Two distinct 18S rRNA secondary structures in Dipodascus (Hemiascomycetes)

Kumiko Ueda-Nishimura1 and Kozaburo Mikata1

Institute for Fermentation, Osaka, 17-85, Juso- honmachi 2-chome, Yodogawa-ku, Osaka 532-8686, Japan1

Author for correspondence: Kumiko Ueda-Nishimura. Tel: +81 6 6300 6555. Fax: +81 6 6300 6814. e-mail: kumiueda{at}mb.infoweb.ne.jp


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The nucleotide sequences of the 18S rRNA gene from ascomycetous yeast-like fungi in the genera Dipodascus, Galactomyces and Geotrichum were determined and the tested strains were separated into two groups by sequence length. In group 1, the length and secondary structure of 18S rRNA corresponded to those of typical eukaryotes. In group 2, the 18S rRNA gene sequences were about 150 nt shorter than those of most other eukaryotes and the predicted secondary structure lacked helices 10 and E21-5. Many substitutions and some deletions in group 2 18S rRNA gene were not only found in variable regions, but also in regions that are highly conserved among ascomycetes. Despite the considerable differences in 18S rRNA gene sequence and secondary structure between group 2 and other fungi, including group 1, phylogenetic analysis revealed that groups 1 and 2 are closely related. These findings suggest that a number of deletions occurred in the 18S rRNA of the common ancestor of group 2 strains.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains studied, their 18S rDNA sequence base and accession numbers

 
Keywords: 18S rRNA gene, 18S rRNA secondary structures, Dipodascus

The DDBJ accession numbers for the sequences reported in this paper are shown in Table 1.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Dipodascus and Galactomyces are ascomycetous yeast-like fungi. Their anamorph genus is Geotrichum. All species strongly assimilate ethanol (de Hoog et al., 1986 ). Dipodascus is characterized by asci which have persistent walls and rupture in the apical region, and by smooth-walled ascospores with mucilaginous sheaths (de Hoog et al., 1986 ). In Galactomyces, on the other hand, the asci do not open at the apex and ascospores are verrucose with median rims (Redhead & Malloch, 1977 ). Dipodascus and Galactomyces are distinguishable by their physiological characters. Domain D1/D2 26S rRNA gene sequence analysis has shown that species of Dipodascus and its anamorph Geotrichum separate into two clades, one of which includes species of Galactomyces (Kurtzman & Robnett, 1995 ).

Since ribosomes are an indispensable component of the protein synthesis apparatus and their structures are strictly conserved, the DNA component of the small subunit ribosome has proved to be an important and useful molecular clock for quantifying evolutionary relationships between organisms (Woese et al., 1990 ). Generally, the rate of base substitutions, deletions or insertions in various regions of the rRNA gene is not uniform; some areas are highly conserved and unchanged through millions of years, some are highly variable and others are semiconserved (Johnson et al., 1988 ). 18S rRNA of most eukaryotes adopts a common secondary structure, having 48 universal helices (De Rijk et al., 1992 ). They are present in all small subunit rRNAs from Archaea, Bacteria, plastids and Eukarya except in those of the Microspora (Microsporidia) and in those of organisms classified as Parabasalia (Trichomonas and relatives), where some of these helices are missing (Van de Peer et al., 1999 ).

In this paper we report two distinct secondary structures of Dipodascus 18S rRNA, one typical of most eukaryotes and an unusual structure missing some helices, including universal helix 10.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Isolation of genomic DNA.
The strains studied are listed in Table 1. The strains were grown in YM broth at 24 or 17 °C (Dipodascus albidus). Genomic DNA was extracted and purified by the procedure of Holm et al. (1986) .

Assimilation test.
Assimilation tests were performed by standard methods (van der Walt & Yarrow, 1984 ).

Viability.
Cultures were preserved at 5 °C on slanted YM agar with a cotton plug. After 6 months preservation, the cultures were transplanted into the new medium and incubated at a given temperature. Viability was evaluated by assessing growth.

