Current Genetics (t984) 8 : 353-358
~
~
~
© Springer-Verlag 1984
Hygromycin B resistance as dominant selectable marker in yeast Kevin R. Kaster, Stanley G. Burgett, and Thomas D. lngolia Lilly Research Laboratories, 307 E. MeCarty Street, Indianapolis, Indiana 46285, USA
Summary. Saccharomyces cerevisiae is normally sensitive to the drug hygromycin B; a hygromycin B concentration of 200 /~g/ml in agar plates is sufficient to completely inhibit growth. We constructed yeast-E, coli bifunctional plasmids which confer hygromycin B resistance to Saccharomyces cerevisiae Promoters and amino terminal coding regions of a heat shock gene, a heat shock cognate gene, and the phosphoglycerate kinase gene from yeast were fused to a bacterial hygromycin B resistance gene. In all three cases, yeast cells containing plasmids with the hybrid hygromycin B resistance gene were resistant to high levels of the drug. Yeast cells containing these ptasmids can also be directly selected after transformation by using hygromycin B. The intact bacterial hygromycin B resistance gene and the kanamycin resistance gene from Tn903 were also tested in yeast for their ability to confer resistance to hygromycin B and G418. The intact bacterial genes were not effective in conferring drug resistance to yeast cells.
Key words: Hygromycin B - Yeast - Plasmids
Introduction Complementation of auxotrophic markers has been used in yeast transformations in order to select transformed cells. For example, the leu2 (I-finnen et al. 1978), ura3 (Struhl et al. 1979), his3 (Struhl 1982), his4 (Hinnen et al. 1979), and trpl (Hitzeman et al. 1980) genes from Saccaromyces cerevisiae have been used as selectable markers, and presumably any auxotrophic marker could be used. However, complementation of an auxotrophic marker does require that the recipient cell contain the
Offprint requests to: T. D. Ingolia
appropriate marker. This limitation can be a minor inconvenience, requiring only a genetic cross to introduce the appropriate marker into the desired background, or a major obstacle, for example, when one wants to introduce DNAinto another species in which the required genetic lesions are not known. A plasmid-bome dominant selectable marker would be a valuable experimental tool in these cases. We therefore examined several fungicidal drugs and the corresponding inactivating genes from prokaryotes for their efficacy as selectable marker in yeast. The aminoglycoside antibiotics G418 andhygromycin B were chosen because yeast cells are sensitive to G418 (Jimenez and Davies 1980), and hygromycin B (Gonzalez et al. 1978) and prokaryotic drag inactivation systems exist for each of these antibiotics. Both of these drugs disrupt ribosomal function and cause translational errors (Gonzalez et al. 1978; Davies and Smith 1978), and the drug resistance genes inactivate the drugs by covalent modification (Gortzalez et al. 1978; Davies and Smith 1978; Rao et al. 1983). Bacterial resistance to G418 can be conferred by several plasmid-encoded modifying enzymes (Davies and Smith 1978)including the aminoglycoside phosphotransferase encoded by Tn601 (903). In E. coli, a plasmid-bome gene encoding a hygromicin B phosphotransferase (Rao et al. 1983) confers resistance to hygromycin B. The DNA sequence of this gene has been determined and the protein coding region has been identified (Kaster et al. 1983). A nearly identical hygromycin B resistance gene has also been described by Gritz and Davies (Gritz and Davies 1983). Our previous results have shown that the first three amino acids of the hygromycin B phosphotransferase can be replaced by up to 15 amino acids from a foreign gene and still retain activity (Kaster et al. 1983). In this paper we report the construction and characterization of plasmids containing a truncated hygromycin
354 B resistance gene f u s e d t o y e a s t genes, a n d d e m o n s t r a t e t h e e f f e c t i v e n e s s o f t h e s e plasmids in c o n f e r r i n g h y g r o m y cin B resistance t o y e a s t cells.
Materials a n d m e t h o d s
Materials. G418 was provided by Dr. Peter Daniels and the Schering Corporation. Hygromycin B was obtained from Eli Lilly and Company. BamHI and EcoR1 linkers were provided by Dr. Rama Belagaje. Plasmids YG100PR1 and YG101HI were from Mike Slater and Elizabeth Craig. [Gamma-32P]-ATP was synthesized from 32P-phosphate and ADP (Johnson and Walseth 1979). Zymolyase 60,000 was obtained from Miles. All other biochemieals and enzymes were obtained from commercial sources. Enzyme reactions. Restriction enzyme digestion of DNA was performed according to suppliers' recommendation. Protruding ends of restricted DNA molecules were removed, when necessary, by incubating up to 10 ~g of DNA with about three units of DNA polymerase I large fragment in the presence of all four dNTPs at 300 /~M concentration. Double stranded regions of DNA were shortened by digestion with Bal31 nuclease. For removing long (greater than 100 base pairs) stretches, about 10 /~g of DNA were incubated at 37 °C with three units of Bal31 nuclease under conditions specified by the manufacturer. Atiquots were removed at different times after addition of enzyme, quenched with phenol, and analyzed on polyacrylamide gels to determine the extent of digestion. For removing shorter stretches of DNA, the units of enzyme per/~g of DNA were reduced and the reaction carried out at 4 °C. Ligation of DNA molecules, including synthetic oligonucleotides, was accomplished using standard conditions (Schliefand Wensink 1981). When necessary, sequences of regions of plasmids were checked by the chemical cleavage method of Maxam and Gilbert (Maxam and Gilbert 1977). Construction o f plasmids. The hygromycin resistance gene was obtained from pKC222 (Kaster et al. 1983). This plasmid was restricted with HphI and Pstl, and the fragment containing the amino-terminal coding region was purified. The 3' protruding end left by the HphI enzyme was removed with E. coli DNA polymerase I large fragment (Klenow fragment) and BamHI linkers (5'-TGGATCCA) ligated. The fragment was restricted with BamHI and EcoRI and ligated into the BamHI and EcoRI sites of pBR322. This intermediate plasmid, called piT122, was cut with EcoRI, and the hygromycin resistance gene was reconstructed by inserting the small EcoRI fragment from pKC222 into piT122. The resulting plasmid, called piT123, contains the hygromycin resistance gene with a BamHI linker substituted for the first three codons of the protein coding region. As shown in Fig. 1, the truncated hygromycin resistance gene on piT123 can be excised on a BamHI-BgllI fragment. The activator sequences from three yeast genes, a heat shock gene, a heat shock cognate gene, and the phosphoglycerate kinase gene, were modified to facilitate linkage to the hygromycin resistance gene. The heat shock gene is the yeast analogue to the major heat shock gene of Drosophila, hsp70, and is contained on pYG100 (Ingolia et al. 1982). Plasmid pYG100PR1, consisting of a PstI-EeoRI fragment from pYG100 containing the 5' end of the heat shock gene subcloned into pBR322, was used as the starting material, as shown in Fig. 2. This plasmid was digested with NeoI, which cuts within the coding information for amino
