preciselywhether or not the mutants were truly nonencap-sulated, cells from all of the strains, plus theparent CIA, were hydrolyzed and examined forcapsular constituents by paper chromatography.The results of these studies (Fig. 2) revealed that

FIG. 1. Encapsulated Cryptococcus neoformans CIA shown microscopically in India ink preparation (A) antd ingross appearance (C) of colony. Nonencapsulated strain shown microscopically in India ink preparation (B) and ingross appearance ofcolony (D).

1476 J. BACTERIOL.

on June 25, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

NONENCAPSULATED CRYPTOCOCCUS NEOFORMANS MUTANTS

TABLE 1. Results of assimilation tests for Crypto-coccus neoformans, parent strain(CIA), and seven nonencapsulated

mutants (MJ-M7)

4) 4) 4) ~~~04)0 0 o uC's 0Z C OC

CIA 0 +++ +++ +++ +++ 0M1 0 +++ +++ +++ +++ 0M2 0 +++ +++ +++ +++ 0M3 + + +++ 0 0 0M4 0 +++ +++ +++ + 0M5 0 +++ +++ +++ +++ 0M6 0 +++ +++ +++ +++ 0M7 0 +++ +++ +++ +++ 0

the parent encapsulated strain, CIA, containedgalactose, mannose, xylose, and glucuronic acid(also shown in the lactone form), all of which,according to Blandamer and Danishefsky (1), areconstituents of the capsule of C. neoformans. M7,as well as all of the other mutants, lacked all fourof these chemicals.

Since, except for the absence of a capsule, themutants appeared to be similar to the encap-sulated parent strain, it was felt that the functionof the capsule in the pathogenesis of crypto-coccosis might be assessed by determiningwhether or not these strains were virulent formice. The results of these studies are recorded inTable 2 (1 month postisolation). The encap-sulated strain CIA killed mice in 8 days (ic inocu-lation) or 24 days (ip inoculation). All seven ofthe nonencapsulated mutants, regardless of theroute of inoculation, failed to kill mice.

All seven of the mutants were subculturedweekly, and colonies were observed for grossappearance. The cells were observed microscop-ically in India ink preparations. Afier 2 months ofsubculturing, all of the mutant strains were stillnonencapsulated (Table 2). The results ofvirulence studies indicated that MI and M2 killedmice slowly when they were inoculated ic, andthe brains of the animals contained encap-sulated yeast cells. All of the mutants wereavirulent after ip inoculation.

After 4 months of weekly subculturing, coloniesof the M4 organisms began to appear mucoid;however, in India ink preparations the cells stillappeared to lack a capsule. A thick, aqueoussuspension of these cells was dispensed in 3-mlamounts into test tubes (13 X 100 mm) andshaken at 37 C for 1 hr. The cells were removedby centrifugation, and ethyl alcohol was addedto the supernatant fluid to give a final concentra-tion of 90%. After storage for 15 hr at 8 C, a

white precipitate was observed. This material wasremoved by centrifugation, hydrolyzed, andchromatographed according to the techniquesdescribed previously. All four capsular constit-uents were present.

After 5 months of subculturing, all but one ofthe mutants produced mucoid-appearing col-onies. The exception (M7) continued to grow inthe rough form. Microscopic examination ofIndia ink preparations revealed that the cellswere nonencapsulated, whereas 60 to 99% of thecells of strains MI to M6 had capsules. The per-centages of encapsulated cells and the results ofvirulence studies on these strains are recorded inTable 2. It can be seen that three of the revertedstrains (Ml, M2, and M3) possessed varyingdegrees of virulence, whereas M4, M5, M6, andM7 were avirulent.

After 12 months of subculturing, from 51 to99% of the cells of strains Ml, M2, M3, M4, M5,and M6 were encapsulated. Cells of M7 remained

FIG. 2. Chromatogram of hydrolyzed Cryptococcusneoformans. (I) Encapsulated parent strain, CIA. (2)A nonencapsulated mutant. (S) Standards (Ga = galac-tose; GI = glucose; Ma = mannose; Xy = xylose;Ri = ribose; Gu. La. = glucuronolactone; GI. A. =

glucuronic acid).

