A reorganized Candida albicans DNA sequence promoting homologous non-integrative genetic transformation

Molecular Microbiology (1992) 6(23), 3567-3574

A reorganized Candida albicans DNA sequence promoting homologous non-integrative genetic transformation E. Herreros,^ M. I. Garcia-Sdez, C. Nombela and M. Sanchez* Departamento de fvticrobiologi'a II, Facultad de Farmacia, Universidad Compiutense, 2840 fvladrid, Spain. Summary In order to develop plasmids adequate for non-integrative genetic transformation of Candida albicans, a DNA fragment of 15.3 kb was cloned from this organism on the basis of its capacity to convert the integrative Saccharomyces cerevisiae vector Ylp5 into a non-integrative one. Southern hybridization analysis, carried out with a labelled DNA probe of 3.6 kb derived from the cloned fragment, showed that it consisted of C. albicans DNA, the hybridization pattern indicating that the corresponding sequences were homologous to several chromosomal regions. The size of the C. albicans DNA promoting autonomous replication In S. cerevisiae was substantially reduced by subcioning. A 5.1 kb subfragment, defined by SamHI and Sa/I restriction sites, retained autonomous replication sequences {ARS) functional in the heterologous S. cerevisiae system and in C. albicans, when inserted in piasmid constructions that carried a S. cerevisiae trichodermin-resistance gene {tcmi) as selection marker. C. albicans transformants were both of the integrative and the non-integrative type and the plasmids recovered from the latter very often carried a reorganized ARS, indicating that recombination of the inserted ARS DNA had occurred in the homologous host. Successive reorganizations of the ARS insert in C. albicans eventually led to a more stabie and much smaller fragment of 687 bp that was subsequently recovered unchanged from transformants. Sequence analysis of the 687 bp fragment revealed four 11-base blocks, rich in A+T, that carried the essentiai consensus sequence considered relevant for yeast ARS elements in addition to other features also described as characteristic of yeast replication origins. Received 29 April, 1992; revised and accepted 10 August, 1992. tPresent address: Centro de Investigaci6n Glaxo, Parque Tecnol6gico de Madrid, 28760 Tres Cantos, Madrid, Spain. *For correspondence. Tel. (1) 394 1744; Fax (1)394 1745.

Introduction Cloning of Candida albicans genes has been carried out in bacteria and in the yeast Saccharomyes cerevisiae (Kurtz et al., 1988). Integrative transformation (Kurtz et al.. 1986) and gene disruption in C. albicans have also been achieved in a limited number cases (Kelly et al., 1988; Kurtz and Marrinan, 1989), with the use of the cloned genes. However, C. albicans has not yet been developed as a host system to be used for all possibilities of gene cloning based on reliable non-integrative genetic transformation of this yeast. The cloning of a C. aibicans DNA fragment capable of directing autonomous replication (ARS) in this yeast, but not in S. cerevisiae. was reported by Kurtz et ai. (1987) and opened the way for developing C. aibicans as a host-vector system. However, plasmids based on this /iflS fragment seem to replicate in C. albicans in the form of multimers of different sizes, which makes it difficult to isolate them by Escherichia coii transformation. In a recent report, Goshorn et ai. (1992) described the use of the Candida ARS DNA isolated by Kurtz et ai. (1987) to develop a shuttle vector, that included some sequences of the 2 (im plasmid, and was capable of replicating in C. aibicans, S. cerevisiae and £. coli. As expected, this plasmid replicated in Candida as mixed multimers, of different sizes, that were resolved into the corresponding monomers by transformation into S. cerevisiae. With the use of this vector, Goshorn et ai. (1992) have succeeded in cloning three nutritional Candida genes by complementing mutations in this yeast. In the last few years our efforts have also been directed towards the development of Candida transforming plasmids and to analysis of the possibility of using positive selection markers, which are needed for full application of recombinant DNA methodology to the study of the biology and pathogenicity of this yeast. We are especially interested in using new experimental strategies for the identification of genetic elements controlling the differentiation of yeast cells to mycelium (Gil et ai.. 1990) and for the clarification of the role of this phenomenon in pathogenicity and characterization of virulence factors. The use of simple vectors and positive selection markers is important with regard to achieving the isolation of genes by complementation of any type of relevant mutations in C. aibicans.