K.R. Kaster et al. : Hygromycin B resistance in yeast acid 110 of the heat shock protein, digested with Ba131 nuclease, and figated to EcoRI linkers. One isolate, called p i t 115, contained a linker inserted into the coding region for amino acid 5 of the heat shock protein. In order to place a BamHI site near the EcoRI site, the XbaI-BamHI fragment from p l T l l 5 was cloned into plNIIIAI (Nakamura and Inouye 1982) yielding piT116. In order to place a BglII site upstream of the activator sequence, the XbaI-BamHI fragment of p l T l l 6 was cloned into pBHI, a recombinant plasmid containing Drosophila DNA (Craig et al. 1979). The resulting plasmid, piT120, contains the activator sequence from the pYG100 heat shock gene on BglII-BamHI fragment. The sequence near the translation initiation site is also shown in Fig. 2. The reading frame at the BamHI site is the same as that used for the hygromycin resistance gene in piT123. The activator sequence from a yeast heat shock cognate gene was also modified. The gene was contained in pYG101H1, a HindlII subclone into pBR322 obtained from Mike Slater and Elizabeth Craig. A 680bp ClaI-EcoRI fragment containing the 5' end of the gene was cloned into Clal-EcoRI-cut pBR322 to yield piT107. This plasmid was cleaved with C/aI, which cuts at amino acid 10 of the coding region for the heat shock cognate gene, treated with Bal31 nuclease and ligated to BamHI linkers. One isolate, called p l T l l 0 , had a BamHI tinker inserted after the ninth codon of the protein coding region. Beginning with the translation initiation site, the DNA sequence of piT110 includes 5 '-ATG GCT GAA GGT GTT TTC CAA GGT GCT cgg gat cc - 3', where the linker is depicted in small case letters. In order to introduce a BgllI site upstream of the activator sequence, an EcoRI fragment from pYG100BH (Ingolla et al. 1982) was cloned into the EcoRI site of p l T l l 0 to yield plTll8. The activator sequence from the heat shock cognate gene can be liberated on a BgllI-BamHI fragments from piT118, and the reading frame at the BamHI site is the same as the reading frame at the BamHI site of the truncated resistance gene on piT123. The phosphoglycerate kinase gene activator sequence was also adapted for use in gene fusions. The gene was isolated from a bank of yeast genome fragments cloned into lambda phage provided by Dr. John Woolford (Woolford and Rosbash 1981). The probe was a pBR322 derivative containing the Saccharomyces cerevisiae SUF16 gene, provided by Richard Gaber and Michael Culbertson. The SUF16 gene resides within about 5 kb of the phosphoglycerate kinase gene (Gaber and Culbertson 1982), so many recombinant phages should carry both genes. One such phage was identified, and a HindIII fragment containing the phosphoglycerate kinase gene and flanking regions (Hitzeman et al. 1982; Dobson et al. 1982) was cloned into pBR322 to yield piT141. A 958 bp HinclI-ClaI fragment containing the 5' end of the phosphoglycerate kinase gene was purified from pIT141, and subsequently cut with MbolI. The 5'-extensions resulting from the MbolI cleavage were filled in with DNA polymerase I large fragment and 8-met BamHI linkers ligated to the molecules. A 240 bp fragment was purified on an acrylamide gel and ligated into BamHI-cut pUC8 (Vieira and Messing 1982). One isolate, called piT143, contained a BamHI fragment which comprised the 5' end of the phosphoglycerate kinase gene. The DNA sequence of piT143, beginning with the translation initiation site, was found to be 5'-Atg gat cc - 3 ' , where the tinker sequence is depicted in small case letters. The BamHI site of • pIT143 is in phase with theBamHI site of the truncated hygromycin resistance gene on pIT123. The backbone for our yeast-E, coli bifunctional vectors was YEp24 (Botstein et al. 1978). The kanamycin resistance gene from Tn601(903) was incorporated into YEp24 by purifying a 1.7 kb PvulI fragment from pNG614 (a derivative of pNG18; (Grindley and Joyce 1980) and ligating it into the sole Sinai
355
K. R. Kaster et al. : Hygromycin B resistance in yeast
piT 123
Table. 1. Plasmids tested for ability to confer hygromycin B resistance to yeast Plasmid
piT201 piT213 piT215 piT216 piT217 piT219
Activator sequence
Level of Hm resistance in yeast,/~g/ml
lambda pL promoter < 200 None < 200 yeast heat shock cognate gene 1000 yeast heat shock cognate gene, opposite orientation as in pIT215 < 200 yeast heat shock gene 500 yeast phosphoglycerate kinase gene >1000
~ Pst
B
1/%
Eco RI I Hind III g l II ~81bp
///~
~NNkN~Hinc I I
//+ I /,
I' "
_
== " , = . , .
. . . . ='
I I-- Psi I ) I--, 90bp / / Eco RI 240bp ' A B a m HI
The construction and structure of the plasmids is described in materials and methods and Fig. 1-3. Only piT201 will confer hygromycin resistance to E. coli
site of YEp24. The truncated hygromycin gene was also added to YEp24 by ligating a BamHI-BgllI fragment from piT123 into the sole BamHI site of YEp24. The resultant plasmid, called pIT213, is shown in Fig. 3. The activator seuqences described above were then ligated into the sole BamH1 site on piT213 to yield the plasmids described in the Table 1.
Fig. 1. Truncated hygromycin resistance gene on piT123. This plasmid contains the truncated hygromycin B resistance gene. The sequence at the BamHI site, showing the reading frame utilized for the hygromycin B resistance gene, is also shown
Transformations. Transformation of DNA into E. coli was accomplished using the standard CaC12 plus heat shock procedure (Schlief and Wensink 1981). The drug concentrations used for selection in E. coli were: ampicillin, 50 ~g/ml; kanamycin, 50 #g/ml; tetracycline, 25/~g/ml; and hygromycin B, 200/~g/ml. Transformation of DNA into yeast strain DBY689 (Mata, leu2(AH), ura3-50, can1-101, from Dr. D. Botstein) was achieved using yeast protoplasts as described by Hinnen et al. (1978) with minor modifications. Instead of adding the yeast to regeneration top agar, the yeast were added to complete medium containing 1.2 M sorbitol and 20 ml aliguots were added to each petri dish. For complementing auxotrophs, the medium contained 0.67% yeast nitrogen base without amino acids (Difco), 2% destrose, 3% agar, 1.2 M sorbitol and other nutritional requirements at 20 I~g/ml. For hygromycin B selections, the suspension medium contained 1% Bacto yeast extract, 2% Bacto peptone, 2% dextrose, 1.2 M sorbitol, and 3% agar. After varying amounts of time the plates were overlaid with the same solution containing hygromycin B. The estimation of final concentration of hygromycin B in the plate assumed uniform diffusion of the drug throughout the plate.
