OUP user menu

Comparative analysis of the intergenic spacer regions and population structure of the species complex of the pathogenic yeast Cryptococcus neoformans

Mara R. Diaz, Teun Boekhout, Traci Kiesling, Jack W. Fell
DOI: http://dx.doi.org/10.1016/j.femsyr.2005.05.005 1129-1140 First published online: 1 December 2005


Cryptococcus neoformans is an opportunistic basidiomycete responsible for the high incidence of cryptococcosis in patients with AIDS and in other immune-compromised individuals. This study, which focused on the molecular structure and genetic variability of the two varieties in the C. neoformans and Cryptococcus gattii species complex, employed sequence analysis of the intergenic spacer regions, IGSI and IGSII. The IGS region is the most rapidly evolving region of the rDNA families. The IGSI displayed the most genetic variability represented by nucleotide base substitutions and the presence of long insertions/deletions (indels). In contrast, the IGSII region exhibited less heterogeneity and the indels were not as extensive as those displayed in the IGSI region. Both intergenic spacers contained short, interspersed repeat motifs, which can be related to length polymorphisms observed between sequences. Phylogenetic analysis undertaken in the IGSI, IGSII and IGSI + 5S rRNA + IGSII regions revealed the presence of six major phylogenetic lineages, some of which segregated into subgroups. The major lineages are represented by genotypes 1 (C. neoformans var. grubii), genotype 2 (C. neoformans var. neoformans), and genotypes 3, 4, 5 and 6 represented by C. gattii. Genotype 6 is a newly described IGS genotypic group within the C. neoformans species complex. With the inclusion of IGS subgenotypic groups, our sequence analysis distinguished 12 different lineages. Sequencing of clones, which was performed to determine the presence of multiple alleles at the IGS locus in several hybrid strains, yielded a single IGS sequence type per isolate, thus suggesting that the selected group of cloned strains was mono-allelic at this locus. IGS sequence analyses proved to be a powerful technique for the delineation of the varieties of C. neoformans and C. gattii at genotypic and subgenotypic levels.

  • Yeast
  • Intergenic spacer (IGS)
  • Length polymorphism
  • Repeated motifs
  • Cryptococcus neoformans

1 Introduction

The basidiomycetous, encapsulated yeast Cryptococcus neoformans (Sanfelice) Vuillemin is a prevalent clinical opportunistic pathogen, which causes life-threatening infections of the central nervous system in immunocompromised and immunocompetent hosts [1,2]. Estimates indicate that cryptococcosis occurs in about 6% to 10% of AIDS patients in the United States and Western Europe, and in 15% to 30% of patients in sub-Saharan Africa [14].

Although C. neoformans var. grubii is the most common cause of fungal meningitis in AIDS and immunocompromised individuals [1], C. neoformans var. neoformans and Cryptococcus gattii have been reported as ethiological agents of cryptococcosis in HIV-infected patients [58]. C. neoformans var. neoformans and C. neoformans var. grubii occur worldwide and are frequently isolated from bird droppings, but also in trees, soil, house dust, and domestic animals, e.g., cats and cows [1,911], whereas C. gattii is mainly found in tropical and subtropical areas (Australia, Asia, South America, South California and Southern Europe). The geographic boundary of this species has been expanded with the recent outbreak of C. gattii in the Vancouver Islands, British Columbia [12,13]. According to a recent proposal, which raised the status of C. neoformans var. gattii to species status, C. neoformans represents a species complex comprising two species: C. gattii (serotypes B and C) and C. neoformans with variety grubii (serotypes A and AD) and variety neoformans (serotype D) [14,15]. The two species differ in their capsule polysaccharide structure, antigenic structure, molecular and morphological characteristics, epidemiology, virulence characteristics, ecology, and geography [1].

Genetic heterogeneity has been observed within the species complex [1620]. AFLP analyses and partial sequence analyses of the intergenic spacer region I (IGSI) has shown considerable genetic divergence between the species and varieties [17,19]. Support for distinct genetic lineages for C. neoformans var. neoformans and C. neoformans var. grubii have also been based on URA5 sequences and restriction fragment length polymorphism (RFLP) patterns employing a CNRE 1 probe [7,21].

The scope of the present study is to elucidate and compare the genetic diversity and phylogeny within the C. neoformans species complex based on a comparative study involving complete sequence analysis of IGSI and an analysis of the 609–675-bp region of the IGSII region. This region has been frequently used as a tool for species identification and phylogenetic studies [19,2225]. A detailed analysis of the molecular structure organization of the two intergenic spacers is presented.

2 Methods

2.1 Isolates

One hundred and seven clinical and environmental DNA isolates originating from different geographic areas were analyzed. The source of isolation, genotypes, and serotypes are described in Table 1. Data were obtained from the CBS collection, Boekhout et al. [17], or by information provided by the depositors of the isolates.

