Cloning and expression of the yeast galactokinase gene in an Escherichia coli plasmid

Cloning and expression of the yeast galactokinase gene in an Escherichia coli plasmid

Gene, 5 (1979) 291--303 291 © Elsevier/North-HollandBiomedical Press, Amsterdam -- Printed in The Netherlands CLONING AND EXPRESSION OF THE YEAST GAL...

1MB Sizes 2 Downloads 41 Views

Gene, 5 (1979) 291--303 291 © Elsevier/North-HollandBiomedical Press, Amsterdam -- Printed in The Netherlands

CLONING AND EXPRESSION OF THE YEAST GALACTOKINASE GENE A N EschericMa coli PLASMID (Galactose gene cluster; BglII; BamHI; pBR322 vector; mRNA colony hybridization; gene bank; Saccharomyces eerevisiae; recombinant DNA) MARK A. SCHELLand DAVID B. WILSON Section of Biochemistry, Molecular and Cell Biology, Cornell University, Ithaca, NY 14853 (U.S.A.) (Received November 28th, 1978) (Accepted February 21st, 1979) SUMMARY This report describes the construction and isolation of a plasmid, derived from pBR322, which carries a BglII restriction fragw~nt of DNA containing the galactokinase gene from Saccharomyces cerevisiae. This was accomplished by the following procedure: (1) Purified galactokinase mRNA, labelled with t2sI, was hybridized to BglII digests of yeast DNA employing Southern's filter transfer technique to identify a restriction fragment containing the galactokinase gene. (2) This fragment was partially purified by agarose gel electrophoresis, ligated into the BamHI site of pBR322 and transformed into £scherichia ¢oU to generate a clone bank containing the galactokinase gene. (3) This bank was screened by in situ colony hybridization with galactokinase mRNA resulting in the identification of a plasmid carrying this gene. This plasmid DNA hybridized with the galactokinase mRNA to the same extent in the presence or absence of a large excess of unlabelled mRNA from cells that were not induced for galactokinase synthesis, while the same amount of unlabelled galactose-induced mRNA reduced the hybridization by 95%. When this plasmid was introduced into an E. coil strain deleted for the galactose operon it caused the synthesis of low levels of yeast galactokinase activity. INTRODUCTION We have previously reported the purification of the gal I gene product (galactokinase) from S. cerevisiae (Schell and Wilson, 1977). A specific antibody was prepared against this protein and subsequently used to purify the Abbreviations: BSA, bovine serum albumin; DTT, dithiothreitol; SD8, sodium dodeeyl sulphate.

292

mRNA coding for galactokinase by immunoprecipitation (Schell and Wilson, 1978). As the next step the mRNA was used as a probe to detect a chimeric plasmid containing the yeast galactokinase gene. The successful isolation of this plasmid would dem(instrate the utility of this type of sequential isolation (protein -~ antibody -~ mRNA -* DNA) to obtain probes to study and clone many other eucaryotic genes, whose mRNA's represent less than 2% of the total cellular mRNA. There are two possible approaches to the cloning of a specific gene fragment utilizing a purified mRNA as a probe for the gene sequence. The first approach, the "shot gun" method, involves the screening of a clone bank containing the entire yeast genome for a clone that would hybridize to galactokinase mRNA. However thi~ approach is complicated by the large number of rDNA clones that are pre~mt in a bank of the entire yeast genome as they will hybridize to the small ~mounts of rRNA which contaminate most mRNA preparations. Alternatively, the fragment containing the desired gene can be identified by the use of the method of Southern (1973) and partially purified before the generation cf the clone banl~ This method has several advantages. First the partially purified clone bank is much smaller than the "shot gun" bank and therefore is much easier to generate and screen. Second, hybridization conditions can be established for the identification of the specific gene fragment. Finally, ribosomal DNA can be eliminated from the clone bank by the choice of the proper restriction enzyme to generate the fragment. METHODS AND PROCEDURES