The ability o f this gene fragment to function in gene fusions was demonstrated by fusing it to the E. coli lac gene contained on the pUC plasmids (Vieira and Messing 1982). The pUC plasmids contain a BamHI site downstream o f t h e / a c promoter and translation initiation site, thus facilitating the fusion o f the protein coding regions o f genes. The reading frame through the BamHI site of the hygromycin B gene is the same as that in pUC7, but out o f frame with the BamHI site in pUC8. If the resistance gene required transcriptional and translational activators for activity, then only the promoter in pUC7 should be effective in driving expression o f the resistance gene, and then only when the resistance gene was inserted in the proper orientation with respect to the promoter. The BamHI-BglII fragment from piT123 (Fig. 1) was ligated in both orientations into t h e B a m H l sites o f pUC7 and pUC8. Out o f these four plasmids, only the insert in the correct orientation into pUC7 was able to confer hygromycin B resistance to E. coli Fusions o f yeast genes to the hygromycin B gene were also constructed. A BamHI site was inserted downstream o f the translation initiation site of three yeast genes: the heat shock gene, YG100 (Fig. 2) (Ingolia et al. 1982); the heat shock cognate gene, YG101 ; and the phosphoglycerate kinase gene on piT141. The fused genes were inserted into a plasmid containing the following genetic elements (Fig. 3): an E. coli origin o f replication and the ampicillin resistance gene from pBR322, the kanamycin resistance gene from Tn601(903), a yeast origin of
Results and discussion A bacterial hygromycin B gene from pKC222 (Rao et al. 1983) was truncated to facilitate gene fusions. BamHI linker (5'-TGGATCCA-3') was substituted for the DNA encoding the first three amino acids o f the resistance gene, as shown in Fig. 1. The sequence downstream of the BamHI site contains the coding information for the resistance gene but without a translation initiation site and transcriptional activator sequence.
356
K . R . Kaster et al. : Hygromycin B resistance in yeast
pBR322 ~ .
YG 100PR : : . ~ - , pBR322 PstI XbaI NcoI EcoRI aa 110 aa 242 Nco I Digestion Bal 31 Eco RI Linkers
p i t 115 pBR322 ~ - : : Pet I Xba I
I~ Eco RI aa 5 Xba ] Digestion
Xba ]
/Eco RI " ~ H i n d III Hi ~[ p l N 1II A I J ~ Bam /"
pBR322
~ Eco R[ Digestion •
•
J'
~¢" Xba I Digestion Eco RI Digestion
piT 1 16 XbaI
pBR322"~'~ I
Eco RI I/I ~-",o,'~-o _ P B R322 Bam HI Hind III ~Xba I Digestion
Hind III /
~ . Bgl II [ "~ Xba I ( B H 1 )
~Bam HI Digestion ~ V Xba I Digestion ~ ' ~ B a m Bam HI Digestion
HI
/
piT 120
Eeo RI YG100 ~Hind HI Hind I[I Bgl II Xba I Activat°r~l,~Bam=. ,.nH]..~9o pBR322 ~ - ~ .;" ' "---'-" ~ - ~ ~". . . . . Bam H! I" ATG TCA AAA GCT GGA GGA ATT CCA AGC TTG GATCC
piT 2 1 3 Eco RI I . H i n d III
]
/
-"o~\ ~
/ /
~ 10.7 kb \
^ ..,~
Pst I
" ~ ~ j Eco RI ', ~-~Hind I I I
~ 1 ~ ""Pst I ~.
,
,,.
Bam HI ,,2~"'--LL..~
/~- Sma: Pvu II
.,~,~'Z
EcoRI
~"N'sma
i
EcoRI / ~ \PvuI . . -Pst I I I'sma: P;U-I-I H,nd III Bam:BgllI Hind III
Fig. 3. Functional units on piT213, a yeast-E, coli bifunctional vector which contains a truncated hygromycin resistance gene. The sole BamHI site in piT213 facilitates fusion of activator sequences with the truncated hygromycin B resistance gene
replication from 2 micron DNA, and the yeast ura3 gene. The various plasmids are listed in the Table. The plasmids containing the yeast gene-hygromycin B resistance gene fusions were found to confer hygromycin B resistance to yeast. The plasmids were transformed
Fig. 2. Construction of a portable activator sequence from a yeast heat shock gene. The details of the plasmid constructions and the origin of the starting materials is described in materials and methods. The sequence of the translated region up to the BamHI site is shown for piT120
into yeast, and cells containing plasmids with the yeast gene fusions (piT215, the heat shock cognate gene; piT217, the heat shock gene; and piT219, the phosphoglycerate kinase gene) were able to grow on YPD plates containing high levels of hygromycin B whereas control cells without plasmid or cells containing the other plasmids listed in the Table were unable to grow on plates containing 200/~g/ml hygromycin B. Since piT215 and piT219 were most effective in conferring resistance to the drug (see the Table), these plasmids were used to study direct selection of drug resistance after transformation. Ceils receiving piT215 or piT219 could be directly selected after transformation without first selecting for uracil prototrophy. The transformed cells were plated in 20 ml of YPD plus 1.2 M sorbitol and 3% agar and incubated at 30 °C for varying lengths of time. They were then overlaid with 10 ml of the same medium containing 1.5 mg/ml hygromycin B. Incubation was continued at 30 °C. Although the recovery time required after transformation varied from experiment to experiment, satisfactory results could be achieved by adding the hygromycin B at 4 h or at 20 h after transformation. The requirement for recovery time may be related to the mechanism of action of hygromycin B, which is the introduction of translational errors through perturbation of the protein synthetic machinery (Gonzalez et al. 1978). Perhaps the phosphotransferase must accumulate before resistance can be achieved. In some experiments
K. R. Kaster et al. : Hygromycin B resistance in yeast adding the hygromycin B 4 h after transformation killed the transformed cells as well as the untransformed cells, in which case adding the drag 20 h after transformation allowed unambiguous selection of transformants. In other experiments, the background growth 20 h after transformation was already too high to allow easy selection of tranformants after adding the hygromycin B, in which case addition of the drag 4 h after transformarion allowed selection of transformed cells. The variability could be due to the extent of cell wall digestion by the zymolyase used in protoplasting. The transformants per ug of piT215 or piT219 recovered after selection with hygromycin B was about 10% of the frequency determined by selection for uracil prototrophy. However, all of the uracil prototrophs selected after transformation with piT215 or piT219 were hygromycin B resistant when patched onto YPD plates containing 1 mg/ml antibiotic. Also, all the hygromycin B resistant cells selected directly after transformation were also prototrophic for uracil. It is possible that increasing the efficiency of expression of the resistance gene would elevate the transformation frequency. Several possible ways to increase the levels of expression of the resistance gene are being tested. One idea is to introduce a yeast transcriprional terminator after the hygromycin B gene. Introduction of a terminator after a foreign gene has been found in other systems to elevate the level of expression of the foreign gene (Hitzeman et al. 1983). Another way to increase the efficiency of the hygromycin B selection might be to use the lithium acetate (Ito et al. 1983) or calcium chloride (Iimura et al. 1983) methods of yeast transformation to eliminate the deleterious effects of cell wall digestion on subsequent achievement of hygromycin B resistance. However, in