3568

E. Herreros, M. I. Garcfa-Saez, C. Nombela and M. Sanchez Vector

C.albicans DNA insert

Plasmid

Transformation freouencies S. cerevisiae

pCM2

B *•

H

H

S

H

S

pEH1

pEH2

pEH3

pEH6

pEH5

C.albicans

S

B m H S

pEH7

10'

Ylp5

10'

Ylp5

2-5 X 10'

5x 10'

Ylp5

2-5 X 10'

5x 10'

Yip352

10'

5x 10'

pEMBL18+tsi!ll

0

3X 10'

pEMBL18 + t £ m l

5 x 10'

Sx 10'

pEMBL18 + t s m l

5 X 10'

10'

Fig. 1. Non-integrative transforming plasmids and transformation frequencies of S. cerevisiae and C. aibicans, derived from yeast integrative-vectors (Ylp5 and Yip352) or from pEMBL18 with an S. cerevisiae trichodermin-resistance gene (tcm1), by insertion of DNA cloned from C. aibicans, and transformation frequencies for S. cerew's/ae TD-28 and C. aibicans.

While our work was in progress, the report of Cannon et al. (1990) described the isolation of a C. aibicans ARS DNA fragment that promoted non-integrative transformation of this yeast and S. cerevisiae; although some of the transformants were of the integrative type, this piece of DNA can be used for the construction of non-integrative vectors that can be recovered from transformants by E. CO//transformation. In this communication, we describe the development of non-integrative plasmids that can also transform C. aibicans and S. cerevisiae and can be recovered unaltered from the corresponding transformants. These plasmids are based on a small /iflS-containing DNA fragment from C. aibicans that was derived from a much larger one, through recombination in the host cell. We also used these plasmids with a positive selection marker, namely a gene that provides resistance to the inhibitor of protein synthesis, trichodermin.

Results and Discussion Non-integrative genetic transformation promoted by C. aibicans DNA S. cerevisiae has often been used as a system for cloning genes and other types of DNA fragments from other organisms, based on the activity of the cloned DNA In the heterologous yeast system. This has been the experimental strategy followed to select for autonomous replication sequences {ARS) from various origins such as fungi, plants and mammals, on the basis of their capacity to

convert an integrative vector of S. cerevisiae into a nonintegrative one (Stinchcomb etai., 1980; Hsu etai., 1983; Roth etai., 1983; Montiel etai., 1984). The present study was undertaken in order to isolate a C. aibicans DNA fragment that could be used to develop a plasmid adequate for non-integrative genetic transformation of this yeast. Figure 1 illustrates the restriction map of several DNA fragments, isolated from C aibicans, that were capable of promoting genetic transformation of S. cerevisiae at a high frequency, when inserted in integrative vectors. A 15.3 kb fragment was initially cloned from C. aibicans 1394, on the basis of Its capacity to convert the integrative S. cerevisiae vector Ylp5 into a non-integrative one. The cloning of this DNA piece was started by isolating genomic DNA from strain 1394 which was partially digested with restriction endonuclease Mbo\, ligated to a preparation of Ylp5 DNA digested with BamHl, amplified in £. coli and used to transform S. cerevisiae TD-28, a strain that carries a wide deletion of the URA3 locus. Twelve Ura* transformant clones were obtained and their DNA extracted and used to transform E. coli for plasmid isolation in order to verify their transforming capacity. The most efficient one was plasmid pCM2 (Fig. 1), carrying the 15.1 kb C. aibicans DNA, which could be substantially reduced in size without loss of the capacity to transform S. cerevisiae, when inserted not only in Ylp5 (the integrative plasmid used to clone it) but also in another integrative plasmid such as Ylp352 (Fig. 1). S. cerevisiae transformants obtained with these plasmid constructions displayed mitotic instability which Is

Candida aibicans DNA promoting non-integrative transformation 1 2

3

4

5

8

kb

smaller genomic DNA fragments that were presumably non-contiguous, a possibility which could explain the hybridization pattern obtained (Fig. 2).

- 23.6 -

9.4

- E.E - 4.3

II PH i I i

3569

ASXB

I III

P I

« I

H

S

Probe HKb

Fig. 2. Southern hybridization analysis of ARS DNA cioned from C. aibicans. The cioned fragment was mapped and the subfragment indicated by an arrow, defined by BamHi and H/ndiii restriction sites, was iabeiied and hybridized to C. aibicans DNA digested with SamHI (1) and H/ndlii (2), SamHi-digested DNA from S. cerevisiae (3) and E. coii (4), and the purified SamHI-H/ndlli DNA fragment used as a probe (5). M, Mbol; P, Pst\; H, H/ndlli; A, Aval, S, Sa/I; X, Xho\] B, BamHi.