:.:i: iia>- -

VOL. 94, 1967 1477

imL..6L- M.-

on June 25, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

BULMER, SANS, AND GUNN

TABLE 2. Virulence and per cent encapsulation (PE) in vitro of Cryptococcus neoformans, parent straint(CIA) and seven mutants (MJ-M7), during 12 months of subculturing

Time postisolation"

1 month 2 months 5 months 12 months

Strain Mortality Survival PE Mortality Survival PE Mortality Survival PE Mortality tSurival PEtmb time time time

ip ic ip ic ip ic ip ic ip ic ip ic ip ic ip ic

% % days % % days days % % days days % % days days

CAI 100 100 24 8 100 100 100 21 7 100 100 100 18 8 100 100 100 13 5 100Ml Oc 0 - 0 0 100 - 28 0 0 100 - 21 60 25 100 19 16 60M2 0 0 - - 0 0 100 - 12 0 50 100 26 7 50 0 100 - 6 51M3 0 0 -- 0 0 - - 0 25 0 175 - 98 100 100 14 10 98M4 O - 0 0 0 -O- 0 - 99 100 100 26 5 75M5 0 0--0 0 0-O- 00 85 100 100 36 16 80M6 0 0--0 0 0-- 00 0 - 85 100 100 57 8 80M7 0 0 - 0 0 0-- 0 0 0 0 0 - 0

a After isolation of mutants.b Survival time-Mean c days survival of animals that died of cryptococcosis.c Avirulent 140 and 250 days post ic and ip inoculations, respectively.

nonencapsulated. Cells of MI through M6 werevirulent to varying degrees, whereas M7 wasavirulent (Table 2).The results of assimilation studies on the

mutants after 12 months of subculturing indicatedthat all of the mutants, except M3, had patternsidentical to that of the parent strain CIA. Theassimilation pattern for M3 was the same as isshown in Table 1.

DIscussIoN

In the present studies, UV irradiation of a

heavily encapsulated strain of C. neoformansinduced the development of mutants that weredevoid of a capsule. Microscopic observations ofIndia ink preparations of the mutants suggestedthat a capsule was lacking. More definitive proofwas obtained by chromatographing hydrolyzedwhole cells. By this method, no capsular con-stituents could be detected.The results of the assimilation tests, growth of

the organisms at 37 C, production of urease, anddevelopment of a brown pigment on Niger seedmedium indicated that six of the seven mutantswere similar to the parent strain. One mutant,M3, had a different assimilation pattern. Thus,according to tests used, six of the seven mutantswere similar to the parent strain except that theylacked a capsule, formed rough colonies, and wereavirulent for mice.Mutant M4, like MI through M6, reverted to

the encapsulated state after 5 months of weeklysubculturing. After 4 months of subculturing, thecolonies of this strain were mucoid. Microscop-ically, however, the cells still appeared to be

nonencapsulated. This suggests that, as a prelim-inary step to the encapsulated state, the organismproduced a soluble capsular material. Thus, thismutant differed from the others, and it seemspossible that at least two distinct genetic loci areinvolved in the development of the typical encap-sulated state. A third locus might also beimplicated if one interprets the instabilities ofstrains MI through M6 and the stability of M7as representing different loci. The results of theexperiments comparing the virulence of encap-sulated and nonencapsulated strains indicate thatvirulence for mice is closely associated with thepresence of a capsule.After 1 month of subculturing, the mutants lackeda capsule and were avirulent. After 2 months ofsubculturing, all of the mutants still lacked acapsule. Two of them killed mice slowly afterintracerebral inoculation; however, encapsulatedcells were observed in the brains of the animals.After 12 months of subculturing, six of the sevenmutants had a capsule and, for the most part,were virulent. The one mutant which remainedconsistently nonencapsulated throughout the12 months of subculturing failed to kill mice.Thus, from the 1, 2, and 12 months data it canbe stated that without a capsule C. neoformansCIA is avirulent and with a capsule it is virulent.