our hands the lithium-catalyzed transformation frequency is at least the fold lower than the protoplast-mediated frequency. Even if the frequency is not improved, though, the lithium-catalyzed transformation procedure might eliminate the variability in recovery rime needed before adding the hygromycin B. Drag resistance genes effective in conferring resistance to E. coli were also tested for their ability to confer drag resistance to yeast. Plasmid piT201 (see the Table) carrying the tambda p L promoter driving transcription of the bacterial hygromycin B resistance gene was particularly effective in conferring drug resistance to E. coli (Kaster et al. 1983) but was ineffective in yeast. We also tested the efficacy of the Tn601(903) genetic element as a selectable marker in yeast. Two groups have reported the use of this genetic element to confer G418 resistance to yeast (Jimenez and Davies 1978; Hollenberg 1982). However, we found that in our plasmid constructions, Tn601(903) did not confer elevated levels of G418 resistance to yeast cells. Perhaps the bacterial transcriptional and translational activator sequences are not
357 capable of driving sufficient levels of expression of the gene product in yeast, and the G418 resistance observed by others was due to gene expression driven by yeast activator sequences upstream of the resistance gene. Supporting this notion is the finding of Jimenez and Davies (1980) that the plasmids recovered from G418 resistant yeast cells were rearranged with respect to the input plasmids, and may have had yeast activator sequences linked to the G418 resistance gene. The yeast genes we successfully used to activate the resistance gene were the heat shock gene YG100, the heat shock cognate gene YG101, and the phosphoglycerate kinase gene. The fusions introduced 12, 12, and 3 foreign amino acids, respectively, in front of the hygromycin B resistance gene by linking the protein coding regions of the genes. Earlier work had shown that up to 15 amino acids from an E. coli gene fused to the truncated resistance gene still allowed expression of hygromycin B resistance in E. coli (Kaster et al. 1983). Fusing genes from an organism to the hygromycin B resistance gene via protein coding-region fusions has the advantage of maintaining possible translational activator sequences in front of the resistance gene. In cases where a certain organism might utilize translational activators, such as the Shine-Dalgarno sequence of prokaryotes, it might be difficult to achieve drug resistance in that organism with fusionsin the 5' untranslated region of the gene. (In yeast, however, transcriptional fusions using this hygromycin resistance gene are also effective, as shown by Gritz and Davies (1983) using a fusion to the yeast cycl gene.) Thus, use of the appropriate activator sequence fused to the protein coding region of the truncated hygromycin B resistance gene should facilitate hygromycin B resistance in any sensitive organism. Since most celis are sensitive to hygromycin B, this antibiotic and corresponding resistance gene should prove useful in a wide variety of systems.
Acknowledgements. We thank John Woolford, Richard Gaber, Mike Culbertson, Mike Slater and Elizabeth Craig for providing us with recombinant plasmids. We also thank Martin Bard and Nagaraja Rao for many helpful discussions, Cheryl Brazzell for communicating studies of direct drug resistance selections in yeast, Julian Davies for sending research results prior to publication, and Brigitte Schoner for carefully reading the manuscript. Finally, we thank Dick Baltz, Paul Burnett, and Eli Lilly and Company for their support of this word, and Cheryl Alexander for typing the manuscript.
References Botstein D, Falco SC, Stewart SE, Brennan M, Scherer S, Stinchcomb DT, Struhl K, Davis RW (1979) Gene 8:17-24 Craig EA, McCarthy EJ, Wadsworth SC (1979) Cell 16:575-588 Davies J, Smith DI (1978) A Rev Microbiol 32:469-518
358 Dobson MJ, Tuite MF, Roberts NA, Kingsman AJ, Kingsman SM (1982) Nucleic Acids Res 10:2625-2637 Gaber RF, Culbertson MR (1982) Gene 19:163-172 Gonzalez A, Jimenez A, Vazquez D, Davies JE, Schindler D (1978) Biochim Biophys Acta 5 2 1 : 4 5 9 - 4 6 9 Grindley NDF, Joyce CM (1980) Proc Natl Acad Sci USA 77: 7176-7180 Gritz L, Davies J (1983) Gene 25:179-188 Hinnen A, Hicks JB, Fink GR (1978) Proc Natl Acad Sci USA 75:1929-1933 Hinnen A, Farabaugh P, Ilgen C, Fink GR (1979) ICN-UCLA Symposium 14. Academic Press, New York, pp 4 3 - 5 0 Hitzeman RA, Clarke L, Carbon J (1980) J Biol Chem 255: 12073-12080 Hitzeman RA, Hagie FE, Hayflick JS, Chen CY, Seeburg PH, Derynck P (1982) Nucleic Acids Res 10:7791-7808 Hitzeman RA, Chen CY, Hagie FE, Patzer E J, Liu CC, Estell DA, Miller JV, Yaffe A, Leid DG, Levinson AD, Opperman H (1983) Nucleic Acids Res 11:2745-2763 Hollenberg CP (1982) Hofschneider PH, Goebel W (eds In: Gene cloning in organisms other than E. coli. Springer, New York Iimura Y, Gotoh K, Ouchi K, Nishiya T (1983) Agric Biol Chem 47:897-901 Ingolia TD, Slater MR, Craig EA (1982) Mol Cell Biol 2 : 1 3 8 8 1398
K.R. Kaster et al.: Hygromycin B resistance in yeast Ito H, Fukuda Y, Murata K, Kimura A (1983) J Bacteriol 153: 163-168 Jimenez A and Davies J (1980) Nature 287:869-871 Johnson RA, Walseth TF (1979) Adv. Cyclic Nucleotide Res 10:135-167 Kaster KR, Burget SG, Rao RN, Ingolia TD (1983) Nucleic Acids Res 11:6895-6911 Maxam A, Gilbert W (1977) Proc Natl Acad Sci USA 7 4 : 5 6 0 564 Nakamura K, Inouye M (1982) EMBO Journal 1:771-775 Rao RN, Allen NE, Hobbs JN, Alborn WE, Kirst HA, Paschal JW (1983) Antimicrob Agents Chemother 24:689-695 Schleif RF and Wensink PC (1981) Practical methods in molecular biology. Springer, New York Struhl K, Stinchcomb DT, Scherer S, Davis RW (1979) Proc Natl Acad Sci USA 76:1035-1039 Struhl K (1982) Nature 300:284-287 Vieira J, Messing J (1982) Gene 19:259-268 Woolford JL, Jr, Rosbash M (1981) Nucleic Acids Res 9 : 5 0 2 1 5036 C o m m u n i c a t e d b y K. Esser Received January 27/February 22, 1984
~
~
~
© Springer-Verlag 1984
Hygromycin B resistance as dominant selectable marker in yeast Kevin R. Kaster, Stanley G. Burgett, and Thomas D. lngolia Lilly Research Laboratories, 307 E. MeCarty Street, Indianapolis, Indiana 46285, USA
Summary. Saccharomyces cerevisiae is normally sensitive to the drug hygromycin B; a hygromycin B concentration of 200 /~g/ml in agar plates is sufficient to completely inhibit growth. We constructed yeast-E, coli bifunctional plasmids which confer hygromycin B resistance to Saccharomyces cerevisiae Promoters and amino terminal coding regions of a heat shock gene, a heat shock cognate gene, and the phosphoglycerate kinase gene from yeast were fused to a bacterial hygromycin B resistance gene. In all three cases, yeast cells containing plasmids with the hybrid hygromycin B resistance gene were resistant to high levels of the drug. Yeast cells containing these ptasmids can also be directly selected after transformation by using hygromycin B. The intact bacterial hygromycin B resistance gene and the kanamycin resistance gene from Tn903 were also tested in yeast for their ability to confer resistance to hygromycin B and G418. The intact bacterial genes were not effective in conferring drug resistance to yeast cells.