Preparation of z2sI-tabeUed galcctokinase mRNA Galactokinase mRNA was purified by immunoprecipitation of total yeast polysomes with galactokinase antibody, deproteinization of the RNA in the immunoprecipitated polysomes, and chromatography of this" RNA on polyuridylic acid linked Sepharose by the procedure described by Schell and Wilson (19"/9). In rive s2P-labelled mRNA prepared by this method did n o t have a high enough specific activity to be used to probe for the galactokinase gene in restriction enzyme digests of the entire yeast genome. Therefore, the galactokinase mRNA was purified after low level s2p labelling (in vivo; for monitoring purification) and iodinated with ~2sI by a modification of the method described by Prensky (1976). The reaction was prepared by the addition of the following reagents: 3.0 pl Na 12sI (0.4 Ci/ml in 50 mM NaOH; carrier free); 2.0 pl 0.15 N HNOs in 0.2 M NaAc (sodium acetate) pH 4.8; 1.5 ~I 0.2 mM KI in 0.1 M NaAc p_H 4.8; 5.0 , I RNA (0.2-0.6 pg)in 0.1 M NaAc pH 4.8; 2.0 ~l 0.1 M NaAc pH 4.8; 5.0 pl 1.4 mM TICIs in 0.1 M NaAc pH 4.8. After the TIC13 catalyst solution was added, the solution was incubated at 60°C for 25 rain; 0.2 ml of TNE (0.1 M Tris. HCI pH 8.0, 0.1 M NaCI, I mM EDTA) was added to the reaction an,d the incubation was continued for an additional 12 min. Then 0.05 mg of £. coli rRNA was added to the

293

solution and the iodinated mRNA was purified away from the unreacted '=sI by chromatography on 0.5 ml columns of CF-11 cellulose as described by Prensky (1976). The resultant RNA sample contained 3-107 cpm giving it a specific activity of 0.5--1.0- l 0 s cpm/~g RNA.

Restriction enzyme analysis of yeast DNA for the galactokinase gene Yeast DNA was prepared by the method described by CiTer et al. (1975) and further purified by equilibrium sedimentation centrifugation in CsCI. The DNA was digested by placing it in 10 mM Tris. HCI pH 7.5, 10 mM MgCI2, 6 mM DTT, 50 mM NaCI, 0.1 mg/ml BSA to a final DNA concentration of 0.05 to 0.3 mg/ml. Restriction enzyme was added to the solution at a ratio of 1 unit of enzyme to 1/~g of DNA and it was incubated at 37°C for 2--5 h. These samples (5--50/~1) were loaded onto 1% agarose slab gels (0.3 crn X 20 cm X 20 cm) containing 50 mM Tris--Ac pH 8.2, 20 mM NaAc, 20 mM NaCI, 3 mM EDTA. The gels were run at 50 mA for 10 h while circulating the electrophoresis buffer between the upper and lower tanks with a peristaltic pump (Wu et al., 1976). The restriction fragments separated on these gels were transferred onto nitrocellulose sheets by a modification of the method of Southern (1975). Hybridization of '25I-labelled galactokinase mRNA to these sheets was performed in "Seal-A-Meal" bags in a minimal volume of 5 X SSC + 0.5% SDS at 65°C for 20--30 h (SSC ffi 0.15 M NaCI, 0.015 M Na-citrate, pH 7.0). Usually 3 ~ of this hybridization solution containing 0.3 mg of E. coli tRNA and 0.5--2.0. l 0 s cpm of '25I-labelled galactokinase mRNA was used for a 10 X 13 cm sheet. After hybridization the nitrocellulose sheet was removed from the bag, washed 3 times with 200 ml of 2 × SSC + 0.5% SDS at 43°C for 30 rain each time, and then finally washed with 200 ml of 2 × SSC for 30 rain at 43°C three more times. The sheet was air-dried on a glass plate, coated with a solution of 20% PPO in toluene and dried again. The sheets were autoradiographed a t - 2 0 ° C using Kodak XR-5 X-ray film with a Picker intensifying screen for 2--14 days.

Preparation of plasmid DNA E. coli strains containing the plasmid pBR322 (Rodriguez et al., 1977) were grown at 37°C in M9 media supplemented with 0.2% glucose, 0.3% casamino acids, 0.1 mg/ml ampicillin, and any specific strain requirements to late log phase (As00nm = 0.9). Chloramphenicol was added to the culture at a final concentration of 0.17 mg/ml and the culture was shaken at 37°C for 12--16 h to amplify the plasmid. Plasmid DNA was prepared from these cells by a modification of the method of Tanaka and Weisblum (1975). The amount of DNA was determined by reading the A260nm and by the diphenylamine assay (Burton, 1968).