18S rRNA gene sequencing.
18S rRNA gene sequences were determined with a Thermo Sequenase fluorescent labelled primer cycle sequencing kit with 7-deaza-dGTP (Amersham Pharmacia Biotech) following the supplier’s protocol. PCR for amplification of the 18S rRNA gene and cycle sequencing using PCR products were performed as described by Ueda-Nishimura & Mikata (1999) .

The 18S rRNA gene sequences determined in this study were deposited in DDBJ with the accession numbers listed in Table 1.

Prediction of 18S rRNA secondary structure.
18S rRNA secondary structure was predicted following the model of De Rijk et al. (1992) . GENETYX-MAC v. 8 (Software Development) was used to help predict stem structures in the area containing many deletions in the sequence.

Phylogenetic analysis.
Sequence data were manually aligned with various 18S rRNA sequences of representatives of related genera obtained from GenBank. Positions that could not be compared among all sequences, corresponding to unknown bases, deletions and insertions, and regions of ambiguity in the total alignment were removed before performing the phylogenetic analysis. A phylogenetic tree was constructed by Kimura’s two-parameter method (Kimura, 1980 ) and the neighbour-joining method (Saitou & Nei, 1987 ) using CLUSTAL V (Higgins et al., 1992 ). Bootstrap values (Felsenstein, 1985 ) were calculated from 1000 replicates.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
18S rRNA gene sequences and secondary structure of 18S rRNA
18S rRNA gene sequences of Dipodascus, Galactomyces and Geotrichum species were amplified in vitro by PCR and their nucleotide sequences were determined directly. Multiple strains of Dipodascus armillariae, Galactomyces citri-aurantii, Geotrichum candidum, Dipodascus ovetensis and Dipodascus capitatus were examined (Table 1) and sequence identity was found for all species except Gal. citri-aurantii, in which a single base substitution between IFO 10821 and IFO 10822 was discovered at nt 1698 (nucleotide positions according to Mankin et al., 1986 ). The sequence of Dipodascus ambrosiae IFO 10801 was found to be identical to those of D. ovetensis IFO 1201 and IFO 10827. These newly determined sequences were separated by length into two groups (Table 1). Group 1 sequences are 1773–1794 nt in length, corresponding to the expected length for most eukaryotes, whereas group 2 sequences are 1634–1636 nt in length, i.e. smaller than most other eukaryotes. Group 1 consists of six Dipodascus species, three Galactomyces species and four Geotrichum species. Group 2 consists of six Dipodascus species and two Geotrichum species.

The 18S rRNA secondary structure of group 1 strains corresponds to that of typical eukaryotes, while that of group 2 strains differs from the general ascomycetous model (De Rijk et al., 1992 ) in that it lacks portions of the V2 and V4 variable regions (area numbering according to De Rijk et al., 1992 ) (Fig. 1). The major deletions correspond to whole helices 10 (helix numbering according to De Rijk et al., 1992 ) in the V2 region and E21-5 in the V4 region, and the ends of helices E10-1 and 11 in the V2 region and helix E21-1 in the V4 region (Fig. 2a, b). Helix 10 is deleted completely (Fig. 2a). UUG in the loop of helix 11 is conserved, but other parts of helix 11 are not conserved (Fig. 2a).



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 1. Secondary structure model for eukaryotic 18S rRNAs. The 5' terminus is symbolized by a dot, the 3' terminus by an arrowhead. Areas of relatively conserved structure are indicated by bold lines, variable areas by thin lines. Deleted regions of group 2 strains compared with group 1 strains are indicated by dotted lines. This figure is modified from De Rijk et al. (1992) .

 


View larger version (50K):
[in this window]
[in a new window]
 
Fig. 2. Sequence deletions and substitutions in the variable regions of 18S rRNA among S. cerevisiae, D. albidus (group 1) and D. ovetensis (group 2): (a) V2 region, (b) V4 region, (c) V8 region. [n (and) n] indicates the upper and lower regions forming helix n. Helix numbering is according to De Rijk et al. (1992) . Underlining indicates nucleotides forming hydrogen bonds with the opposite side. Asterisks represent conserved bases in the three sequences. Nucleotide positions are according to Mankin et al. (1986) .