View this table:
Table 1

List of experimental strains

StrainSource of isolationSerotypeIGS genotypeAFLP TypeGenBank #
C. neoformans var. grubii
CBS 879Ulcerated cheekA1a1=DQ007972
CBS 886UnknownA1a1DQ007978
CBS 916UnknownA1a1=DQ007972
CBS 935UnknownA1a1=DQ007978
CBS 1143Cerebrospinal fluidA1a1DQ007974
CBS 1144Cerebrospinal fluidA1a1DQ007973
CBS 1931SoilA1a1DQ007972
CBS 1932SoilA1a1ADQ007977
CBS 1933Mastitic cow, USAA1a1=DQ007979
CBS 4572Cerebrospinal fluidA1a1=DQ007972
CBS 4868Sputum, The NetherlandsA1a1DQ007983
CBS 5756UnknownA1a1=DQ007978
CBS 6961Man, Oklahoma, USAA1a1DQ007975
CBS 7779Urease negative isolate from AIDS patient, ArgentinaA1a1=DQ007972
CBS 7812Cerebrospinal fluid, USAA1a1DQ007982
RV 46115Plants, IndiaA1b1=DQ007985
RV 52733Pigeon droppings, BelgiumD1a3=DQ007972
RV 53794Canary bird droppings, BelgiumD1a3=DQ007978
RV 55446House dust, ZaireA1a1=DQ007979
RV 55447Air inside house, ZaireA1a?=DQ007979
RV 55451Cockroach, ZaireA1b1DQ007984
RV 58145Wood, ZaireA1a1=DQ007979
RV 58146Wood, ZaireA1c1ADQ007986
RV 59351Parrot dropping, BelgiumA1b1=DQ007984
RV 59369Parrot dropping, BelgiumA1b1=DQ007984
RV 60074Skin, cryptococcosis, BelgiumA1a1DQ007981
RV 61756Man, Belgium (visited Zaire)AD1a1A=DQ007972
RV 62210Cerebrospinal fluid, Belgium fluid from AIDS patient, BelgiumA1b1=DQ007984
RV 64610AIDS patient, RwandaA1a1A=DQ007980
RV 65662Man, Portugal (visited Venezuela)A1a1ADQ007980
RV 66025Cryptococcoma, BelgiumA1a1DQ007979
RDA 1335 AVB0AIDS patient, The NetherlandsA1b1=DQ007984
RDA 1371 AVB2AIDS patient, The NetherlandsA1a1=DQ007978
RDA 1369 AVB3AIDS patient, The NetherlandsA1a1=DQ007978
RDA N/A AVB5AIDS patient, The NetherlandsA1a1=DQ007972
RDA 1549 AVB7AIDS patient, The NetherlandsA1a1=DQ007978
RDA 4092 AVB10AIDS patient, The NetherlandsA1a1=DQ007979
RDA 4094 AVB11AIDS patient, The NetherlandsA1b1=DQ007984
RDA 4054 AVB12AIDS patient, The NetherlandsA1a1=DQ007978
RDA 4091 AVB13AIDS patient, The NetherlandsA1b1=DQ007985
BD2AIDS patient, FranceA1a1DQ007976
WM 164*Pigeon droppings, AustraliaA1c1ADQ007987
WM 553*House dust, BrazilA1c1A=DQ007987
WM 554Dust from pigeon, BrazilA1b1DQ007985
WM 555Dust from pigeon, BrazilA1a1=DQ007972
WM 715Pine needlesA1a1=DQ007972
WM 716Woody debris Eucalyptus camaldulensis, AustraliaA1a1=DQ007978
Hamdan 214L*AIDS patient, BrazilA1c1A=DQ007987
C. neoformans var. neoformans
CBS 131Institut Pasteur, FranceAD2c3=DQ007943
CBS 132Fermenting fruit juice, ItalyD2c3DQ007944
CBS 464Laboratoire de Parasitologie, Paris, FranceA2c3DQ007945
CBS 882Nasal tumor of horse, type strain of Torula nasalis, USAD2a2=DQ007940
CBS 888UnknownD2a2DQ007940
CBS 918Dead white mouse, The NetherlandsD2c2=DQ007943
CBS 939UnknownD2c3=DQ007943
CBS 950TumorAD2c3=DQ007943
CBS 1584UnknownA2c?=DQ007943
CBS 4194Spleen, GermanyD2a2DQ007942
CBS 5467Milk from mastitic cow, SwitzerlandD2c2=DQ009483
CBS 5474Mastitic cowD2c2=DQ007943
CBS 5728Nonmeningitic cellulitis and osteomyelitis, USAD2c2=DQ007943
CBS 6885Lesion on bone in man, USAD2c2DQ007943
CBS 6886Dropping of pigeon, DenmarkD2a2DQ007941
CBS 6900Genetic offspring of CBS 6885 × CBS 7000D2b2DQ007946
CBS 6901Genetic offspring of CBS 6885 × CBS 7000D2b2DQ007947
CBS 7814Air, BelgiumD2c2=DQ007943
CBS 7815Pigeon droppings, CzechoslovakiaD2c2=DQ007943
CBS 7816Cuckoo dropping, ThailandD2b2DQ007950
CBS 7824UnknownD2bN/ADQ007948
CBS 7826UnknownAD2bN/ADQ007949
RV 52755Cerebrospinal fluid, BelgiumD2c3=DQ007943
RV 62692Skin cryptococcosis, BelgiumD2c2DQ007951
BA1AIDS patient, FranceD2c3=DQ007943
BA3AIDS patient, FranceAD2c3=DQ007943
BA4AIDS patient, FranceAD2c3=DQ007943
C. gattii
CBS 919Meningoencephalic lesion, type strain Torulopsis neoformansB4b4DQ007958
CBS 1930Sick goat, ArubaB36DQ007952
CBS 5757UnknownB4a4DQ007955
CBS 5758UnknownC55DQ007964
CBS 6290Man, ZaireB4c4DQ007959
CBS 6289Subculture of type strain RV 20186B4c4DQ007963
CBS 6955Spinal fluid, type strain of Filobasidiella bacillispora, USAC55DQ007966
CBS 6956Sputum, USAB36DQ007954
CBS 6992ManB4a4=DQ007955
CBS 6994Cerebrospinal fluid, USAC55=DQ007966
CBS 6996ManB55DQ007965
CBS 6998Cerebrospinal fluid, ThailandB4a4DQ007956
CBS 7748Air in hollow, Eucalyptus camaldulensis, AustraliaB4b4DQ007957
CBS 7750Bark debris of E. camaldulensis, USAB36=DQ007954
RV 5265Cerebrospinal fluid, ZaireB4c4DQ007962
NIH 139Patient, USAC55=DQ007966
NIH 178Patient, USAC55=DQ007966
IMH 1658 CBS 8684Nest of wasp, UruguayB36DQ007953
48ALung of a goat, SpainB4c4DQ007960
52ABrain of a goat, SpainB4c4=DQ007960
56AGut of a goat, Spain,B4c4=DQ007960
59ALung of a goat, SpainB4c4DQ007961
60ALung of a goat, SpainB4c4=DQ007961
CGBMA1*Pink shower tree, BrazilB36DQ007971
CGBMA6*Pink shower tree, BrazilB36DQ007970
CGBMA15*Pink shower tree, BrazilB36=DQ007971
OITIGYPI10*Pottery tree, BrazilBC36DQ007969
OITIGYPI15*Pottery tree, BrazilB36=DQ007969
WM 161*UnknownB55DQ007968
WM 176Eucalyptus tree, USAB4b4=DQ007958
WM 779Human, cerebrospinal fluid, IndiaC67=DQ007967
B-5742Eucalyptus tree, USAC67DQ007967
  • CBS, from Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands; RV, from collection previously held at the Laboratory of Mycology, Institute of Tropical Medicine, Antwerp, Belgium, presently in the Scientific Institute of Public Health, Brussels, Belgium; RDA, from Erasmus Medical Center Rotterdam, The Netherlands; BD, BA, from Institut Pasteur, Paris, France; WM, from Wieland Meyer, Westmead Hospital, Sydney, Australia; NIH and B-5742, from National Institutes of Health, Bethesda, MD, USA; IMH, from Instituto de Hygiene, Montevideo, Uruguay; 48A, 52A, 56A, 59A 60A, from Dr. J. Torres, Barcelona, Spain; CGBMA & OITIGYP from M. Lazera, Instituto Oswaldo Cruz, Rio de Janeiro, Brazil.; GenBank#: GenBank accession number; *: IGSII sequence unavailable; = Identical sequence; N/A = not available