Extraction of DNA from agarose gels DNA was extracted from agarose gel sections by the following method. Gel

294

sections (1.5 ml) were slowly frozen and thawed 3 times and homogenized in a siliconized dounce homogenizer. 3 ml of TNE (20 mM Tris- HCI pH 7.5, 0.4 M NaCI, 5 mM EDTA)were added and the suspension rehomogenized. 3 ml of TNE-saturated phenol were added, the suspension was rehomogenized, allowed to sit at room temperature for 30 min, and centrifuged at 9000 rev./ rain in an HI~4 rotor. After removal of the aqueous phase the phenol and agarose were extracted 2 more times as described above with 3 ml TNE. The combined aqueous phases were extracted with TNE-saturated phenol and ethanol-precipitated. Approx. 50% of the DNA in the gel sections was recovered by this method.

Ligation and transformation Ligation of BglII restriction enzyme fragments into pBR322 plasmid DNA was performed by a modification of the method of Tanaka and Weisblum (1975). The conditions are described as follows. Plasmid pBR322 cut with BamHI restriction enzyme was diluted to 22 t~g/ml in the following solution: 30 mM Tris-HCI pH 7.5, 0.05 mM ATP, 12 mM MgCI2, 50 mM NaCI, 10 mM DTT. BglII restriction enzyme fragments isolated from agarose gels were added at a final concentration of 80 #g/ml; 0.3 units of T4 ligase were added, and the reaction incubated for 12 h at 12°C. The ligation of the yeast DNA fragments from the gel into the plasmid was monitored by agarose gel electrophoresis in a 1% agarose gel containing 90 mM Tris. borate pH 8.3, 3 mM EDTA (Greene et al., 1975). Competent ceils of K coil strain M94 (CGSH/5346) were prepared as described by Morrison (1977). Transformation of these cells with plasmid DNA was performed by a modification of the method of Tanaka and Weisblum (1975). The ligation reaction mix (0.05 ml) was added to 0.06 ml of 10 mM Tris. HCI pH 7.5, I mM EDTA, 20 mM NaCI, and 100 mM CaCI2 and 0.2 ml of competent cells were added to this solution. After incubation at 4°C for 30 rain the cells and DNA solution were incubated at 420(3 for 3 mhl with shaking. Then 2.7 wd of L-broth + 0.2% glucose + 0.1 mg/ml ampicillin were added to the cell suspension and incubated at 37°C with moderate shaking for 40 min. Aliquots (0.1 ml ffi 150 transformants) were spread on L-broth plates containing 0.1 mg/ml ampicillin and 0.2% glucose, and the plates incubated at 37°C for 20 h. The chilled plates were replica plated (Miller, 1972), onto L-broth + 0.2% glucose plates containing 0.02 rag/m! tetracycline, and these plates incubated for 15 h at 370(3. The original master plate (ampicillin) was superimposed onto the tetracycline plate and the colonies that did not grow on the tetracycline plates were picked onto new plates (L-broth + 0.2% glucose + 0.1 mg/ml ampicillin).

Screening of plasmid containing strains for the galactokinase gene The strains which were found to c o n ~ plasmid pBR322 with inserted yeast DNA (ampiciilin resista.~.,~tetracycline sensitive) were grown a t 37°C

295 for 15 h on L-broth plates containing 0.1 mg/ml ampicillin. Filters which contained plasmid DNA from each individual strain at the site of each colony were prepared on Whatman 540 paper by a modification of the method of Beckmann et al. (1977). To detect the clones containing the galactokinase gene, the filters were hybridized with ~25I-labelled galactokinase mRNA. 20 filters (6 X 8 cm, containing 48 colonies each) were placed in a "Seal-aMeal" bag and then 1 5 m l of 50% formamide, 4 × SSC, 0.5% SDS, 0.2-106 cpm/ml galactokinase mRNA were added to the bag. The bag was sealed and incubated at 41°C for 24 h. After hybridization the filters were washed twice with 50% formamide in 4 X SSC for 20 rain with 200 ml each time at room temperature, and three to four times with 2 × SSC, again for 20 rain each. The filters were dried and autoradiographed for 1 week.

Hybridization analysis of putative galactokinase containing plasmids To prepare plasmid DNA fixed to nitrocellulose filters (Cooper et al., 1974), 50 ng of plasmid DNA was added to 0.5 ml of 0.01 × SSC containing 20 ~g/ ml E. coli DNA. The DNA was denatured by heating at 98°C for 12 min and quickly chilled to 0°C. The solution was adjusted to 6 × SSC in a final volume of 2.0 ml, passed slowly through 25 mm nitrocellulose filters; the filters were washed and then baked. From each filter 5 mm diameter circles were punched out with a standard paper punch; each of these filters contained 3--4 ng of plasmid DNA. Competition hybridization of the ~2SI-labelled galactokinase mttNA to these immobilized plasmid DNAs was performed by placing each disc in 0.05 ml of hybridization solution (50% formamide, 0.8 M NaCI, 0.5% SDS, 50 mM Tris. HCI pH 7.5, 10 mM EDTA, 260 pg/ml induced or non-induced polysomal mRNA) in conical microfuge tubes. After pre-hybridization for 40 h at 41°(3, 10 ng of ~=SI-labelled galactokinase mRNA (0.2.106 cpm) in 0.01 ml of hybridization solution were added to each hybridization tube. After an addition. al incubation at 41°C for 40 h, the nitrocellulose filters were assayed for hybrid formation by a modification of the method of Cooper et al. (1974).