Key words: Hygromycin B - Yeast - Plasmids
Introduction Complementation of auxotrophic markers has been used in yeast transformations in order to select transformed cells. For example, the leu2 (I-finnen et al. 1978), ura3 (Struhl et al. 1979), his3 (Struhl 1982), his4 (Hinnen et al. 1979), and trpl (Hitzeman et al. 1980) genes from Saccaromyces cerevisiae have been used as selectable markers, and presumably any auxotrophic marker could be used. However, complementation of an auxotrophic marker does require that the recipient cell contain the
Offprint requests to: T. D. Ingolia
appropriate marker. This limitation can be a minor inconvenience, requiring only a genetic cross to introduce the appropriate marker into the desired background, or a major obstacle, for example, when one wants to introduce DNAinto another species in which the required genetic lesions are not known. A plasmid-bome dominant selectable marker would be a valuable experimental tool in these cases. We therefore examined several fungicidal drugs and the corresponding inactivating genes from prokaryotes for their efficacy as selectable marker in yeast. The aminoglycoside antibiotics G418 andhygromycin B were chosen because yeast cells are sensitive to G418 (Jimenez and Davies 1980), and hygromycin B (Gonzalez et al. 1978) and prokaryotic drag inactivation systems exist for each of these antibiotics. Both of these drugs disrupt ribosomal function and cause translational errors (Gonzalez et al. 1978; Davies and Smith 1978), and the drug resistance genes inactivate the drugs by covalent modification (Gortzalez et al. 1978; Davies and Smith 1978; Rao et al. 1983). Bacterial resistance to G418 can be conferred by several plasmid-encoded modifying enzymes (Davies and Smith 1978)including the aminoglycoside phosphotransferase encoded by Tn601 (903). In E. coli, a plasmid-bome gene encoding a hygromicin B phosphotransferase (Rao et al. 1983) confers resistance to hygromycin B. The DNA sequence of this gene has been determined and the protein coding region has been identified (Kaster et al. 1983). A nearly identical hygromycin B resistance gene has also been described by Gritz and Davies (Gritz and Davies 1983). Our previous results have shown that the first three amino acids of the hygromycin B phosphotransferase can be replaced by up to 15 amino acids from a foreign gene and still retain activity (Kaster et al. 1983). In this paper we report the construction and characterization of plasmids containing a truncated hygromycin
354 B resistance gene f u s e d t o y e a s t genes, a n d d e m o n s t r a t e t h e e f f e c t i v e n e s s o f t h e s e plasmids in c o n f e r r i n g h y g r o m y cin B resistance t o y e a s t cells.
Materials a n d m e t h o d s
Materials. G418 was provided by Dr. Peter Daniels and the Schering Corporation. Hygromycin B was obtained from Eli Lilly and Company. BamHI and EcoR1 linkers were provided by Dr. Rama Belagaje. Plasmids YG100PR1 and YG101HI were from Mike Slater and Elizabeth Craig. [Gamma-32P]-ATP was synthesized from 32P-phosphate and ADP (Johnson and Walseth 1979). Zymolyase 60,000 was obtained from Miles. All other biochemieals and enzymes were obtained from commercial sources. Enzyme reactions. Restriction enzyme digestion of DNA was performed according to suppliers' recommendation. Protruding ends of restricted DNA molecules were removed, when necessary, by incubating up to 10 ~g of DNA with about three units of DNA polymerase I large fragment in the presence of all four dNTPs at 300 /~M concentration. Double stranded regions of DNA were shortened by digestion with Bal31 nuclease. For removing long (greater than 100 base pairs) stretches, about 10 /~g of DNA were incubated at 37 °C with three units of Bal31 nuclease under conditions specified by the manufacturer. Atiquots were removed at different times after addition of enzyme, quenched with phenol, and analyzed on polyacrylamide gels to determine the extent of digestion. For removing shorter stretches of DNA, the units of enzyme per/~g of DNA were reduced and the reaction carried out at 4 °C. Ligation of DNA molecules, including synthetic oligonucleotides, was accomplished using standard conditions (Schliefand Wensink 1981). When necessary, sequences of regions of plasmids were checked by the chemical cleavage method of Maxam and Gilbert (Maxam and Gilbert 1977). Construction o f plasmids. The hygromycin resistance gene was obtained from pKC222 (Kaster et al. 1983). This plasmid was restricted with HphI and Pstl, and the fragment containing the amino-terminal coding region was purified. The 3' protruding end left by the HphI enzyme was removed with E. coli DNA polymerase I large fragment (Klenow fragment) and BamHI linkers (5'-TGGATCCA) ligated. The fragment was restricted with BamHI and EcoRI and ligated into the BamHI and EcoRI sites of pBR322. This intermediate plasmid, called piT122, was cut with EcoRI, and the hygromycin resistance gene was reconstructed by inserting the small EcoRI fragment from pKC222 into piT122. The resulting plasmid, called piT123, contains the hygromycin resistance gene with a BamHI linker substituted for the first three codons of the protein coding region. As shown in Fig. 1, the truncated hygromycin resistance gene on piT123 can be excised on a BamHI-BgllI fragment. The activator sequences from three yeast genes, a heat shock gene, a heat shock cognate gene, and the phosphoglycerate kinase gene, were modified to facilitate linkage to the hygromycin resistance gene. The heat shock gene is the yeast analogue to the major heat shock gene of Drosophila, hsp70, and is contained on pYG100 (Ingolia et al. 1982). Plasmid pYG100PR1, consisting of a PstI-EeoRI fragment from pYG100 containing the 5' end of the heat shock gene subcloned into pBR322, was used as the starting material, as shown in Fig. 2. This plasmid was digested with NeoI, which cuts within the coding information for amino