characteristic of non-integrative transformation (Zakian and Kupfer, 1982); i.e. approximately 70% of the transformant cell population reverted to uracil auxotrophy after 20 h growing in non-selective complete medium. The plasmids were easily recovered from yeast transformants by E coli transformation, indicating that they were maintained as non-integrative plasmids. It follows that the cloned DNA must carry autonomous replicating sequences {ARS), and, in order to analyse this DNA by Southern hybridization, a 3.6 kb subfragment, derived from it and defined by SamHI and H/ndlll-restriction sites, was used as a labelled probe. The results in Fig. 2 clearly show that the cloned DNA hybridized with DNA from C. aibicans, whereas it did not hybridize with E. coii or S. cerevisiae DNA. The pattern of hybridization showed homoiogy with multiple fragments of different sizes, suggesting that the cloned sequences were represented in several of the chromosomal regions. The conditions used for /Wtool digestion of Candida genomic DNA in the initial cloning experiments were adjusted to generate fragments of 2-5 kb. However, the selected plasmid pCM2 carried an insert of 15 kb, with at least 10 /W/jol sites, indicating that this particular fragment resulted either from partial digestion or from the ligation of several

Deveiopment of Candida transforming plasmids with a dominant seiection marker The results described above clearly indicate that the cloned C. aibicans DNA was able to promote non-integrative genetic transformation of S. cerevisiae. In order to test for the same activity in C, aibicans, we decided to develop plasmid constructions with this DNA piece and with a putative dominant selection marker. We had access to a number of genes that convey resistance to several inhibitors such as kanamycin, hygromycin, G418, cycioheximide, etc. All these substances were tested but their minimal inhibitory concentrations (MICs) against C. aibicans were far beyond the reasonable level considered to be an adequate selection marker. However, trichodermin, a protein synthesis inhibitor, was reasonably effective; MICs of trichodermin against S. cerevisiae TD-28 and C. aibicans 1001 were determined to be 1.5 \ig mp' and 4 |ig ml"^ respectively. Plasmid constructions used for transformation of C. aibicans are depicted in Fig. 1. A trichodermin-resistance gene {tcm1), which formed part of plasmid pTCM (Fried and Warner, 1981) and was included in a 0.9 kb HpaW-HpaW segment, was inserted in the C/al site of plasmid pEMBL18; this construction was used to develop plasmids pEH5 and pEH6, by incorporating two subfragments derived from the originally cloned Candida DNA (Fig. 1). Both pEH5 and pEH6 transformed C. aibicans 1001 protoplasts; trichodermin-resistant (Tcm") transformants were selected at a frequency of 10^-10^ per ^lg of DNA, by regenerating protoplasts in the presence of 8 ng ml"^ trichodermin. However only pEH5 determined the transformation of S. cerevisiae, transformants being selected, in this case with 3 ^g ml"^ of the inhibitor, at the same frequency as in C. aibicans. Transformation of Candida usually gave rise to two types of colonies: the most abundant ones were small colonies which consisted of cells that displayed mitotic instability, with approximately 20% of the cells reverting to trichodermin sensitivity upon subculture in non-selective medium. The other type was transformants that formed large colonies consisting of cells that did not revert to sensitivity with significant frequency. Plasmid DNA was extracted from several of the Candida 'small colony' transformants and amplified by transformation of E coii. Restriction analysis of plasmids, recovered from many transformants obtained in different transformation experiments, clearly demonstrated that the ARS C. aibicans DNA insert was very often modified, probably because of recombination in the Candida host

3570

E Herreros, M. I. Garcia-Saez, C. Nombeia and M. Sanchez 1 2

3 4

pEH7

Fig. 3. Southern hybridization analysis of several C. aibicans transformants. A labelled probed consisting of pBR322 DNA was hybridized to DNA extracted from C. a/b/cans 1001 (1), a 'large colony' (2) and a 'small colony' (3) transformant of the same strain obtained with pEH7 (2), and plasmid pEH7 as a control (4). The exposure of lanes 1 and 2, during the autoradiographic procedure, was much iongerthan that of lanes 3 and 4, resulting in a difference in intensity of the bands.

that led to some kind of reorganization of the ARS fragment. Several of the modified plasmids were used for new transformations of the same strains and eventually a much more stable plasmid was obtained. This plasmid, pEH7, could transform C. aibicans reproducibly and could be recovered from many transformant clones without any apparent modification. The restriction map of pEH7, relative to that of pEH5, clearly showed that a substantial deletion had occurred in the ARS sequences of the latter, which had lost most of the intermediate sequences but retained the two restriction sites at both ends. The reduction in the size of the /4RS through the successive reorganizations was from 5.3 kb to only 0.7 kb. The recombinant ARS fragment retained the two restriction sites at both ends (Fig. 1). The generation of plasmid pEH7 by recombination of pEH5 and intermediate derivatives, in the host cell, improved the characteristics of this plasmid construction as a vector for Candida transformation; pEH7 transformed C. aibicans and S. cerevisiae with higher frequencies (10^-10" transformants per |xg of DNA) than pEH5, the appearance of large colonies among transformant clones being rare with this new plasmid. The results of hybridization experiments (Fig. 3) with DNA extracted from small and large colony transformants with a labelled probe of pBR322 DNA enabled us to interpret the significance of both types of transformants. pBR322 includes some of the sequences of pEMBL18 and therefore of