Assay of galactokinase activity Galactokinase assays were performed by a modification of the method of Schell and Wilson (1977) using [~4C] galactose with a specific activity of 10 000 cpm/nmole. 10 mM glucose was included in some assays to block phosphorylation of galactose by enzymes other than yeast galactokinase. Protein was measured by the method of Lowry et al. (1951) using bovine serum albumin for reference.

Preparation of mRNA Total polysomal mRNA from galactose-induced or uninduced yeast was prepared from 8. cerevisiae strain X108D as described by Schell and Wilson (1979). Total induced or uninduced mRNA was prepared by deproteinization

296 of a post-mitochondri~_ supematant extract of the same cells with chloroform: phenohisoamyl alcohol and chromatography on polyuridylic acid-linked sepharose.

Preparation of media M9 medium was prepared by adding the following chemicals to 1.0 liter of distilled water: 6 g Na2HPO4, 3 g KH2PO4, 0,5 g NaCI, I g NH4CI. After autoclaving the medium was made to I mM MgSO4 and 0.1 mM CaCI2 by the addition of appropriate amounts of sterile 1.0 M stock solutions. L-broth was prepared by adding 10 g of bacto-tryptone, 5 g NaCI, and 5 g yeast extract to 1.0 liter of water; plates contained 2.0% agar. Carbon sources were included at a final concentration of 0.2%.

Chemicals TIC13 (K and K Laboratories); Na~2SI (New England Nuclear; carrier free); KI (Mallinckrodt); Agarose (Seakem); Nitrocellulose sheets (Millipore); CsCI (Kawecki-Berylco); phenol (Mallinckrodt; distilled and stored at-20°C); formamide (Fisher; distilled in vacuo and stored at-20°C); DNA enzymes: BamHI, BglII, EcoRI, T4 ligase (New England Biolabs); ampiciUin (ParkeDavis; Amsfl-S); tetracycline (Sigma). All other chemicals were purchased from Sigma Chemical Co. or were of reagent grade purity.

Recombinant DNA procedures These experiments were carried out in strict compliance with the NIH Guidelines for research involving recombinant DNA molecules. The level of containment employed was P2 EK1 as prescribed by these guidelines. RESULTS

Identification of a restriction fragment containing the galactokinase gene Yeast DNA was digested with BglII restriction enzyme, subjected to agarose gel electrophoresis, and the resulting pattern of DNA fragments transferred to nitrocellulose filters as described in METHODS.These DNA filters were hybridized with ~2SI-labelled galactokinase mRNA and autoradiographed. Three discrete bands (A,B,C) were consistently detected in the BglII restriction enzyme digest. This is illustrated in Fig. 1 (slot 2) which is a typical autoradiograph of a nitrocellulose sheet containing a BglII digest hybridized with galactokinase mRNA. Due to the larger amount of DNA coding for rENA relative to single copy genes it was likely that at least one of these bands resulted from the hybridization of small amounts of rRNA which contaminates most RNA preparations. This conclusion was supported by preliminary Southern filter hybridization experiments with EcoRI di~ests of yeast DNA, where the galactokinase mRNA preparation hybridized to known rDNA restriction enzyme ~agments (Bell et al., 1977). Furthermore fragment C in the hybridization pattern in Fig. I has a molecular weight which is

297

Fig. 1. Hybridization of l~SI-labelled galactokinase m R N A to BglII restriction enzyme digests of yeast DNA. Yeast DNA was digested with BglII restriction enzyme, fractionated by agarose gel electrophoresis and transferred to nitrocellulose sheets. The DNA digests were hybridized with: Slot 1 = 10 ~g of BglH digested DNA hybridized with 10' cpm of '~sIlabelled total induced polysomal mRNA. Slot 2 = 10 ~g of BgllI digested DNA hybridized with 10' cpm of is'I-labelled galactokinase mRNA. Slot 3 = 10 ~g of BglII digested DNA hybridized with 10 s cpm of 'ssI-iabeiled yeast ribosomal RNA. Molecular weight values were calculated using EcoRI digested k-phage DNA that was run on the same gel as a reference.