2.2 DNA isolation and PCR reaction

DNA was isolated from culture cells as described by Fell et al. [26] using lysing enzyme and QIAmp Tissue kit (Qiagen, Valencia, CA, USA) or by the CTAB method[27].

The DNA amplification of the IGS regions employed a primer walking technique using different sets of universal and specific primers yielding amplicon sizes ranging ∼1.5 to 2 kb. The set of primers, which involved a forward and a reverse primer, used to generate amplicons in the IGSI region, were: (a) Lr12F: 5′CTGAACGCCTCTAAGTCAGAA (universal forward primer located in 28SrDNA) and 5SR: 5′GCACCCTGCCCCGTCCGATCC (reverse primer located at position 44–25 of the 5SrDNA gene; (b) Lr12F and IG3R: 5′TGATTCAGCTAGCCAGTAA (reverse primer located at position 619–600 of the IGSII region); (c) Lr12F and IG4R: 5′GTCGCACCCAGTCGCACCTC (reverse primer located at position 820–801 of the IGSII region). Analysis of the IGSII region used the primer combinations: (a) IG1F: 5′CAGACGACTTGAATGGGAACG (forward primer located at position 3613–3633 of the LrRNA region) and IG4R; (b) IG1F and NS1R: 5′GAGACAAGCATATGACTAC (reverse primer located in the18SrDNA); (c) IG2F: 5′CAACAGCTTTCTATGCA (forward primer located at position 777–793 of the IGSI and IG4R; (d) IG2F and NS1R. The strain, CBS 882, was used as reference strain for primer positions, except for the IG4R primer, which employed ATCC 24067.

The PCR reaction for sequence analysis was carried out in microtubes containing a master mix with a final volume of 100 μl. The master mix contained: Target DNA; 10 mM Tris HCl (pH 9); 5 mM KCl; 0.1% Triton X-100; 2 mM MgCl2; 100 nM forward and reverse primers; 2.5 U of AmpliTaq DNA polymerase; dNTPs containing 200 μM each of dGTP, dCTP, dGTP and dATP. The PCR reaction was performed for 40 cycles in a MJ Research PTC 100 thermocyler (GMI, Ramsey, MN, USA) as follows: 2 min denaturating step at 94 °C, 1 min annealing at 57 °C and 3 min extension at 72 °C, followed by a final extension 7 min at 72 °C.