pEH7, DNA extracted from small colony transformants carried a hybridizing band of the size of the transforming plasmid pEH7, whereas large colony transformants carried a band of hybridizing DNA of a much greater size (Fig. 3). The differences in intensity of hybridization bands of Fig. 3 can be explained by the longer exposure of lanes 1 and 2 during the autoradiographic procedure. pEH7 transformed both C. aibicans and S. cerevisiae viWU similar frequencies; mitotic stability measurements, carried out as described by Murray and Cesareni (1986), showed that C. aibicans 'large colony' transformants reverted to trichodermin sensitivity at a frequency of 0,5% per generation, whereas reversion of C. aibicans 'small colony' and S. cerevisiae transformants was in the order of 10% per generation. It follows that in large colony transformants the transforming DNA must be integrated in the chromosome, whereas small colony transformants were of a non-integrative type consistent with their mitotic instability. The number of non-integrative vectors developed for S. cerevisiae is very high, making this yeast one of the most accessible cloning systems. However, these vectors are usually non-functional in other yeasts, thus the use of other yeast species as hosts for genetic transformation depends on the development of the specific vectors for the corresponding species. The isolation of a DNA fragment from any organism on the basis of ARS activity in S. cerevisiae does not neccessarily mean that the isolated DNA will promote non-integrative transformation in the organism of origin. Regarding C. aibicans, while our work was in progress, at least two reports documented data indicating that Candida DNA fragments, selected on the basis of >4RSactivity in S. cerevisiae, may (Cannon etai., 1990) or may not (Kurtz etai., 1987) carry sequences promoting non-integrative transformation in the cells of origin. It is clear from the evidence reported above that the DNA we cloned in S. cerevisiae promoted non-integrative transformation in C, aibicans. However, plasmid constructions carrying long fragments of the cloned DNA frequently recombined in the transformed host cells so that the plasmids recovered from transformants were modified judging from the restriction pattern of the inserted ARS fragment. The recombination process eventually reduced the size of the ARS DNA to approximately 0.7 kb, a reduction which is very convenient from the point of view of the size of the transforming vector. No further recombination was detected and vectors such as pEH7 were recovered from the corresponding transformants without any apparent changes. The reorganized ARS should therefore be ideal for developing non-integrative shuttle vectors suitable for cloning in C. aibicans. Our results also reveal that it is possible to use trichodermin resistance as a dominant selection marker for transformation of C. aibicans, gene tcm1 codes for a S.

Candida aibicans DNA promoting non-integrative transformation GGATCCAGAC TTGTTAAATC TGATGTrfTG ACGATTATCA TTGAATGTAA GATCAATAAA CCTAGGTCTG AACAATTTAG ACTACAAAAC TGCTAATAGT AACTTACATT CTAGTTATTT

61

CTATTCTCTT TCTTTTTATT TCAAAACCTT GCAGTACCAT GGAATTATAT CAAGCCATGT GATAAGAGAA AGAAAAATAA AGTTTTGGAA CGTCATGGTA CCTTAATATA GTTCGGTACA

121

^''TTAA^TGA TCAGTAGTCT AATCGGGCTT CAAGGGAAAA GGTGGTTTAC TCGCTTAACG AAATTTAACT AGTCATCAGA TTAGCCCGAA GTTCCCTTTT CCACCAAATG AGCGAATTGC

181

ATACTTTTGA CAil^TCTACT ACTTAGATTG TAAGAGAGCA TTGGTAGCAT CAAAAACACA TATGAAAACT GTAAAGATGA AGAATCTAAC ATTCTCTCGT AACCATCGTA GTTTTTGTGT

241

CAAAATATAA TCATTTTCCC AATTGGGAAC AATTCAAAAT AAAAACAACT TTGATGCGAA GTTTTATATT AGTAAAAGGG TTAACCCTTG T T ^ G T T ^ A TTTTTGTTGA AACTACGCTT

301

GAACTTCAAT ATTTTAGCTA ATTATATCAA AATTGGAATG GAGTCTCTAC TACTACATGC CTTGAAGTTA TAAAATCGAT TAATATAGTT TTAACCTTAC CTCAGAGATG ATGATGTACG

361

TTTTCTCAAT GAAATTACGT TAAAACTTTT CTCGTCTTCT TCCATTTTTT TTT_ATATTCT AAAAGAGTTA CTTTAATGCA ATTTTGAAAA GAGCAGAAGA AGGTAAAAAA [ A A A T A ^ U V G A