identical to the size of the BglII restriction enzyme fragments containing the yeast rDNA (J. Szostak, personal communication). This hypothesis was confirmed by hybridizing BglII digests of yeast DNA on nitrocellulose sheets with 12SI-labelled yeast rRNA or total mRNA from non-induced yeast cells. The results of these hybridizations are illustrated in Fig. 1 (slots 1 and 3). It is clear that fragments A and C are probably rDNA fragments since they are the only bands in BglII digests that hybridize yeast rRNA and non-induced mRNA. Fragment B did not hybridize to either rRNA or non.induced mRNA and therefore it was likely that fragment B with a molecular weight of 3.7.106 was the BglII restriction fragment containing the galactokinase gene.

Partial purification of the galactokinase gene fragment In order to isolate preparative amounts of this DNA fragment, 400 ~g of yeast DNA were digested with 30 units of BglII restriction enzyme for 22 h

298

at 37°C in the solution described in METHODS. After ethanol precipitation the redissolved DNA was run on a preparative 1% agarose gel (4 slots; 0.6 cm X 4 cm) at 70 mA for 18 h; 2, DNA cut with EcoRI was run on the same gel as a molecular weight marker. Since the molecular weight of the putative galactokinase gene ¢on'-r~dning Bg/II restriction fragment was 3.7.10 ~, the area of the gel containing the restriction enzyme fragments of molecular weight 3.4-106 to 4.0-106 was cut from the gel and the DNA extracted from the agarose pieces as described in METHODS. The final yield of DNA from this partial purification procedure was 2 0 - 3 0 #g. Since 50% of the DNA was lost during manipulation, it can be estimak~l that this partially purified DNA fraction containing the galactokinase gene fragment had been enriched 10--20.fold for this specific fragment.

Generation o f a clone bank The DNA isolated from the gel was ligated into pBR322 plasmid DNA at the BamHI restriction enzyme site as described in METHODS. The ligation appeared to be greater than 50% complete as estimated by electrophoresis using a 1% agarose gel in Tris. borate buffer (Greene et al., 1977). The mixture of ligated plasmid and plasmid with inserted yeast DNA f~gments was used to transform £. coli strain M94 by the procedure described in METHODS. Selection for transformed cells containing pBR322 on plates containing 0.1 mg/nd ampicillin showed that 1.5-103 cells/ml were ampicillin-resistant out of a total of 1.7.10 s viable cells/ml indicating a frequency of 10 -s transformation to ampicillin resistance per cell with the plasmid preparation. The entire transformation mixture was plated out on 25 ampicillin plates (0.1 ml = 150 colonies/plate). When these plates were replica plated on tetracycline plates approx. 15% of the total ampicillin-resistant transformants were found to be tetracycline-sensitive. Since the yeast DNA fragments are inserted into the tetracycline-resistance gene on the plasmid (Rodriguez et al., 1977), these 15% were presumed to contain plasmid with yeast DNA ligated into it. From three individual transformations 1500 ampicillin-resistant/tetracyclinesensitive transformants were isolated. It was calculated from the purification factor of the galactokinase gene fragment by agarose gel electrophoresis, that there should be from 300--800 unique DNA fragments in this clone bank and therefore it should contain from 2--5 plasmid clones with the BglII restriction enzyme fragment containing the galactokinase gene. Detection o f plasmid containing the galactokinase gene Filters were prepared with DNA from each of the plasmid strains fixed at discrete locations on Whatman 540 paper as described in METHODS. After hybridization of these filters and autoradiography, only 6 colonies out of the 1500 transformants hybridized to 12SI-labelled galactokinase mRNA. Each of these strains were grown and plasmid DNA isolated as described by Meagher et al. (1977). After digestion of these plasmids with EcoRI and gel electrophoresis in 1% agarose none of the six plasmids appeared to have