2.3 Cloning

PCR products generated by the primer combination IG1F and IG2R (5′ATG CAT AGA AAG CTG TTG G) were separated and eluted with QIAquick PCR purification kit from Qiagen, and cloned into the pCR 2.1 vector kit according to the manufacturer's instructions (Invitrogen-Carlsbad, CA, USA). Of each ligation, 2 μl were transformed into INVα F′ competent cells following the TA cloning kit procedure (Invitrogen-Carlsbad, CA, USA). Twenty-four white colonies were isolated from each cloned strain and grown overnight in LB liquid media. Plasmid preparations were performed using the Wizard Plus SV Miniprep DNA Purification System from Promega (Madison, WI, USA). All 24 plasmids were sequenced with an Applied Biosystems 3730 DNA Analyzer (Foster City, CA, USA) using a standard manufacturer protocol. Sequence analysis was performed using the SeqMan program from DNASTAR (Madison, WI, USA).

2.4 Cycle sequence analysis and data analysis

For cycle sequence analysis, the IGSI region was divided in two segments: IGSI(a) and IGSI(b). Cycle sequence primers for IGSI(a) employed the forward primer, IG1F and the reverse primer, IG2R. IGSI(a) sequences of this segment had been deposited in EMBL [19]. IGSI(b) segment employed IG2F as a forward primer and 5SR as the reverse primer. The IGII region sequences were obtained using the forward primer 5SF and the reverse primer IG3R. For Genbank accession numbers we refer to Table 1.

The sequences were obtained with a Li-Cor (Lincoln, NE, USA) NEN Global IR2 Automated Sequencer using the manufacturer's protocol. Nucleotide sequences were read using Base image IR (Li-Cor) software and edited using AlignIR (Li-Cor). Sequence alignment was made with MegaAlign (DNA Star, Inc, Madison) and by manual adjustment. The phylogenetic trees were computed with PAUP*4.0 using Parsimony analyses (heuristic search, stepwise addition, random addition, nearest neighbor, 100 maximum tree). Gaps were represented as missing data. Each character was treated as an independent, unordered, multiple character of equal weight. The reliability of the clusters was calculated using bootstrap analysis with 500 replications.

3 Results

3.1 IGS sequence analysis

The sequence strategy used three sets of primers, which yielded three different segments varying in sizes: IGSI(a): (657–822 bp), IGSI(b): (436–599 bp) and IGSII (609–657 bp). Our partial IGSII sequence analyses lacked ∼460 characters at the 3′ end. This segment of the IGSII region was a fairly homogenous area with approximately 14-bp base pair differences between C. neoformans var. neoformans and C. gattii. Complete sequences of the IGSI region were obtained combining the segments, IGSI(a) and IGSI(b). When all three segments (IGSI(a), IGSI(b) and IGSII) were pooled, sequences containing over 2 kb were generated (including 118 bp from the 5S rRNA gene) and were denominated as IGSI + 5S rRNA + IGSII.

A map depicting the organization of the intergenic spacer regions is illustrated in Fig. 1. The first 55 bp of the IGSI(a) region, which is located between the 28SLrDNA and the 5S RNA gene, showed nearly identical sequences for all the genotypes. Beyond this point, divergence in sequences was pronounced between the genotypes. These divergences were the result of nucleotide substitutions and the presence of short indels. Details about the sequence of the IGSI(a) region have been published elsewhere [19]. The first segment of the IGI(b) (∼355 bp in length), which starts at position 877 bp in our alignment, displayed random nucleotide substitutions and few indel areas of one or two bp, except at position 1049, where a 7-bp indel was located. Several long indel areas (location: 1232–1467), consisting of up to 118 bp, were also observed and were the most prominent feature of the region. At the 3′ end of the IGSI, a modified TATA box (position 1500) with a TATAAT consensus sequence was identified flanking the 5S RNA gene of IGS genotypes 1 and 2 (Fig. 1). A high similarity in sequences was found in the 5S rRNA gene, which divided the IGS region into two regions (Fig. 1). Immediately following the 3′ end of the 5S RNA, there was a CT- and T-rich area containing two consecutive blocks of seven T motifs separated by GC residues. This area, which represents the 5′ end of the IGSII, has been described as the site where the RNA polymerase III terminator binds [28]. About 30 bases downstream, an area comprising over 80 nucleotides downstream, contained mostly purine residues, G + A, arranged in GAGA, GAAAA or GAAA tandem repeats. The incidence of GAGA motifs, which extended from position 1733 to 1765, was more frequent in genotypes 1 and 2. The GAGA repeats also occured in the IGI(a) region [19]. Another feature of the IGSII was the presence of consecutive modules of TATA and ATAATA, extending from position 1938 to 2010.

Figure 1

Diagrammatic presentation of the repeat motifs, locations and their variations in the IGSI and IGSII regions of C. neoformans species complex. Numbers indicate location.