K.R. Kaster et al. : Hygromycin B resistance in yeast acid 110 of the heat shock protein, digested with Ba131 nuclease, and figated to EcoRI linkers. One isolate, called p i t 115, contained a linker inserted into the coding region for amino acid 5 of the heat shock protein. In order to place a BamHI site near the EcoRI site, the XbaI-BamHI fragment from p l T l l 5 was cloned into plNIIIAI (Nakamura and Inouye 1982) yielding piT116. In order to place a BglII site upstream of the activator sequence, the XbaI-BamHI fragment of p l T l l 6 was cloned into pBHI, a recombinant plasmid containing Drosophila DNA (Craig et al. 1979). The resulting plasmid, piT120, contains the activator sequence from the pYG100 heat shock gene on BglII-BamHI fragment. The sequence near the translation initiation site is also shown in Fig. 2. The reading frame at the BamHI site is the same as that used for the hygromycin resistance gene in piT123. The activator sequence from a yeast heat shock cognate gene was also modified. The gene was contained in pYG101H1, a HindlII subclone into pBR322 obtained from Mike Slater and Elizabeth Craig. A 680bp ClaI-EcoRI fragment containing the 5' end of the gene was cloned into Clal-EcoRI-cut pBR322 to yield piT107. This plasmid was cleaved with C/aI, which cuts at amino acid 10 of the coding region for the heat shock cognate gene, treated with Bal31 nuclease and ligated to BamHI linkers. One isolate, called p l T l l 0 , had a BamHI tinker inserted after the ninth codon of the protein coding region. Beginning with the translation initiation site, the DNA sequence of piT110 includes 5 '-ATG GCT GAA GGT GTT TTC CAA GGT GCT cgg gat cc - 3', where the linker is depicted in small case letters. In order to introduce a BgllI site upstream of the activator sequence, an EcoRI fragment from pYG100BH (Ingolla et al. 1982) was cloned into the EcoRI site of p l T l l 0 to yield plTll8. The activator sequence from the heat shock cognate gene can be liberated on a BgllI-BamHI fragments from piT118, and the reading frame at the BamHI site is the same as the reading frame at the BamHI site of the truncated resistance gene on piT123. The phosphoglycerate kinase gene activator sequence was also adapted for use in gene fusions. The gene was isolated from a bank of yeast genome fragments cloned into lambda phage provided by Dr. John Woolford (Woolford and Rosbash 1981). The probe was a pBR322 derivative containing the Saccharomyces cerevisiae SUF16 gene, provided by Richard Gaber and Michael Culbertson. The SUF16 gene resides within about 5 kb of the phosphoglycerate kinase gene (Gaber and Culbertson 1982), so many recombinant phages should carry both genes. One such phage was identified, and a HindIII fragment containing the phosphoglycerate kinase gene and flanking regions (Hitzeman et al. 1982; Dobson et al. 1982) was cloned into pBR322 to yield piT141. A 958 bp HinclI-ClaI fragment containing the 5' end of the phosphoglycerate kinase gene was purified from pIT141, and subsequently cut with MbolI. The 5'-extensions resulting from the MbolI cleavage were filled in with DNA polymerase I large fragment and 8-met BamHI linkers ligated to the molecules. A 240 bp fragment was purified on an acrylamide gel and ligated into BamHI-cut pUC8 (Vieira and Messing 1982). One isolate, called piT143, contained a BamHI fragment which comprised the 5' end of the phosphoglycerate kinase gene. The DNA sequence of piT143, beginning with the translation initiation site, was found to be 5'-Atg gat cc - 3 ' , where the tinker sequence is depicted in small case letters. The BamHI site of • pIT143 is in phase with theBamHI site of the truncated hygromycin resistance gene on pIT123. The backbone for our yeast-E, coli bifunctional vectors was YEp24 (Botstein et al. 1978). The kanamycin resistance gene from Tn601(903) was incorporated into YEp24 by purifying a 1.7 kb PvulI fragment from pNG614 (a derivative of pNG18; (Grindley and Joyce 1980) and ligating it into the sole Sinai
355
K. R. Kaster et al. : Hygromycin B resistance in yeast
piT 123
Table. 1. Plasmids tested for ability to confer hygromycin B resistance to yeast Plasmid
piT201 piT213 piT215 piT216 piT217 piT219
Activator sequence
Level of Hm resistance in yeast,/~g/ml
lambda pL promoter < 200 None < 200 yeast heat shock cognate gene 1000 yeast heat shock cognate gene, opposite orientation as in pIT215 < 200 yeast heat shock gene 500 yeast phosphoglycerate kinase gene >1000
~ Pst
B
1/%
Eco RI I Hind III g l II ~81bp
///~
~NNkN~Hinc I I
//+ I /,
I' "
_
== " , = . , .
. . . . ='
I I-- Psi I ) I--, 90bp / / Eco RI 240bp ' A B a m HI
The construction and structure of the plasmids is described in materials and methods and Fig. 1-3. Only piT201 will confer hygromycin resistance to E. coli
site of YEp24. The truncated hygromycin gene was also added to YEp24 by ligating a BamHI-BgllI fragment from piT123 into the sole BamHI site of YEp24. The resultant plasmid, called pIT213, is shown in Fig. 3. The activator seuqences described above were then ligated into the sole BamH1 site on piT213 to yield the plasmids described in the Table 1.
Fig. 1. Truncated hygromycin resistance gene on piT123. This plasmid contains the truncated hygromycin B resistance gene. The sequence at the BamHI site, showing the reading frame utilized for the hygromycin B resistance gene, is also shown
Transformations. Transformation of DNA into E. coli was accomplished using the standard CaC12 plus heat shock procedure (Schlief and Wensink 1981). The drug concentrations used for selection in E. coli were: ampicillin, 50 ~g/ml; kanamycin, 50 #g/ml; tetracycline, 25/~g/ml; and hygromycin B, 200/~g/ml. Transformation of DNA into yeast strain DBY689 (Mata, leu2(AH), ura3-50, can1-101, from Dr. D. Botstein) was achieved using yeast protoplasts as described by Hinnen et al. (1978) with minor modifications. Instead of adding the yeast to regeneration top agar, the yeast were added to complete medium containing 1.2 M sorbitol and 20 ml aliguots were added to each petri dish. For complementing auxotrophs, the medium contained 0.67% yeast nitrogen base without amino acids (Difco), 2% destrose, 3% agar, 1.2 M sorbitol and other nutritional requirements at 20 I~g/ml. For hygromycin B selections, the suspension medium contained 1% Bacto yeast extract, 2% Bacto peptone, 2% dextrose, 1.2 M sorbitol, and 3% agar. After varying amounts of time the plates were overlaid with the same solution containing hygromycin B. The estimation of final concentration of hygromycin B in the plate assumed uniform diffusion of the drug throughout the plate.