4 21

TTACTTTTTC TTTAATTTTT TTTTGTCATT CCTTATTTAA ATTl'dCAGAA TACCCTTTAA AATGAAAAAG AAATTAAAAA AAAACAGTAA GGAATAAATT TAAAGGTCTT ATGGG^AATT

481

ACTACATAAT ATCATGAAGA ATCCATTTGA CAGTGGCAGT GACGATGAAG ATCCATTTCT 3GATGTATTA TAGTACTTCT TAGGTAAACT GTCACCGTCA CTGCTACTTC TAGGTAAAGA

541

TAGTAATCCA CAATCTGCAC CATCAATGCC CTACGCAGCA TATACTAGTG ACAGAACATC ATCATTAGGT GTTAGACGTG GTAGTTACGG GATGCGTCGT ATATGATCAC TGTCTTGTAG

601

GCCCCGCAAG ACATACCAAC CATTGAATTT TGACAGTGAG GACGAAGATG CTAAAGAAAG CGGGGCGTTC TGTATGGTTG GTAACTTAAA ACTGTCACTC CTGCTTCTAC GATTTCTTTC

661

CGAATTTATG GCTTTCCCAC TGTCGAC GCTTAAATAC CGAAAGGGTG ACAGCTG

ARS2

ARS 3

X

3571

Fig. 4. Nucleotide sequence of the reorganized C. aibicans ARS DNA inserted in piasmid pEH7. The SamHI-Sa/i fragment was sequenced. The four biocks showing homoiogy with the major core consensus sequences, described in S. cerevisiae, are indicated as ARS 1 to 4. Framed are consensus eiements 5' with regard to the corg consensus, and indicated by arrows are consensus, eiements in 3'. These sequence data wiii appear in the EMBLVGenBank/DDBJ Nucieotide Sequence Data Libraries under the accession number X65035.

-^

AHS4

V

z

cerevisiae modified ribosomal protein, L3, of the 60S subunit (Schultz and Friesen, 1983) and it seems to be expressed in C. aibicans, thus leading to resistant transformant clones. The use of dominant selection markers represents a clear advantage from the point of view of the development of vectors, since it may enable strategies of genetic transformation that will not depend on the use of strains with specific markers. However, the very slow growth of transformants was a difficulty from the practical point of view. This slow growth was probably due to the limited expression of the tcm1 allele In C. aibicans under conditions of inhibition of protein synthesis, which might require a long time to build a sufficient level of modified ribosomal protein to overcome the toxic effect of the inhibitor. Studies are in progress in our laboratory designed to improve the efficiency of this type of selection by enhancing the expression of the resistance gene with an appropriate promoter,

Nucleotide sequence of the recombinant ARS fragment Sequencing of the recombinant ARS DNA 687 bp fragment revealed several features that have been considered characteristic of S. cerevisiae (Fig. 4). The 11nucleotide consensus sequence, 5'-(A/T)TTTAT(A/G) JJJ{/VT)-3' (underlined are the bases identified as being essential for the function), has been described as characteristic of S. cerevisiae ARS elements (Broach et al..

1983). Palzkill and Newlon (1988) reported that ARS function in S. cerevisiae requires a copy of the indicated core consensus sequence and additional near matches in the 3' flanking region. Four blocks homologous to this A+T-rich consensus sequence were apparent in our sequence at starting positions 19, 242,431 and 455; all of them carried at least the aforementioned five bases considered essential for promoting autonomous replication, judging from data obtained by site-directed mutagenesis (Van Houten and Newlon, 1990; Kipling and Kearsey, 1990). Of these four blocks, ARS2 and ARS4 (Fig. 4) are the ones most similar to the originally proposed S. cerevisiae consensus sequence. These four consensus blocks are highly homologous to sequences CaARSI (TTTTGTATTTT) and CaARS2 (TTTTATGTTTT), carried by the C. aibicans autonomous replicating DNA fragment cloned by Cannon etai. (1990). The regions flanking the core consensus sequence have also been considered relevant for ARS activity. For example, Marunouchi et al. (1987) proposed that a 5'TNT(A/G)AA-3' box, at the 5' end of the consensus sequence in the thymine rich strand, is also required for activity. We also found this sequence at positions -120 and -14 of the second and third core consensus sequence blocks, respectively. With regard to the 3' end, Palzkill et al. (1986) defined another block, with the sequence 5'-CTTTTAGC(A/T)(A/T)(A/T)-3', associated with ARS activity. We also found a block of this type at

3572

E. Herreros, M. I. Garcfa-Saez, C. Nombeia and M. Sanchez

position +47 of the second core consensus sequence ARS2 (Fig, 4), The ARS sequence described fits the two models that have been proposed to explain the requirements for autonomous replicating activity. However, the results shown still do not allow us to distinguish which specific requirements for /(flS activity in C, aibicans must be different from those of S, cerevisiae. This is the second DNA fragment to be described with ARS activity in both yeast species. A detailed comparison of our /\RS with the other one described by Cannon et al. (1990) revealed 44% overall homoiogy; in addition to the aforementioned consensus sequences we found several identical sequence blocks in both fragments, ranging from 4 to 18 bases. Further studies are in progress in our laboratory, in order to improve genetic transformation of C, aibicans by several means including the development of piasmids with both ARS, the one described in this study and the one cloned by Cannon etai. (1990).