The longest consensus of repeats consisted of a 19-bp motif: GTCATGGGGGACTTGGGAG. This motif was found at five locations within the IGSI(a) region. Variations of the motif between locations and genotypes are displayed in Fig. 1. Genotypes 3–4–5–6 of C. gattii displayed the most divergent sequences at position 432, with a similarity value of 15.8%. Low identity values (26.3–31.6%) were also observed at position 421 for genotypes 1 and 2 of C. neoformans. In contrast, at location 341 genotypes 1 and 2 displayed similarity values ranging from 95% to 100%. Other short repeat and imperfect copies were identified, among them CAAAAAATT occurred four times in IGSI(a), four times in IGSI(b) and three times in IGSII (Fig. 1). Another repeat, GTTTTT, occurred once in IGSI(a) and nine times in IGSII (Fig. 1). Beginning at position 1783, genotype 1c displayed five consecutive copies of the repeat, followed by genotypes 1a and 1b, with three copies each. In contrast, genotypes 2, 3, 4, 5 and 6 showed nucleotide variations and/or deletions of the repeat. Overall, most of the motifs were found as consecutive tandem arrays and/or interspersed throughout the sequence.

3.2 Identity values

The percentage of identity, which was calculated by sequence comparisons using the DNA Star Program (Clustal W alignment with manual adjustments), within and between the members of the main phylogenetic clusters in IGSI + 5S rRNA + IGSII regions are shown in Table 2. High identity or similarity values were obtained within each genotypic group. The levels of similarity values within the main genotypes were as follows: (a) genotype 1: 97.3–100%; (b) genotype 2: 98.0–100%; (c) genotype 3: 99.5–100%; (d) genotype 4: 99–100%; (e) genotype 5: 99.7–100% and genotype 6: 100%. Due to the sequence variability between the genotypes, lower sequence similarity values were obtained. For example, some strains within genotype 1 differed by a maximum of 78.5%, 66.9%, 66.0%, 66.6% and 66.3% from genotypes 2, 3, 4, 5 and 6, respectively (Table 2).

View this table:
Table 2

Sequence identities among and within the genotypes of the C. neoformans species complex

Genotype 1Genotype 2Genotype 3Genotype 4Genotype 5Genotype 6
  • Percent identity values were determined with Clustal W alogorithm alignment generated by MegAlign. IGSI + 5SrRN + IGSII.

3.3 Length polymorphism

The length heterogeneity between the genotypes in the IGSI region was highly variable, with genotype 1 differing by an average of 185, 146, 152, 151 and 145 bp from genotypes 2, 3, 4, 5 and 6. The highest variation was found between some members of genotypes 1b and 2a, which differed by approx. 194 bp. The IGSII region displayed less length heterogeneity. The highest variation in the IGSII region (48 bp) was observed between genotypes 2b and 3. When both intergenic spacers where combined, including the 5S rRNA gene, (IGSI–5S rRNA–IGSII), length polymorphism among and within the genotypes were: genotype 1: (2157–2167 bp); genotype 2: (1963–1983 bp); genotype 3: (2046–2047 bp); genotype 4: (2006–2024 bp); genotype 5: (2017–2018 bp) and genotype 6: (2034 bp). A maximum length variation of 203 bp was documented between genotypes 1b and 2b.

3.4 Phylogenetic analysis

Phylogenetic analysis of each intergenic spacer (IGSI, IGSII) recovered the same major genotypic distribution as the combined sequence IGSI + 5S rRNA + IGSII. Therefore, to avoid redundancy, we present the tree derived from IGSI + 5S rRNA + IGSII sequence analyses (Fig. 2). The tree topology displayed six distinct genotypes representing the three varieties as three distinct phylogenetic lineages. The first clade or IGS genotype 1 was mostly represented by serotype A isolates belonging to C. neoformans var. grubii. Few isolates, which belong to serotypes D and AD, namely RV 52733 (D), RV 61756 (AD), and RV 53794 (D), also occurred in this phylogenetic group. Genotype 2, represented by C. neoformans var. neoformans, included serotype D and few isolates with serotypes AD and A (CBS 1584 (A), CBS 950 (AD), CBS 131 (AD), CBS 464 (AD), and CBS 7826 (AD). Both genotypes segregated in three distinct subgroups. Overall, genotype 1 subgroups were phylogenetically more distantly related from each other than genotype 2 sub clusters (Fig. 2). The other main phylogenetic group comprised all C. gattii isolates (serotype B and C) and consisted of four distinct clades represented by genotype 3 (serotype B), genotype 4 with sub-genotypes 4a, 4b and 4c (serotype B), genotype 5 (serotypes B and C) and genotype 6 (serotypes C). Genotype 6 is a newly discovered IGS genotypic group comprising two isolates, namely WM 779 and B 5742. All trees showed topological congruency, but they differed in the level of diversity among the three taxa. For example, the sequence diversity inferred from IGSI showed that C. neoformans var. grubii differed from C. neoformans var. neoformans and C. gattii by a maximum of 178 and 376 bp, respectively, whereas in the IGSII region, C. neoformans var. grubii differed by a maximum of 80 and 138 bp, respectively. Analysis of IGSI + 5S rRNA + IGSII showed that C. gattii diverged from var. neoformans by 548 and 506 bp from C. neoformans var. grubii. Considerable genetic diversity was also observed between the four genotypes of C. gattii. For instance, analysis from the IGSI + 5S rRNA + IGSII region showed that genotypes 4, 5 and 6 differed from genotype 3 by a maximum of 171, 173 and 164 bp, respectively, whereas genotype 4 and 5 differed from genotype 6 by over 125 and 127 bp.