299 identical band patterns. F u r t h e r m o r e , two of the six unique plasmids appeared to contain a piece of yeast D N A that was t o o small to be the 3.7.106 BglII fragment. The origin of these small plasmids is n o t known although cellular modification of the plasmid is a possible explanation. However f o u r of the plasmids contained yeast D N A fragments which might contain t h e galactokinase gene. Since it has been shown t h a t m R N A isolated from non-induced polysomes contained n o galactokinase m R N A (Schell and Wilson, 1979), t h e plasmid containing the yeast galactokinase gene could be identified by competition hybridization. Competition hybridizations of the ~2SI-labelled galactokinase m R N A t o t h e various plasmid DNAs fixed o n nitrocellulose filters were performed after pre-hybridization with an excess of unlabelled m R N A from both induced and non-induced y eas t cells as described in METHODS. The results are presented in Table I. As ex p ect ed the hybridization of the s2sI galactokinase m R N A to the various plasmid DNAs pre-hybridized with an excess (12 #g m R N A / 2 ng plasmid) of unlabelled induced m R N A is completely blocked. Hybridization of two of the plasmid DNAs (GB-4 and GB-7) was n o t blocked by an excess of non-induced m R N A while the rest of the plasmids were totally blocked. We conclude from this experiment t h a t the plasmids designated GB-4 and GB-7 contain a sequence homologous to a galactose induced m R N A since the hybridization of galactokinase m R N A to them cannot be blocked by unlabelled non.induced (i.e. non galactokinase-containing) m R N A . Furthermore, when these plasmids (labelled by nick translation [Rigby et al., 1977] ) were hybridized to BglII digests of whole yeast DNA on nitrocellulose filters both plasmids hybridized to t h e 3.7.106 dalton restriction fragment t h a t TABLE I HYBRIDIZATION OF GALACTOKINASE mRNA TO DIFFERENT PLASMID DNAs PRE-HYBRIDIZED WITH VARIOUS mRNA FRACTIONS Galactokinase mRNA (10 ng = 0.5 • 10' cpm z2sI) was hybridized to nitrocellulose filters containing 3 ng of the various plasmid DNAs as described in METHODS. The filters were first pre-hybridized with no unlabelled mRNA, 0.012 mg of unlabelled polysomal mRNA from non-induced yeast cells (+ NI mRNA), or 0.012 mg of polysomal mRNA from galactose induced yeast cells (+ I mRNA). Values are expressed both as the s2sI cpm of galactokinase mRNA hybridized or as the percentage of hybridization in the absence of pre-hybridization with unlabelled mRNA. Plasmid designation

z2sI cpm hybridized No pre-hybridization

+ NI mRNA (%) + I mRNA (%)

GB-2 GB-3 GB-4 GB-5 GB-6 GB-7

830 2120 S60 5100 750 350

10 (1) 23 (2) S40 (9S) 50 (1) 2"/ (4) 325 (93)

20 (2) 15 (1) 25 (3) 60 (1) 20 (3) 15 (5)

300 hybridized only galactokinase mRNA shown in Fig. 1. The plasmid GB-7 also contained an additional fragment of slightly smaller molecular weight; this fragment was "co-cloned" into a plasmid with the galactokinase fragment to produce the larger plasmid. This is supported by the fact that GB-4 DNA was resistant to BglH digestion, while GB-7 was cleaved by BglH to produce a linear DNA molecule. Thus GB-4 and GB-7 are the same plasmid except that GB-7 contains an additional BglIIfragment of DNA derived from an insertion of another individual DNA fragment into the plasmid molecule,

Expression of yeast galactokinase gene in E. coli Proof that the cloned fragment, which is homologous to a galactose-induced mRNA, does contain the functional galactokinase gene was provided by the following experiment. The GB-4 plasmid DNA was used to transform E. coli strain $165 which contains a deletion of the K coil galactose operon (Starlinger, pets. commun.). When extracts of this strain containing the GB-4 plasmid were analysed for galactokinase activity, a low but significant activity (Table II) was detected. Extracts of this strain containing the GB-4 plasmid exhibited a galactokinase activity of 2-10 -3 U./mg, while extracts of the same strain containing no plasmid or the pBR322 plasmid have an apparent activity of 1-10 -3 U./mg. When the same assays were performed in the presence of 10 mM glucose to block phosphorylation of galactose not catalysed by yeast galactokinase (Schell and Wilson, 1977), the GB-4 plasmid bearing strain still had an activity of 1.10 -3 U./mg while the control strains (with pBR322 only or with no plasmid) have galactokinase activity which is at the limit of detection of the g~actokinase assay procedure. This experiment shows that the cloned fragment contains a functional yeast galactokinase gene, which can be expressed at a low level in E. coll. TABLE H EXPRESSION OF GALACTOKINASE ACTIVITY IN VARIOUS PLASMID-CONTAINING STRAINS Ceils of E coli strain $165 containing the designated plasmids were grown in 10 ml of L-broth containing 0.2% galactose and 0.1 mg/ml ampicillin to A,00 am = 0.8. The cells were harvested, washed, and cell extracts prepared by sonication. Extracts were t h e n assayed for galactokinase activity in t h e presence and absence of 10 mM glucose in the assay mix. Activity levels in these strains is expressed in the uv;ts/mg protein • 103. Extract source S165 + 0 S165 + pBR322 8165 + GB-4