 
The 18S rRNA of group 2 strains has many small deletions. Most of these occur in the variable regions, but three minor deletions of 2–4 nt were found in the conserved structural region, corresponding to loops in helices (Fig. 1). Group 1 18S rRNA is more similar to Saccharomyces cerevisiae 18S rRNA than to that of group 2 organisms in some conserved regions around the deleted helices (Fig. 2). For example, Fig. 2(c) shows that the helix 43 sequence of D. albidus (group 1) is almost identical to the S. cerevisiae sequence, but not to the D. ovetensis sequence (group 2).

D-Xylose assimilation and viability
Group 1 strains assimilated D-xylose, but group 2 strains did not. D-Xylose assimilation was the only physiological character that differed between groups 1 and 2 (data not shown). Group 1 strains survived after 6 months on slanted YM agar medium at 5 °C, but again group 2 strains did not.

Phylogenetic analysis
Fig. 3. shows a phylogenetic tree calculated from published fungal 18S rRNA sequences. Basidiomycetous 18S rRNAs were chosen as the outgroup. Some regions, corresponding to positions 643–740 (V4), 1041–1070 (V5), 1336–1382 (V7) and 1680–1716 (V9), had too many gaps and substitutions for the sequences to be aligned. All alignment positions with a gap or an unknown residue and regions of ambiguous alignment were ignored for phylogenetic analysis. In this study 1389 positions were used for phylogenetic analysis. The tree shows that the tested strains form a monophyletic cluster supported at a bootstrap level of 95%. Group 1 species separate into three clusters. Species forming asci containing over 10 ascospores, i.e. D. albidus, Dipodascus geniculatus and Dipodascus australiensis, form subgroup 1a, supported at a bootstrap level of 100%. Galactomyces geotrichum, Galactomyces reessii, Gal. citri-aurantii and Geo. candidum, which is an anamorph of Gal. geotrichum, form subgroup 1b at a 100% bootstrap confidence level. The remaining species of group 1 form subgroup 1c. The Galactomyces strains are closely related to each other and form a monophyletic subgroup, 1b, making Dipodascus a paraphyletic taxon.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 3. Phylogenetic tree showing the relationships between species of the Dipodascus complex and related taxa. The tree is based on 18S rRNA gene sequence data and was constructed by using the neighbour-joining method. Bootstrap values were calculated from 1000 trees and values of <70% were omitted. GenBank, EMBL and DDBJ accession numbers are shown in parentheses. Bar, sequence dissimilarity of 5%.

 
Despite dozens of base substitutions between Gal. geotrichum and Geo. candidum, they map to the same position on the phylogenetic tree (Fig. 3), as do Gal. reessii and Gal. citri-aurantii. However, the base substitutions are located in variable and deleted regions which were not used for the analysis because of ambiguous alignment. Accordingly, some distinctions between sequences could not be detected.

All species of group 2 are closely related to each other with a 100% bootstrap value and the stem of group 2 is very long. D. ambrosiae and D. ovetensis have identical 18S rRNA gene sequences. Only one base substitution was found between Dipodascus spicifer and Geotrichum clavatum, and four between Dipodascus magnusii and Geotrichum fragrans. Three closely related species, D. spicifer, D. capitatus and Geo. clavatum, are thermotolerant, being able to grow at 40 °C (de Hoog et al., 1986 ).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
de Hoog et al. (1986) reported that D. spicifer CBS 244.85T could assimilate D-xylose and that D. spicifer and Geo. clavatum are similar physiologically but distinguished by D-xylose assimilation. However, in this study D. spicifer IFO 10809T (=CBS 244.85T) (group 2) could not assimilate D-xylose and D. spicifer and Geo. clavatum could not be distinguished by D-xylose assimilation. D. spicifer and Geo. clavatum, whose 18S rRNA gene sequences are similar except for one substitution, have been suggested to be conspecific because of the similarity of their domain D1/D2 26S rRNA gene sequences (Kurtzman & Robnett, 1998 ). Therefore, it is suspected that the ability of D. spicifer to assimilate D-xylose as reported by de Hoog et al. (1986) is incorrect.