It should be pointed out, however, that thisconclusion is not supported by the virulencestudies conducted with mutants that were sub-cultured for 5 months. In this case, six of the sevenmutants had a degree of encapsulation similar tothat seen at the 12-month period; yet three ofthese strains were avirulent and three had some

1478 J. BACTERIOL.

on June 25, 2018 by guesthttp://jb.asm

.org/D

ownloaded from

NONENCAPSULATED CRYPTOCOCCUS NEOFORMANS MUTANTS

degree of virulence. These results suggest that thecapsule, as it existed at that time, did not play akey role in the pathogenesis of cryptococcosis.It is possible that the capsule at the 5-monthperiod differed somehow from that seen at the12-month period. The only criterion used todetermine the nature of the capsular material atthe 5-month period was microscopic examinationof India ink preparations. Such observations offerno indication as to the chemical structure orantigenicity of the material. The observation thatone mutant (M4) progressed through at least twostages of development before an intact capsulecould be seen microscopically indicates thatreversion from a nonencapsulated to encapsulatedstate may require several steps. The capsulesobserved after 5 months of subculturing may,then, have been in an immature state as far asbeing a functional virulence factor.

If the capsule is not a virulence factor, itbecomes difficult to explain its apparent closeassociation with virulence at the 1-, 2-, and 12-month periods. Furthermore, M7 never revertedto an encapsulated form, and was consistentlyavirulent. If the results of further investigationssupport the contention that the capsule of C.neoformans is a virulence factor, it will be thefirst such entity described in fungi which arepathogenic for animals. The fact that Crypto-coccus spp. possess a capsule and are avirulentdoes not detract from this suggestion, sincevirulence need not necessarily be linked solely toa single factor, but rather to the overall integrityof the organism.

It is interesting that the results of earlier studieson the significance of the capsule of C. neoformansin virulence are conflicting. It is possible that thestrains used previously were not entirely non-encapsulated. Thus, the degree of encapsulation

may not be important, but rather the presence orabsence of a capsule, and also the possibility thatcapsular material may be present in a diffusiblestate in vivo.

ACKNOWLEDGMENTSThis investigation was supported by Public Health

Service Research Career Program Award (G. S. B.)5-K03-AI-13188 from the National Institute ofAllergy and Infectious Diseases and Public HealthService grant Al 05022.

LITERATURE CITED1. BLANDAMER , A., AND I. DANISHELSKY. 1966. In-

vestigations on the structure of capsular poly-saccharide from Cryptococcus neoformans, typeB. Biochim. Biophys. Acta 117:305-313.

2. DROUHET, E., G. SEGRETAIN, AND J. P. AUBERT.1950. Polyoside capsulaire d'un champignonpathogene Torulopsis neoformans relation avecla virulence. Ann. Inst. Pasteur 79:891-900.

3. HASENCLEVER, H. F., AND W. 0. MITCHELL. 1960.Virulence and growth rates of Cryptococcus neo-formans in mice. Ann. N.Y. Acad. Sci. 89:156-162.

4. KAO, C. J., AND J. SCHWARZ. 1957. Isolation ofCryptococcus neoformans from pigeon nests,with remarks on the identification of virulentcryptococci. Am. J. Clin. Pathol. 27:652-663.

5. KASE, A., AND J. F. METZGAR. 1962. Isolation of adry variant from a mucoid strain of Cryptococ-cus neoformans and preliminary comparativestudies. J. Bacteriol. 83:926-928.

6. LrITMAN, M. L., AND E. TSUBURA. 1959. Effect ofdegree of encapsulation upon virulence ofCryptococcus neoformans. Proc. Soc. Exptl. Biol.Med. 101:773-777.

7. STAIB, F. 1962. Cryptococcus neoformans andGuizotia abyssinica. Z. Hyg. Infektionskrankh.148:466-475.

8. WICKERHAM, L. J. 1951. The taxonomy of yeasts.I. Techniques of classification. U.S. Dept. Agr.Tech. Bull. 1029.

1479VOL. 94, 1967

on June 25, 2018 by guesthttp://jb.asm

.org/D

ownloaded from