Figure 2

Phylogenetic tree of C. neoformans species complex derived from IGSI + 5S rRNAt + IGSII sequence data. Presented is one of 100 most-parsimonious tree (length 948 bp; CI: 0.91; RI: 0.995) computed with PAUP*4 (heuristic search, stepwise addition, random addition sequence, nearest neighbor interchange, 100 maximum trees). Data consisted of 2350 characters (constant characters: 1582; uninformative variable characters: 46; informative characters: 722). Gaps were treated as missing data. Numbers indicate bootstrap values of 100 replicates.

A bootstrap confidence test, which is one of the most commonly used tests of reliability of an inferred tree, demonstrated that the six main clades were well supported in all trees (Fig. 2). Similar bootstrap values, ranging from 97 to 100, were obtained among the sub-genotypes of C. neoformans var. grubii (genotypes 1a, 1b and 1c) and genotype 4 of C. gattii (genotypes 4a, 4b and 4c) (Fig. 2). In contrast, lower bootstrap values, ranging from 71% to 86%, were observed among some of the sub-clades representing genotypes 2a, 2b and 2c of C. neoformans var. neoformans (Fig. 2).

3.5 Cloning analysis

To investigate the hybrid nature of some strains and the possible presence of multiple alleles of the IGS locus, amplicons from serotype AD strains, e.g., RV 61756 and BA4, and IGS genotype 3 isolates, e.g., CBS 1930 (serotype B) and IMH1658 (serotype B), were cloned. In addition, cloning analyses were undertaken with discordant serotype strains that did not follow the typical serotype boundaries of C. neoformans var. neoformans, C. neoformans var. grubii and C. gattii, e.g., CBS 464 (IGS 2c – serotype A), RV 52733 (IGS 1a – serotype D); CBS 6956 (IGS 3 – serotype B), CBS 6996 (IGS 5 – serotype B), CBS 5756 (IGS 1a – serotype A), CBS 7750 (IGS 3 – serotype B). The cloned sequences were compared with sequences derived from clones of non-hybrid origin, e.g., CBS 1143 (IGS 1a – serotype A), CBS 5474 (IGS 2c – serotype D), 48A (IGS 4c – serotype B) and CBS 6955 (IGS 5 – serotype C).

Cloning sequences derived by the primer set IGF and IG2R yielded a single band with amplicon sizes ranging from ∼910 (RV 52733, RV 61756, CBS 1143; CBS 464, CBS 5474) to ∼430 bp (CBS 6956, CBS 6996, CBS 1930, CBS 7750, IMH 1658, CBS 6955, 48A). All cloned sequences recovered from each strain were identical and displayed identical genotypic sequences of the archived strains.

4 Discussion

4.1 Molecular structural organization and characterization of the IGS region

The overall molecular structure of the rDNA IGS of C. neoformans species complex follows the same characteristic features as those reported in other fungi and plants [29]. As in most basidiomycetes, the intergenic spacers of C. neoformans species complex are separated by the 5S rRNA gene, whose transcription occurs in the same direction as that of subunits [30,31]. Both regions consist of several repeated motifs, indels and variable regions with nucleotide substitutions. Besides the common occurrence of point mutations and short indels involving few nucleotides, the prominent feature of IGSI is the presence of extensive indel areas, which were more frequent than those in the IGSII. The different indel lengths can result from different operating mechanisms. For example, short indels can be mostly explained by processes of DNA replication such as slipped strand miss-pairing, whereas long insertions and deletions might be explained by unequal crossing-over or by DNA transposition [32,33]. Although, the two regions showed a similar phylogenetic pattern, the IGSI region contained more phylogenetic informative characters and more sequence diversity than the IGSII. The degree of sequence heterogeneity in the IGSI region suggests that this region possibly functions as a hot spot for unequal crossing-over and miss-pairing events, which is one of the multiple functions attributed to intergenic spacers to maintain homogeneity of the rDNA repeats [34]. Even though, the IGSII region revealed several areas that can act as hot spots, this region is relatively more conserved and, therefore, can accommodate more functions related to rRNA production or processing [35].

Length variation in spacer size was found between the different IGS genotypes. Most of the observed length polymorphisms can be attributed to the presence of indels associated with repeats. The incidence of the commonly occurring GTTTTT and TATA motifs are examples of repeats that contributed to the presence of length polymorphisms between the genotypes. Also, inserts unrelated to repeat motifs are responsible for the differential size variation seen within members of the same genotypic group. The presence of repeats has also been reported in fungi like Pleurotus cornucopiae and Clavispora opuntiae[36,37].