Experimental procedures Vectors, strains, media and other materials The shuttle vectors Ylp5 (Struhl etai., 1979) and Ylp352 (Hill et ai., 1986) were the yeast integrative vectors used. Plasmids pEMBL18 and pEMBL19 (Dente et al., 1983) were used to clone in E. coli. Plasmids Bluescript-KS and -SK (Stratagene,) were used for sequencing, Plasnfiid pTCM (Fried and Warner, 1981), carrying tcm1 gene, was kindly supplied by A. Jimenez. Standard nnethods were used for plasmid preparation (Holmes and Quigley, 1981; Maniatis etai., 1982), Trichodermin was a gift from A. Godtfredsen, C. albioans 1394 (used for the preparation of genomic DNA) and 1001 (used as a host for genetic transformation), were wild-type strains from the Spanish Type Culture Collection (CECT), S, cerevisiae TD-28 {MATa, ura3-52, inost-151, cani) was kindly supplied by F, del Rey, The strains of E. ooli used in cloning procedures were DH5 (Hanahan, 1985), JM109 (Yanisch-Perron et al., 1985) and HB101 (Boyer and Roulland-Dussoix, 1969), YED ( 1 % yeast extract and 2% glucose) broth or agar and YEPD (YED plus 2% peptone) were the complete media used for growing yeast strains. Minimal medium (MM), broth or agar, used for yeast strains, contained 0,67% Yeast Nitrogen Base without amino acids and 2% glucose. To prepare yeast protoplast regeneration medium (RM), MM broth was supplemented with 1 M sorbitol, 0,4% of YEPD broth and 3% agar. When required, uracil or uridine was supplemented at a final concentration of 80 ng mr^, For growing bacteria, Luria-Bertani (LB) medium, containing 1 % bacto-tryptone, 0,5% yeast extract and 1 % NaCI, and SOB medium, consisting of 2% bacto-tryptone, 0,5% yeast extract, 10 mM NaCI, 2,5 mM KCI, 10 mM MgCIa and 10 mM MgS04, were used, Ampicillin was supplemented, when required, at a concentration of 100 ng mP^, Yeast genetic transformation Genetic transformation of S. oerevisiae, by the sphaeroplast procedure, was done according to Beggs (1978), Essentially

the same procedure, with some modifications, was followed for transformation of C, aibicans. Cells grown in YED, up to a concentration of 1-2 X10^ cells m r \ were collected by centrifugation, washed, resuspended in a suspension buffer, at pH 8, consisting of 1 M sorbitol, 25 mM EDTA and 50 mM dithiothreitol, and incubated for 10 min at 30°C with gentle shaking. Cells were again centrifuged and resuspended in 0.1 M sodium citrate buffer pH 5,8, with 1 M sorbitol and 25 mM EDTA, This suspension was introduced in an Erienmeyer flask with the enzyme complex p-glucuronidase (Sigma Chemical) to a final concentration of 1 % followed by an incubation at 30°C for 1 h with gentle shaking. The formation of sphaeroplasts was checked under the phase-contrast microscope by observing osmotic fragility upon addition of distilled water. The preparation was manipulated thereafter with precautions to avoid protoplast breakage. Protoplasts were sedimented by centrifugation at 1100 X gi for 5 min, washed with 1 M sorbitol and washed again with 10 mM Tris-HCI buffer pH 7,4 containing 1 M sorbitol and 10 mM CaClg, Sphaeroplasts were suspended in the same buffer, and divided Into 100-nl aliquots and the transforming DNA added, gently mixed and maintained at room temperature for 15 min. Each tube received 1 ml of 20% polyethylene-glycol (PEG) in 10 mM Tris-HCI buffer pH 7.4, containing 10 mM CaCl2, They were allowed to stand for another 15 min at room temperature and centrifuged for 5 min at 1100 x g. Sphaeroplasts were resuspended in 150 |xl of a solution containing 33% of YEPD medium, 1 M sorbitol and 10 mM CaCl2, and were incubated at 30°C for 24 h in order to favour regeneration and expression of trichodermin resistance. Trichodermin was added after this 24 h period to a final concentration of 8 |xg mP^ for selection of C, albioans transformants and 3 ng mP^ for S, oerevisiae. The sphaeroplast suspension was added to a regeneration medium, maintained at 47°C, which contained 2% glucose, 0,67% Yeast Nitrogen Base without amino acids, 0,4% YEPD medium, 1 M sorbitol and 3% agar. The suspension was gently mixed, poured into Petri dishes and cooled. Plates were finally incubated at 30°C until the appearance of transformant colonies embedded in the agar.