Galactokinase activity (U.Img • 10s.) no glucose

+ 10 mM glucose

1.0 ± 10% 1.0 ± 10% 2.0 ± 10%

0.05 ± 20% 0.05 ± 20% 1.0 ± 10%

301

DISCUSSION

The isolation of a plasmid containing the yeast galactokinase gene in an K coil host is the final step in the purification of the components of the expression of the galactokinase gene. The successful isolation of this plasmid was dependent on the sequential purification of galactokinase (Schell and Wilson, 1977), galactokinase antibody, and galactokinase mRNA (Schell and Wilson, 1979) since each of these purified components was necessary to produce an assay system for the galactokinase gene. This successful cloning experiment provides support for the general use of this sequential isolation method (protein -~ antibody -> mRNA -~ DNA) to clone and study many other eukaryotic genes and their components. Although the methods used to isolate the galactokinase gene are not new, the cloning of a specific gene whose mRNA represents less than 2% of the total cellular mRNA utilizing these methods has not been reported. The isolation of the galactokinase plasmid was dependent on the use of very high specific activity ~2SI-labelled purified galactokinase mRNA to specifically hybridize to the galactokinase gene fragment in BglII digests of the entire yeast genome, allowing subsequent partial purification of the fragment 10- to 20-fold. Although other restriction enzymes were investigated, the enzyme BglII was found to produce the largest DNA fragment (3.7-10 e) containing the galactokinase gene which was easily separated from ribosomal DNA. Due to its large size there was a possibility that this DNA fragment would also contain other parts of the galactose gene cluster in addition to the galactokinase gene. The partially purified DNA preparation was ligated in the BamHI site of the plasmid pBR322 inserting the yeast DNA into the plasmid gene for tetracycline resistance. This allowed elimination of the large number of transformants which contain re-ligated plasmid without inserted yeast DNA (85%). In situ colony hybridization to 1500 of these clones with 12SI-labelled galactokinase mRNA detected six individu"~ plasmids that strongly hybridized the galactokinase mRNA preparation. However, for four of the six plasmids this hybridization could be completely blocked by an excess of unlabelled mRNA from cells that were not induced for the synthesis of galactokinase. Two different plasmids (GB-4 and GB-7) hybridized galactokinase mRNA in the presence of the same amount of non-induced mRNA that completely blocked the hybridization by the other plasmids. It can be concluded from these results that these two plasmids contain a sequence coding for a galactose inducible mRNA. The identity of the other four plasmids that hybridize galactokinase mRNA but are blocked by an excess of non-induced mRNA is not known. They could contain sequences of DNA that are homologous to the non-coding portion of the galactokinase mRNA, or represent DNAs homologous to contaminating mRNAs in the galactokinase mRNA preparation. Hybridization experiments with the galactokinase plasmids showed that a large excess of total cellular mRNA or total polysomal mRNA from uninduced

302

cells does not block the hybridization of galactokin~.se mRNA to the plas~d. This suggests that there is an tmdetectable amount of galactokinase mRNA in yeast cells that are not e x p ~ ~ galactose. 'i~nerefore it seems that the induction of gal~to~nase synthesis (and probably the entire galactose gene cluster) ~ ~ a c t o ~ re~dlt ~0fthe induction of the ~nthesis of mRNA's c ~ ~ for these p r o t e ~ . ~ r i p t i o n ~ Control therefore appears to ~ the predominant m e c h ~ by which the induction of galactose enzyme ~ n thesis is mediated. ~ ~~ ...... The proof that one of the isolated pl~mids contains a ~nctional yeast galactokinase gene was provided by the experiments thatdemonstrated the ~ expression of the yeast galactoYm~ geve in £. coll. An E. ¢oli s ~ , deleted for its galactose operon, carrying the g~actokinase plasmid exhibited a galacto. kinase activity of 0.001 U./mg when assayed under conditions that are specific for yeast galactokinase. Galactokinase assays of the same strain conta/ning either no plasmid or only pBR322 pla~mid showed no apparent galactokinase activity under the same conditions. Although this level of expression of the yeast galactokinase gene in E. ¢oli is low, the activity is real and therefore strongly suggests that the isolated plasmid does indeed contain a ftmcdonal yeast galactokinase gene. Therefore the yeast galactokinase gene c a n ~ added to the list of yeast genes which have been cloned and shown t o be expressed in E. coil (Clarke and Carbon, 1978), ACKNOWLEDGEMENTS