The intergenic spacer region contains sequences that are essential for the initiation of transcription, RNA processing, transcription termination and replication processes of ribosomal DNA [35,38]. The repeat motifs, which are among the prominent features of the IGS region, are found in the transcribed region or upstream of transcription start [35,39,40]. They function as promoters, enhancers and regulators of transcription of rRNA, and possibly have originated from processes involving duplication and amplification of short sequences and from slippage replication mechanisms [40,41]. Many of the repeat motifs have been described as highly conserved in different species. For example, the repeat CAAAAA, which is a shorter version of our motif CAAAAAAT, has been described as a conservative motif in the promoter region of crucifers such as Brassica spp. [42], Raphanus spp. [43] and Arabidopsis[44]. Although the role of these repeats is yet to be determined, some functional significance related to transcriptional regulation is suspected. Other common repeat motifs, e.g., TATA, which are believed to be involved in the assembly of the pre-initiation complex and selection of the transcription site [45], were present in all six genotypes and have been reported to occur as a common element in various fungi such as Schizophyllum commune[46], Laccaria bicolor[47] and Neurospora species [48].

4.2 Phylogeny, genetic diversity and geographic substructure

Our phylogenetic analyses expanded on our earlier findings, which were based on partial sequence analysis of the IGSI region [19]. In addition, we included a new IGS genotypic group (IGS genotype 6) based on unique sequences and a discrete clustering pattern of two serotype-C isolates originating from India and South Africa. These isolates have recently been reported to belong to a new PCR-fingerprinting molecular type VGIV, and AFLP type 7 [12,49]. Isolates from this molecular type have also been found in Colombia, Mexico and Canada [12,16]. Based on our phylogenetic analyses which were supported by high bootstrap values, the C. neoformans species complex consists of six major IGS types. Besides these genotypic groups, our sequence analyses of IGSI + 5S rRNA + IGSII revealed the presence of additional sub-genotypes, e.g., genotypes 4a, 4b and 4c (Fig. 2). As in our previous study [19], our IGS data were not fully concordant with the traditional classification based on serotypes, which described serotype A isolates as C. neoformans var. grubii[15] and serotype D as C. neoformans var. neoformans[1].

The level of divergence between the genotypes increased when both IGS regions were included in the phylogenetic analysis. Overall, our IGS genetic lineages are in agreement with the genotypic classification as inferred by AFLP [17], M13/URA 5 analysis [16], PCR-fingerprinting/RAPD analysis [50], and ITS/RAPD profiles [49]. Boekhout et al. [17] have described six molecular AFLP types, which correlate with our current classification. The equivalent types are: IGS genotype 1 = AFLP 1 + AFLP 1A; IGS genotype 2 = AFLP 2; IGS genotype 3 = AFLP 6; IGS genotype 4 = AFLP 4; IGS genotype 5 = AFLP 5 and IGS genotype 6 = AFLP 7 [12]. AFLP analysis recognized the presence of an additional phylogenetic lineage (i.e., AFLP 3), comprising isolates of hybrid origin between serotype A and D [17]. In contrast, our phylogenetic analysis placed these isolates (i.e., CBS 131, CBS 132, CBS 464, CBS 939, CBS 950, RV 52733, RV 52755, BA1, BA3, BA4 and RV 53794) as members of IGS genotypes 1 or 2 (Fig. 2). Even though, our IGS genotypic classification correlated with major AFLP groups, differences in subgroup classification were evident. For example, AFLP analysis failed to discern the IGS sub-genotypes 1b and 1c. Similarly, AFLP type 2 did not segregate into subtypes, which contrasts with our findings on the IGS genotype 2. In addition, isolates comprising the molecular subtypes AFLP 4B and AFLP 4A clustered randomly in our IGS genotypes 4a, 4b and 4c. This apparent lack of correlation among the subtype-related phylogenetic patterns may be attributed to inherent differences in the nature of both techniques. Based on sequence analysis of the ITS region, seven genotypes have also been described within the species complex. Via combinations of eight bp differences at various locations in the ITS1 and ITS2 regions, Katsu et al. [49] have described a genotyping technique that agreed with our IGS classification. Other fingerprinting techniques, such as PCR fingerprint using minisatellite (M13), URA 5-RFLP typing and PCR–RFLP of the PLB1 genes, grouped the varieties of C. neoformans into eight major molecular types [16,51]. However, if we include our IGS subtypes as independent lineages, our phylogeny could delineate 12 molecular subtypes. The further discrimination into 12 molecular sub-genotypes was possible since the IGS region is a fast-evolving region, which portrays the highest amount of sequence variation within the rDNA [25]. The observed degree of genetic heterogeneity among the IGS genotypes and sub-genotypes can be a valuable tool to understand the global epidemiology of cryptococcosis.

The segregation of C. neoformans var. grubii and C. neoformans var. neoformans into two distinct clades has been confirmed by multiple gene analysis, i.e., IGS, ITS, laccase gene, mitochondrial large ribosomal-subunit RNA (mtLrRNA), topoisomerase (TOPI), Cap59, 26S LrRNA and URA5 gene [7,19,20,52,53]. Some of these studies have reported the presence of considerable sequence divergence between these varieties, namely up to 6% for the URA5 gene [7], 3.6% for the topoisomerase I gene [1] and 8% for a partial analysis of the IGSI region [19]. When both intergenic spacers (IGSII + 5S rRNA + IGSII) were combined, the sequence divergence increased to nearly 12%. Other studies based on fingerprinting-PCR techniques, i.e., UT-4p probe, CNRE1 probe and multilocus enzyme electrophoresis profile, confirmed the separation of the serotypes A and D [16,54,55]. In view of these significant differences, Franzot et al. [15], have proposed variety grubii for serotype A isolates. We consider it likely that IGS genotype 2 (=AFLP genotype 2) represents an individual species, but multilocus sequence typing data are needed to corroborate this hypothesis. The different phylogenetic placement of serotype-AD isolates, and the switching pattern of some isolates from serotype AD to serotype A or D, further complicates the current classification, as these isolates may represent intervarietal or even interspecies hybrids [17,19,52,56].