It was reported that D. ambrosiae and D. ovetensis are superficially similar and distinguished by ascus size and G+C content (de Hoog et al., 1986 ). However, they have been suggested to be conspecific because of the identity of their domain D1/D2 26S rRNA (Kurtzman & Robnett, 1998 ) and 18S rRNA gene sequences.

Between Geo. candidum (IFO 4599 and IFO 5959) and Gal. geotrichum (IFO 9541T), there are 22 substitutions and one deletion in their 18S rRNA genes. Considering that Geo. candidum is an anamorph of Gal. geotrichum this difference is huge because the difference between the sequence of multiple strains of identical species, i.e. D. armillariae, Gal. citri-aurantii, D. ovetensis or D. capitatus, is usually only one substitution or no substitutions at all. Therefore, it is suggested that IFO 4599 and IFO 5959 should not be treated as Geo. candidum.

The predicted 18S rRNA secondary structure of group 2 species lacks helices 10 and E21-5. Helix 10 is one of the universal helices present in all hitherto known small subunit rRNAs from Archaea, Bacteria, plastids and Eukarya, but not the Microspora, which lack helices 10, 11 and 44 (De Rijk et al., 1992 ). De Rijk et al. (1992) showed that 18S rRNA secondary structures of Schizosaccharomyces pombe and Yarrowia lipolytica were different from those of all other fungi. The 18S rRNAs of these species have an insert in helix E43-1, which is not found in other fungi, and Y. lipolytica 18S rRNA lacks helix E21-5. In this light, the lack of helices 10 and E21-5 in 18S rRNA of group 2 species shows them to be very different from other fungi.

The conserved regions are generally highly conserved among related organisms. Accordingly, it is unusual that many substitutions and some deletions in group 2 species are found not only in variable regions but also in regions that are conserved between ascomycetes. Fig. 2 shows that D. albidus (group 1) is more similar to S. cerevisiae than to D. ovetensis (group 2) in some conserved regions around the deleted helices. The phylogenetic analysis, using 1389 reliable positions, shows that groups 1 and 2 are related (Fig. 3). Therefore, it is suggested that the observed substitutions and deletions in the conserved regions of group 2 species happened after the branching of the Dipodascus/Galactomyces clade.

The grouping of species into group 1 (subgroups 1a, 1b and 1c except Geotrichum fermentans) and group 2 species is observed in the domain D1/D2 26S rRNA parsimony tree (Kurtzman & Robnett, 1995 ). Group 1 sequences of domain D1/D2 26S rRNA sequences from GenBank are 545–554 nt in length, corresponding to that of most ascomycetous yeasts (Kurtzman & Robnett, 1998 ), whereas group 2 sequences are 404–407 nt in length, i.e. smaller than those of most other ascomycetous yeasts. We suggest that there might be as many deletions in group 2 whole 26S rRNA as in 18S rRNA.

Group 1, as shown by Kurtzman & Robnett (1995) , is monophyletic. However, subgroup 1a and group 2, as shown in Fig. 3, are joined. The branch leading to group 2 is much longer than for all other strains examined and the long branch indicates that the rate of evolutionary change is increased by substitutions in conserved regions. The branch of subgroup 1a is also relatively long. The joining of group 2 and subgroup 1a was therefore possibly due to a long-branch attraction (Felsenstein, 1978 ).