The authors would like to thank Drs. J. Calve, R. Wu, R. Rothstein, S. Wessler, and J. Szostak for their advice, comments, and assistance. This research was supported in part by Grant PCM 77-15487 from the National Science Foundation. REFERENCES Beekmann, J.8., Johnson, P.F. and Abelson, J., Cloning of yeast transfer RNA genes in Escherichia coli, Science, 196 (1977) 205--208. Bell, G.L, DeGennaro, L.J., Gelfand, D.I-L, Bishop, R.J., Valenzuela, P. and Rutter, W.J., Ribosomal genes of 8accharomyces cerevisiae, L Physical map of the repeatin~ unit and location of the regions coding for 58, 5.88: 188, and 258 ribosomal RNAs, J. Biol. Chem., 252 (1977) 8118--8125. Burton, K., Determination of DNA concentration with diphenylamine, in Groasman, L. and Moldave, K. (Eds.), Methods in Enzymology, Vol. XHB, Academic Press, New York, 1968, pp. 163--166. Clarke, L. and Carbon, J., Expression of cloned yeast DNA in Escherichia co/i: specific eomplementation of argininosuceinate lyase (arglI) mutations, J. Mol. Biol., 120 (1978) 517--532. Cooper, T.G., Whitney, P. and Magasanik, B., Reaction of/at-specific ribonucleic acid from Escherichia coli with la¢ deoxyribonucleic acid, J, Biol. Chem., 249(1974) 6548-'6555. CrYer, D., Eecleshall, R .~and M ~ u r , J ' Isolation Of Yeast DNA in Prescott, D, (Ed.), Methods in Cell Biology, VOL XII, Chapter 3, Academic Press, New York; r1975, pp. 89--44.

303 Greene, P.M., Betlach, M., Goodman, J., and Boyer, H.W., Gel electrophoresis of plasmid DNA, in Wickner, R.B. (Ed.), Methods in Molecular Biology, Vol. 7, Marcel Dekker, New York, 1974, pp. 87--111. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265--275. Meagher, R.B., Tait, R.C., Betlach, M. and Boyer, H.W., Protein expression in E. coli minicells by recombinant plasmids, Cell, 10 (1977) 521--536. Miller, J.H., Replica plating, in Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, New York, 1972, pp. 56--59. Morrison, D.A., Transformation in Escherichia coli: cryogenic preservation of competent cells, J. Bacteriol., 132 (1977) 349--351. Prensky, W., Radiodination of DNA and RNA in Prescott, D. (Ed.), Methods in Cell Biology, Vol. XIII, Chapter 7, Academic Press, New York, 1976, pp. 122--152. Rigby, P.W.J., Dreckmann, M., Rhodes, C. and Berg, P., Labeling deoxyribonucleic acid to high spbcific activity in vitro by nick translation with DNA polymerase I, J. Mol. Biol., 113 (1977) 237--251. Rodriguez, R.L., Tait, R., Shine, J., Bolivar, F., Heyneker, H., Betlach, M. and Boyer, H.W., Characterization of tetracycline and ampicillin resistant plasmid cloning vehicles, in Scott, W.A. and Werner, R. (Eds.), Molecular Cloning of Recombinant DNA (Miami Winter Symposia; Vol. 13) Academic Press, New York, 1977, pp. 73--84. Schell, M.A. and Wilson, D.B., Purification and properties of galactokinase from 8accharomyces ¢erevisme, J. Biol. Chem., 252 (1977) 1162--1166. Schell, M.A., and Wilson, D.B., Purification of galactokinase mR~A from S. cerevisiae by indirect immunoprecipitation, J. Biol. Chem. (1979) in press. Southern, E.M., Detection of specific sequences among DNA fragments separated by gel electrophoresis, J. Mol. Biol., 98 (1975) 503--517. Tanaka, T. and Weisblum, B., Construction of a colicin E1-R factor composite plasmid in vitro: means for amplification of deoxyribonucleic acid, J. Bacteriol., 121 (1975) 354--362. Wu, R., Jay, E., and Roychoudhury, R., Nucleotide sequence analysis of DNA, in Busch, H. (Ed.), Methods in Cancer Research, Vol. XII, Chapter III, Academic Press, New York, 1976, pp. 88--177. Communicated by F.E. Young.