Lack of concordance between serotype and genotype delineation has also been documented for C. gattii. Our phylogenetic studies demonstrated that C. gattii showed more genetic diversity than C. neoformans. This was demonstrated by the presence of four phylogenetic lineages (i.e., genotypes 3–6), which were all supported by high bootstrap values. This genetic diversity among C. gattii isolates has also been documented by Latouche et al. [51], who reported five distinct lineages employing PCR–RFLP of the PLBI gene. The phylogenetic clustering based on IGSI + 5S rRNA + IGSII indicated that IGS genotype 3 is closer related to genotype 4 than to genotypes 5 and 6, whereas genotype 6 was phylogenetically closer related to genotypes 4 and 5 than to genotype 3. Genotype 6 differed from genotypes 3, 4 and 5 by ∼6%, 4.2%, 4.9% nucleotide substitutions, respectively. Based on rate of nucleotide substitutions in the ITS region, Katsu et al. [49] have shown that ITS type 6, which corresponds to IGS genotype 6, was phylogenetically closer related to ITS type 5 (=IGS genotype 5) and ITS type 4 (=IGS genotype 3).

Overall, the IGS of C. gattii differed by up to 28.4% and 30.1% from C. neoformans var. neoformans and C. neoformans var. grubii, respectively. In contrast, the URA5 gene revealed only 8% sequence divergence between C. gattii and the varieties of C. neoformans[57], and nearly 100% similarity has been reported for the other ribosomal genes (16SrDNA: two differences; 2SrDNA: ten differences; 5.8S and 5S rDNA: no differences). Our data support the previous recognition of the occurrence of at least two species within the C. neoformans species complex [14].

Sequence analysis of cloned PCR amplicons of the IGS did not indicate the presence of hybridization events, contrary to results obtained by PCR analysis employing specific primers for CNA1, CLA4 and GPA1 genes [58], PCR-fingerprinting with (GACA)4 primers [17,59,60], AFLP analysis [17,60], RAPD analysis of the PLB1 gene [51, and cloning analysis of the ITS region [49]. Our IGS cloning analysis revealed that the two serotype-AD strains RV61756 and BA4 appear to have a single allele or that there is a strong selection for a single locus. However, more samples need to be analyzed to elucidate the allelic structure of serotype-AD isolates at the IGS locus. Although serotype-AD strains are known to be diploids [17,49,51,58,59,61,62], there are many AD isolates reported as haploids [6265] and aneuploids [7]. For example, the existence of a single STE20 allele and one mating-type allele has been reported to occur in serotype-AD strains [63,64].

Hybridization events may also explain part of the genetic heterogeneity displayed within members of C. gattii isolates. For instance, Sugita et al. [52] have found that serotype B exhibited higher genetic divergence than serotype C isolates. Although the latter observation is supported by our and previous studies [17,19], the inclusion of IGS genotype 6 (serotype C) as a new phylogenetic group within C. gattii indicates that serotype C isolates are phylogenetically diverse. Contrary to previous suggestions [17], our cloning analysis with C. gattii isolates (IGS genotype 3 = AFLP 6) indicated that these strains are not heterozygotous diploids for the IGS locus. Similar findings based on cloning analysis of PLB1 fragments have been reported by Latouche et al. [51]. IGS genotype 3, which was previously thought to be restricted to tropical and subtropical America [17,19], has been reported in Brazil [60], Thailand, Australia [49], and Canada [12,13].

Overall, the apparent lack of geographic concordance with phylogeny suggests that C. neoformans has undergone recent global dispersal [20]. This recent dispersal model might explain why IGS sequences did not discriminate among isolates from different geographic locations.

This study provided further evidence on the genetic diversity among the C. neoformans species complex and further supports the recognition of at least two, and probably three species within the complex. The present study confirms the utility of the IGS region as a powerful tool for species or varietal identification. The divergences, which are mostly represented by nucleotide variations and by deletion/insertion areas, can serve as a platform to design specific probes for rapid and concise identification of the species, varieties and genotypes of C. neoformans and C. gattii, [66] which may be very useful for a better understanding of the causes of infection and the epidemiology of cryptococcosis.


Our sincere gratitude to B. Theelen and A. Statzell-Tallman for their assistance in DNA isolation and to Drs. M. Lazera, F. Coenjaerts, K.J. Kwon-Chung, S. Hamdan, J. Torres-Rodriguez, D. Swinne, W. Meyer, A. van Belkum, M.T. Barreto de Oliviera, and F. Frommer for providing strains. This work was funded by National Institutes of Health Grant 1-UO1 AI53879-01.


  • 1 Department of Biochemistry, Georgetown University Medical Center, 4000 Reservoir Rd NW, Washington, DC 20057, USA.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
  61. [61].
  62. [62].
  63. [63].
  64. [64].
  65. [65].
  66. [66].
View Abstract