Cioning and sequencing Restriction analysis and cloning in E. ooli were carried out as described by Maniatis etai. (1982), Restriction endonucleases and T4 DNA ligase were from Boehringer Mannenheim, Bethesda Research Laboratories and New England Biolabs. Selected cloned fragments were inserted into Bluescript plasmids, amplified and both strands separated by alkaline denaturation and sequenced by the Sanger dideoxy sequencing technique (Sanger etai., 1977), using the sequencing kit supplied by Pharmacia LKB Biotechnology and appropriate oligonudeotide primers.

DNA anaiysis Yeast genomic DNA, for use in Southern blot analysis, was extracted as described by Sherman etai. (1986), The extracted DNA was digested with restriction enzymes and transferred to Pall BiodyneA nylon membranes and hybridized with a probe, labelled by nick translation as described by Rigby etai. (1977), with the use of the Amersham nick translation kit. After transfer

Candida albicans DNA promoting non-integrative transformation of the fragmented DNA to the nylon membranes, it was denatured and hybridized with the iabelled probe, by overnight incubation at 42°C with gentie shaking in the presence of 50% formamide. Membranes were finally washed: four 5-min washes at room temperature with strong shaking, in a solution containing 0.03 M sodium citrate, 0.3 M NaCi and 0.1% SDS, and two 15-min washes at 50°C, in 0.0015M sodium citrate, 0.015 NaCI and 0.1% SDS, were carried out. The pH of the washing solutions was adjusted to 7.0 with citric acid crystals.

Acknowledgements This work was supported by grants 89/266 and 90/565, awarded to C.N., from Fondo de Investigaciones Sanitarias. E.H. was a recipient of a fellowship from Ministerio de Educaci6n y Ciencia and M.I.G.-S. from FIS. We wish to thank Ana I. Santos for her involvement in C. aibicans work and for much heip and discussions during this investigation. We are indebted to Dr W. O. Godtfredsen (Leo Pharmaceutical Products, Balierup, Denmark) for repeated generous gifts of trichodermin.

References Beggs, J.D. (1978) Transformation of yeast by a replicating hybrid piasmid. Nature275:104-109. Boyer, H.W., and Rouiland-Dussoix, D. (1969) A complementation analysis of the restriction and modification of DNA in Escherichia coli. J Mol e/o/41:459-472. Broach, J.R., Li, Y.-Y., Feldman, J., Jayaram, M., Abraham, J., Nasmyth, K.A., and Hicks, J.B. (1983) Localization and sequence analysis of yeast origins of DNA replication. Cold Spring Harbor Symp Quantitative Bioi 47:1165-1173. Cannon, R.D., Jenkinson, H.F., and Shepherd, M.G. (1990) Isolation and nucleotide sequence of an autonomously replicating sequence (ARS) element functional in Candida aibicans and Saccharomyces cerevisiae. Mol Gen Genet 221: 210-218. Dente, L., Cesareni, G., and Cortese, R. (1983) pEMBL: a new family of single stranded plasmids. NucI Acids Res 1 1 : 1645-1655. Fried, H.M., and Warner, J.R. (1981) Cloning of a yeast gene for trichodermin resistance and ribosomal protein. Proc NatI Acad Sci USA 78: 238-242. Gil, C Pomes, R., and Nombeia, C. (1990) Isolation and characterization of Candida aibicans morphological mutants derepressed for the formation of filamentous hypha-type structures. J Bacterioi M2: 2384-2391. Goshorn, A.K., Grindie, S.M., and Scherer, S. (1992) Gene isolation by complementation in Candida aibicans and applications to physical and genetic mapping. Infect Immun 60: 876-886. Hanahan, D. (1985) Techniques for transformation of £. coli. In DNA Cloning. Vol. I. Glover D.M. (ed.). Oxford: IRL Press, p. 120. Hiil, J.E., Myers, A.M., Koerner, T.J., and Tzagoloff, A. (1986) yeasVEscherichia coii shuttle vectors with multiple unique restriction sites. Yeast2:163-167. Holmes, D.S. and Quigley, M. (1981) A rapid boiling method for