The ability o f this gene fragment to function in gene fusions was demonstrated by fusing it to the E. coli lac gene contained on the pUC plasmids (Vieira and Messing 1982). The pUC plasmids contain a BamHI site downstream o f t h e / a c promoter and translation initiation site, thus facilitating the fusion o f the protein coding regions o f genes. The reading frame through the BamHI site of the hygromycin B gene is the same as that in pUC7, but out o f frame with the BamHI site in pUC8. If the resistance gene required transcriptional and translational activators for activity, then only the promoter in pUC7 should be effective in driving expression o f the resistance gene, and then only when the resistance gene was inserted in the proper orientation with respect to the promoter. The BamHI-BglII fragment from piT123 (Fig. 1) was ligated in both orientations into t h e B a m H l sites o f pUC7 and pUC8. Out o f these four plasmids, only the insert in the correct orientation into pUC7 was able to confer hygromycin B resistance to E. coli Fusions o f yeast genes to the hygromycin B gene were also constructed. A BamHI site was inserted downstream o f the translation initiation site of three yeast genes: the heat shock gene, YG100 (Fig. 2) (Ingolia et al. 1982); the heat shock cognate gene, YG101 ; and the phosphoglycerate kinase gene on piT141. The fused genes were inserted into a plasmid containing the following genetic elements (Fig. 3): an E. coli origin o f replication and the ampicillin resistance gene from pBR322, the kanamycin resistance gene from Tn601(903), a yeast origin of
Results and discussion A bacterial hygromycin B gene from pKC222 (Rao et al. 1983) was truncated to facilitate gene fusions. BamHI linker (5'-TGGATCCA-3') was substituted for the DNA encoding the first three amino acids o f the resistance gene, as shown in Fig. 1. The sequence downstream of the BamHI site contains the coding information for the resistance gene but without a translation initiation site and transcriptional activator sequence.
356
K . R . Kaster et al. : Hygromycin B resistance in yeast
pBR322 ~ .
YG 100PR : : . ~ - , pBR322 PstI XbaI NcoI EcoRI aa 110 aa 242 Nco I Digestion Bal 31 Eco RI Linkers
p i t 115 pBR322 ~ - : : Pet I Xba I
I~ Eco RI aa 5 Xba ] Digestion
Xba ]
/Eco RI " ~ H i n d III Hi ~[ p l N 1II A I J ~ Bam /"
pBR322
~ Eco R[ Digestion •
•
J'
~¢" Xba I Digestion Eco RI Digestion
piT 1 16 XbaI
pBR322"~'~ I
Eco RI I/I ~-",o,'~-o _ P B R322 Bam HI Hind III ~Xba I Digestion
Hind III /
~ . Bgl II [ "~ Xba I ( B H 1 )
~Bam HI Digestion ~ V Xba I Digestion ~ ' ~ B a m Bam HI Digestion
HI
/
piT 120
Eeo RI YG100 ~Hind HI Hind I[I Bgl II Xba I Activat°r~l,~Bam=. ,.nH]..~9o pBR322 ~ - ~ .;" ' "---'-" ~ - ~ ~". . . . . Bam H! I" ATG TCA AAA GCT GGA GGA ATT CCA AGC TTG GATCC
piT 2 1 3 Eco RI I . H i n d III
]
/
-"o~\ ~
/ /
~ 10.7 kb \
^ ..,~
Pst I
" ~ ~ j Eco RI ', ~-~Hind I I I
~ 1 ~ ""Pst I ~.
,
,,.
Bam HI ,,2~"'--LL..~
/~- Sma: Pvu II
.,~,~'Z
EcoRI
~"N'sma
i
EcoRI / ~ \PvuI . . -Pst I I I'sma: P;U-I-I H,nd III Bam:BgllI Hind III
Fig. 3. Functional units on piT213, a yeast-E, coli bifunctional vector which contains a truncated hygromycin resistance gene. The sole BamHI site in piT213 facilitates fusion of activator sequences with the truncated hygromycin B resistance gene
replication from 2 micron DNA, and the yeast ura3 gene. The various plasmids are listed in the Table. The plasmids containing the yeast gene-hygromycin B resistance gene fusions were found to confer hygromycin B resistance to yeast. The plasmids were transformed
Fig. 2. Construction of a portable activator sequence from a yeast heat shock gene. The details of the plasmid constructions and the origin of the starting materials is described in materials and methods. The sequence of the translated region up to the BamHI site is shown for piT120
into yeast, and cells containing plasmids with the yeast gene fusions (piT215, the heat shock cognate gene; piT217, the heat shock gene; and piT219, the phosphoglycerate kinase gene) were able to grow on YPD plates containing high levels of hygromycin B whereas control cells without plasmid or cells containing the other plasmids listed in the Table were unable to grow on plates containing 200/~g/ml hygromycin B. Since piT215 and piT219 were most effective in conferring resistance to the drug (see the Table), these plasmids were used to study direct selection of drug resistance after transformation. Ceils receiving piT215 or piT219 could be directly selected after transformation without first selecting for uracil prototrophy. The transformed cells were plated in 20 ml of YPD plus 1.2 M sorbitol and 3% agar and incubated at 30 °C for varying lengths of time. They were then overlaid with 10 ml of the same medium containing 1.5 mg/ml hygromycin B. Incubation was continued at 30 °C. Although the recovery time required after transformation varied from experiment to experiment, satisfactory results could be achieved by adding the hygromycin B at 4 h or at 20 h after transformation. The requirement for recovery time may be related to the mechanism of action of hygromycin B, which is the introduction of translational errors through perturbation of the protein synthetic machinery (Gonzalez et al. 1978). Perhaps the phosphotransferase must accumulate before resistance can be achieved. In some experiments
K. R. Kaster et al. : Hygromycin B resistance in yeast adding the hygromycin B 4 h after transformation killed the transformed cells as well as the untransformed cells, in which case adding the drag 20 h after transformation allowed unambiguous selection of transformants. In other experiments, the background growth 20 h after transformation was already too high to allow easy selection of tranformants after adding the hygromycin B, in which case addition of the drag 4 h after transformarion allowed selection of transformed cells. The variability could be due to the extent of cell wall digestion by the zymolyase used in protoplasting. The transformants per ug of piT215 or piT219 recovered after selection with hygromycin B was about 10% of the frequency determined by selection for uracil prototrophy. However, all of the uracil prototrophs selected after transformation with piT215 or piT219 were hygromycin B resistant when patched onto YPD plates containing 1 mg/ml antibiotic. Also, all the hygromycin B resistant cells selected directly after transformation were also prototrophic for uracil. It is possible that increasing the efficiency of expression of the resistance gene would elevate the transformation frequency. Several possible ways to increase the