One possible explanation for the increased rate of substitutions is that a number of deletions may have occurred in the 18S rRNA gene of the common ancestor of group 2 strains by chromosomal rearrangement and homologous recombination. For 18S rRNA to interact with other rRNAs and ribosomal proteins for construction of the ribosome, the change in three-dimensional structure might have required many base substitutions in the general conserved regions. In addition to deletions in the 18S rRNA gene, it is expected that the length of the 26S rRNA gene of group 2 species might be shorter than that of group 1 species. Furthermore, group 2 species might have lost the ability to assimilate D-xylose and the ability of group 2 strains to survive on slanted agar medium was inferior to that of group 1. From these characteristics, it is suggested that the common ancestor of group 2 lost parts of its rDNA operon and the genes related to D-xylose assimilation and survival.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
De Rijk, P., Neefs, J.-M., Van de Peer, Y. & De Wachter, R. (1992). Compilation of small ribosomal subunit RNA sequences.Nucleic Acids Res 20, 2075-2089.[Medline]

Felsenstein, J. (1978). Cases in which parsimony or compatibility methods will be positively misleading.Syst Zool 27, 401-410.

Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using the bootstrap.Evolution 39, 783-791.

Higgins, D. G., Bleasby, A. J. & Fuchs, R. (1992). CLUSTAL V: improved software for multiple sequence alignment.Comput Appl Biosci 8, 189-191.[Abstract]

Holm, C., Meeks-Wagner, D. W., Fangman, W. L. & Botstein, D. (1986). A rapid, efficient method for isolating DNA from yeast.Gene 42, 169-173.[Medline]

de Hoog, G. S., Smith, M. Th. & Guého, E. (1986). A revision of the genus Geotrichum and its teleomorphs.Stud Mycol 29, 1-131.

Johnson, A. M., Illana, S., Hakendorf, P. & Baverstock, P. R. (1988). Phylogenetic relationships of the apicomplexan protist Sarcocystis as determined by small subunit ribosomal RNA comparison.J Parasitol 74, 847-860.[Medline]

Kimura, M. (1980). A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences.J Mol Evol 16, 111-120.[Medline]

Kurtzman, C. P. & Robnett, C. J. (1995). Molecular relationships among hyphal ascomycetous yeasts and yeastlike taxa. Can J Bot 73 (Suppl. 1), S824–S830.

Kurtzman, C. P. & Robnett, C. J. (1998). Identification and phylogeny of ascomycetous yeasts from analysis of nuclear large subunit (26S) ribosomal DNA partial sequences.Antonie Leeuwenhoek 73, 331-371.

Mankin, A. S., Skryabin, K. G. & Rubtsov, P. M. (1986). Identification of ten additional nucleotides in the primary structure of yeast 18S rRNA. Gene 44, 143-145.[Medline]

Redhead, S. A. & Malloch, D. W. (1977). The Endomycetaceae: new concepts, new taxa. Can JBot 55, 1701-1711.

Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees.Mol Biol Evol 4, 406-425.[Abstract]

Ueda-Nishimura, K. & Mikata, K. (1999). A new yeast genus, Tetrapisispora gen. nov.: Tetrapisispora iriomotensis sp. nov., Tetrapisispora nanseiensis sp. nov. and Tetrapisispora arboricola sp. nov., from the Nansei Islands, and reclassification of Kluyveromyces phaffii (van der Walt) van der Walt as Tetrapisispora phaffii comb. nov.Int J Syst Bacteriol 49, 1915-1924.[Abstract]

Van de Peer, Y., Robbrecht, E., de Hoog, S., Cares, A., De Rijk, P. & De Wachter, R. (1999). Database on the structure of small subunit ribosomal RNA sequences.Nucleic Acids Res 27, 179-183.[Abstract/Free Full Text]

van der Walt, J. P. & Yarrow, D. (1984). Methods for isolation, maintenance, classification and identification of yeasts. In The Yeasts, a Taxonomic Study, pp. 76-97. Edited by N. J. W. Kreger-van Rij. Amsterdam: Elsevier.

Woese, C. R., Kandler, O. & Wheelis, M. L. (1990). Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya.Proc Natl Acad Sci USA 87, 4576-4579.[Abstract]

Received 13 September 1999; revised 29 November 2000; accepted 17 February 2000.