3573

the preparation of bacterial piasmids. Anai Biochem 114: 193-197. Hsu, W.H., Magee, P.T., Magee, B.B., and Reddy, C.A. (1983) Construction of a new yeast cloning vector containing autonomous repiication sequences from Candida utiiis. J Bacterioi 154:1033-1039. Keily, R., Miller, S.M., and Kurtz, M.B. (1988) One-step gene disruption by cotransformation to isoiate double auxotrophs in Candida aibicans. Moi Gen Genef 214: 24-31. Kipiing, D., and Kearsey, S.E. (1990) Reversion of autonomousiy repiicating sequence mutations in Saccharomyces cerevisiae: creation of a eukaryotic replication origin within procaryotic vector DNA. Moi Ceii Bioi 10: 265-272. Kurtz, M.B., and Marrinan, J. (1989) Isolation of Hem3 mutants from Candida aibicans, by sequential gene disruption. Moi Gen Genet2^7: 47-52. Kurtz, M.B., Cortelyou, M.W., and Kirsch, D.R. (1986) Integrative transformation of Candida albicans, using a cioned Candida ADE2 gene. Moi Cell Biol 6:142-149. Kurtz, M.B., Cortelyou, M.W., Milier, S.M., Lai, M., and Kirsch, D.R. (1987) Development of autonomously replicating piasmids for Candida aibicans. Moi Ceil Biol7:209-217. Kurtz, M.B., Kirsch, D.R. and Kelly, R. (1988) The moiecuiar genetics of Candida aibicans. Micro Sci5:58-63. Maniatis, T., Fritsch, E.F., and Sambrook, J. (1982) Moiecuiar doning: A Laboratory Manual. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Marunouchi, T., Matsumoto, Y., Hosoya, H., and Okabayashi, K. (1987) In addition to the ARS core, the ARS box is necessary for autonomously replicating sequences. Mol Gen Genet 206: 60-65. Montiei, J.F., Norbury, C.J., Tutite, M.F., Dobson, M.J., Milis, J.S., Kingsman, A.J., and Kingsman, S.M. (1984) Characterization of human chromosomal DNA sequences which replicate autonomously in Saccharomyces cerevisiae. Nuci Acids Res 12:1049-1068. Murray, J.A.H., and Cesareni, G. (1986) Functional anaiysis of the yeast plasmid partition locus STB". Embo J 5: 3891-3399. Palzkiil, T.G. and Newion, C S . (1988) A yeast replication origin consists of multiple copies of a smali conserved sequence. Cell 53:44^-A50. Palzkili, T.G., Oliver, S.G., and Newion, C.S. (1986) DNA sequence analysis of /IRSelements from chromosome III of Saccharomyces cerevisiae: idenfication of a new conserved sequence. Nuci Acids Res 14: 6247-6264. Rigby, P.W.J., Dieckmann, M.I., Rhodes, C , and Berg, P. (1977) Labelling DNA to high specific activity in vitro by nicktranslation with DNA polymerase I. Moi Bioi 113: 237-251. Roth, G.E., Blanton, H.M., Hager, L.J., and Zakian, V.A. (1983) Isolation and characterization of sequences from mouse chromosomal DNA with ARS function in yeast. Moi Ceil Bioi 3:1898-1908. Sanger, F., Nicklen, S., and Coolson, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc Nati Acad Sci USA 74: 546^-5467. Schultz, L.D., and Friesen, J.D. (1983) Nucleotide sequence of the tcm1 gene (ribosomal protein L3) of Saccharomyces cerevisiae. J Bacterioi ^55: 8-14. Sherman, F., Fink, G.R. and Hicks, J.B. (1986) Methods in Yeast Genetics. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. Stinchcomb, D.T., Thomas, M., Kelly, J., Selker, E., and Davis,

3574

E. Herreros, M. I. Garcia-Saez, C. Nombela and M. Sanchez

R.W. (1980) Eukaryotic DNA segments capable of autonomous replication in yeast. Proc Nati Acad Sci USA 77: 4559-4563. Struhi, K., Stinchcomb, D.T., Scherer, S., and Davis, R.W. (1979) High-Frequency transformation of yeast: autonomous replication of hybrid DNA molecules. Proc Nati AcadScUB: 1035-1039. Van Houten, J.V., and Newion, C.S. (1990) Mutationai analysis

of the consensus sequence of a replication origin from yeast chromosome III. Wo/Ce//S/o/10: 3917-3925. Yanisch-Perron, C , Vieira, J., and Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucieotide sequences of the M13mp18 and pUC vectors. Gene 33:109-119. Zakian, V.A., and Kupfer, D.M. (1982) Repiication and segregation of an unstable piasmid in yeast. Piasmid8:15-28.