levels of expression of the resistance gene are being tested. One idea is to introduce a yeast transcriprional terminator after the hygromycin B gene. Introduction of a terminator after a foreign gene has been found in other systems to elevate the level of expression of the foreign gene (Hitzeman et al. 1983). Another way to increase the efficiency of the hygromycin B selection might be to use the lithium acetate (Ito et al. 1983) or calcium chloride (Iimura et al. 1983) methods of yeast transformation to eliminate the deleterious effects of cell wall digestion on subsequent achievement of hygromycin B resistance. However, in our hands the lithium-catalyzed transformation frequency is at least the fold lower than the protoplast-mediated frequency. Even if the frequency is not improved, though, the lithium-catalyzed transformation procedure might eliminate the variability in recovery rime needed before adding the hygromycin B. Drag resistance genes effective in conferring resistance to E. coli were also tested for their ability to confer drag resistance to yeast. Plasmid piT201 (see the Table) carrying the tambda p L promoter driving transcription of the bacterial hygromycin B resistance gene was particularly effective in conferring drug resistance to E. coli (Kaster et al. 1983) but was ineffective in yeast. We also tested the efficacy of the Tn601(903) genetic element as a selectable marker in yeast. Two groups have reported the use of this genetic element to confer G418 resistance to yeast (Jimenez and Davies 1978; Hollenberg 1982). However, we found that in our plasmid constructions, Tn601(903) did not confer elevated levels of G418 resistance to yeast cells. Perhaps the bacterial transcriptional and translational activator sequences are not
357 capable of driving sufficient levels of expression of the gene product in yeast, and the G418 resistance observed by others was due to gene expression driven by yeast activator sequences upstream of the resistance gene. Supporting this notion is the finding of Jimenez and Davies (1980) that the plasmids recovered from G418 resistant yeast cells were rearranged with respect to the input plasmids, and may have had yeast activator sequences linked to the G418 resistance gene. The yeast genes we successfully used to activate the resistance gene were the heat shock gene YG100, the heat shock cognate gene YG101, and the phosphoglycerate kinase gene. The fusions introduced 12, 12, and 3 foreign amino acids, respectively, in front of the hygromycin B resistance gene by linking the protein coding regions of the genes. Earlier work had shown that up to 15 amino acids from an E. coli gene fused to the truncated resistance gene still allowed expression of hygromycin B resistance in E. coli (Kaster et al. 1983). Fusing genes from an organism to the hygromycin B resistance gene via protein coding-region fusions has the advantage of maintaining possible translational activator sequences in front of the resistance gene. In cases where a certain organism might utilize translational activators, such as the Shine-Dalgarno sequence of prokaryotes, it might be difficult to achieve drug resistance in that organism with fusionsin the 5' untranslated region of the gene. (In yeast, however, transcriptional fusions using this hygromycin resistance gene are also effective, as shown by Gritz and Davies (1983) using a fusion to the yeast cycl gene.) Thus, use of the appropriate activator sequence fused to the protein coding region of the truncated hygromycin B resistance gene should facilitate hygromycin B resistance in any sensitive organism. Since most celis are sensitive to hygromycin B, this antibiotic and corresponding resistance gene should prove useful in a wide variety of systems.
Acknowledgements. We thank John Woolford, Richard Gaber, Mike Culbertson, Mike Slater and Elizabeth Craig for providing us with recombinant plasmids. We also thank Martin Bard and Nagaraja Rao for many helpful discussions, Cheryl Brazzell for communicating studies of direct drug resistance selections in yeast, Julian Davies for sending research results prior to publication, and Brigitte Schoner for carefully reading the manuscript. Finally, we thank Dick Baltz, Paul Burnett, and Eli Lilly and Company for their support of this word, and Cheryl Alexander for typing the manuscript.
References Botstein D, Falco SC, Stewart SE, Brennan M, Scherer S, Stinchcomb DT, Struhl K, Davis RW (1979) Gene 8:17-24 Craig EA, McCarthy EJ, Wadsworth SC (1979) Cell 16:575-588 Davies J, Smith DI (1978) A Rev Microbiol 32:469-518
358 Dobson MJ, Tuite MF, Roberts NA, Kingsman AJ, Kingsman SM (1982) Nucleic Acids Res 10:2625-2637 Gaber RF, Culbertson MR (1982) Gene 19:163-172 Gonzalez A, Jimenez A, Vazquez D, Davies JE, Schindler D (1978) Biochim Biophys Acta 5 2 1 : 4 5 9 - 4 6 9 Grindley NDF, Joyce CM (1980) Proc Natl Acad Sci USA 77: 7176-7180 Gritz L, Davies J (1983) Gene 25:179-188 Hinnen A, Hicks JB, Fink GR (1978) Proc Natl Acad Sci USA 75:1929-1933 Hinnen A, Farabaugh P, Ilgen C, Fink GR (1979) ICN-UCLA Symposium 14. Academic Press, New York, pp 4 3 - 5 0 Hitzeman RA, Clarke L, Carbon J (1980) J Biol Chem 255: 12073-12080 Hitzeman RA, Hagie FE, Hayflick JS, Chen CY, Seeburg PH, Derynck P (1982) Nucleic Acids Res 10:7791-7808 Hitzeman RA, Chen CY, Hagie FE, Patzer E J, Liu CC, Estell DA, Miller JV, Yaffe A, Leid DG, Levinson AD, Opperman H (1983) Nucleic Acids Res 11:2745-2763 Hollenberg CP (1982) Hofschneider PH, Goebel W (eds In: Gene cloning in organisms other than E. coli. Springer, New York Iimura Y, Gotoh K, Ouchi K, Nishiya T (1983) Agric Biol Chem 47:897-901 Ingolia TD, Slater MR, Craig EA (1982) Mol Cell Biol 2 : 1 3 8 8 1398
K.R. Kaster et al.: Hygromycin B resistance in yeast Ito H, Fukuda Y, Murata K, Kimura A (1983) J Bacteriol 153: 163-168 Jimenez A and Davies J (1980) Nature 287:869-871 Johnson RA, Walseth TF (1979) Adv. Cyclic Nucleotide Res 10:135-167 Kaster KR, Burget SG, Rao RN, Ingolia TD (1983) Nucleic Acids Res 11:6895-6911 Maxam A, Gilbert W (1977) Proc Natl Acad Sci USA 7 4 : 5 6 0 564 Nakamura K, Inouye M (1982) EMBO Journal 1:771-775 Rao RN, Allen NE, Hobbs JN, Alborn WE, Kirst HA, Paschal JW (1983) Antimicrob Agents Chemother 24:689-695 Schleif RF and Wensink PC (1981) Practical methods in molecular biology. Springer, New York Struhl K, Stinchcomb DT, Scherer S, Davis RW (1979) Proc Natl Acad Sci USA 76:1035-1039 Struhl K (1982) Nature 300:284-287 Vieira J, Messing J (1982) Gene 19:259-268 Woolford JL, Jr, Rosbash M (1981) Nucleic Acids Res 9 : 5 0 2 1 5036 C o m m u n i c a t e d b y K. Esser Received January 27/February 22, 1984
