Current Genetics 2, 223-228 (1980)

~

~

~

© by Springer-Verlag 1980

Transcriptional Units of GAL Genes in Saccharomyces cerevisiae Determined by Ultraviolet Light Mapping Takashi Segawa* and Toshio Fukasawa Research Unit for MolecularGenetics,Keio UniversitySchool of Medicine,35, ShinanomachiShinjuku,Tokyo, Japan 160

Summary. The size of the transcriptional unit of the structural genes for three galactose-metabolizing enzymes which form a cluster on chromosome II in Saccharomyces cerevisiae was studied by the ultraviolet light (UV)mapping technique. Thus the size of the primary transcripts of GAL7 for galactose-l-phosphate uridylyl transferase, GALIO for uridine diphosphoglucose 4-epimerase, or GALl for galactokinase were estimated to be 0.81 x 106, 1.1 x 106, or 1.3 x 106 respectively. In the light of these data together with the known directions of transcription of the genes, we concluded that each of three genes was transcribed from its own promoter.

Key words: Transcriptional Units - GAL Genes Saccharomyces cerevisiae - UV mapping.

Introduction In Saccharomyces cerevisiae galactose is metabolized to glucose-l.phosphate through the so-called Leloir pathway involving three enzymes, galactokinase, galactose-1phosphate uridylyl transferase, and uridine diphosphoglucose-4-epimerase (hereafter called kinase, transferase, and epimerase, respectively, and gal enzymes, collectively) as in other organisms including Escherichia coli and man (Kalckar 1958). The structural genes for the gal enzymes of E. coli form a polycistronic operon, a gene cluster whose expression is under the control of a promoter and operator and is negatively regulated by a specific repressor (Buttin offprint requests to: T. Fukasawa * l~esent address: Department of BiologicalSciences,Columbia University, New York, N. Y 10027, USA

1963; Saedler et al. 1968). The genes of S. cerevisiae, GALl, GAL 7, and GALIO (collectively called gal genes) coding for kinase, trahsferase, and epimerase, respectively, also form a gene cluster which is located on chromo. some II. The expression of three genes is coordinately regulated either negatively or positively by two genes, GALi (GALSO) or GAL4, respectively (Mortimer and Hawthorne 1969), probably at the transcriptional level (see Hopper et al. 1978). These regulatory genes are linked neither to the gal genes nor to each other. There is no evidence, however, that the gal genes of yeast are transcribed as a unit from a single promoter. Instead, recent experiments with ~ phage-cloned gal genes suggest that both GALIO and GAL 7 are transcribed toward the centromere from one strand of DNA and GALl from the other strand (St. John and Davis 1979 and personal communication). It is not yet clear whether or not GAL7 and GALIO are transcribed separately from each other. Polyadenylated mRNA molecules for kinase and transferase which are identified by template activity in a cell-free translation system were shown to be distinct to and separable from each other by size (Hopper and Rowe 1978). However, experiments dealing with the functional mRNA do not rule out the possibility that the primary transcript is polycistronic and processed into functional monocistronic mRNAs. Therefore, we feel it is important to know the size of the primary transcripts of gal genes to elucidate the molecular mechanism of the regulation of their expression. In this work we determined the size of the transcriptional units of the three gal genes by the UV (ultraviolet light) mapping technique; a simple method which has proved useful for determining the size of the primary transcripts of eukaryotic as well as prokaryotic genes. For example, sizes of the transcriptional units of the genes coding for the histone proteins of HeLa cells (Hackett et al. 1978) or the heat-shock proteins of 0172-8083/80/0002/0223/$01.20

224

T. Segawa and T. Fukasawa: Transcriptional Units of Yeast Galactose Genes

0

1

2

UV Dose (sec) t, 6

Materials and Methods

|0

~

10-1 10.2

>

'7, ;10. 4 oo

10- ~ 10-6

10 Fig. 1. UV (ultraviolet light)-survival curves for wild-type and sensitive strains. Cells of the strains 108-3C a trpl ural, XS774-6A a tad1-1 rnal gal2 or ST-3A were plated on YEPD agar (1% Bacto yeast extract, 2% Polypeptone, 2% glucose and 2% Bacto agar) and exposed to UV for indicated periods at a dose rate of 28 erg/mm2/s. Colonies were scored after incubation for 3 days at 30°C in the dark. (o) 108-3C, (e) XS774-6A, (A) ST1-3A

Drosophila melanogaster (Carlson and Pettijohn 1979) were successfully determined with the primary transcript of rRNA gene of the respective organism as the standard. The rationale of the method is as follows: UV-irradiation of cells introduces lesions in DNA, the most important of which are thymine dimers, which cause premature termination of transcription (Michalke and Bremer 1969; Sauerbier et al. 1970). The UV-sensitivity of a given gene is expressed by the equation, R ( k , d) = e - k ' d , where R is the fraction of intact gene copies (including leader sequences) which escaped UV irradiation at a dose of d, and k is the inactivation coefficient of the gene. The value of k is directly proportional to the distance between the promoter a n n the end of the gene, that is, the size of the primary transcript (Sauerbier et al. 1970). R is proportional to the rate of synthesis of the primary product of the gene. Thus we determined the value of R for the respective gal gene b y measuring the rate of induced synthesis of the gal enzyme in a cell population which had been irradiated at various doses of UV. In a parallel experiment, the rate of 25S rRNA synthesis was also determined. Based on the k n o w n molecular weight of the primary transcript for rRNA, we calculated the absolute value of the primary transcript for the gal genes. In the light of the present results together with the directions of transcription of the genes elucidated by St. John and Davis (see above), we conclude that each of three gal genes is primarily transcribed as a monocistronic RNA molecule.

Yeast Strains. Saceharornyces eerevisiae Strain ST1-3A, a/~ radl-1/radl-1, a galaetose-fermenting diploid strain defective in excision of UV-induced pyrimidine dimers (Urtrau et al. 1971; Waters and Moustacehi 1974; Prakash 1975) was used throughout the present work to avoid dark repair. This strain was constructed by mating two strains XS774-6A, a tad1-1 rnal gal2 (Nakai and Matsumoto 1967) and N77-1A, aural leul ade2 and/or ade6 (our stock). The strain ST-3A is as sensitive as its parental strain SX 774-6A to UV but the survival curve of the former strain exhibited a shoulder characteristic of diploid cells (Nakai and Matsumoto 1967) (Fig. 1). Media and Chemicals. YM5L was the same as YM-5 (Hartwell 1967) except that sodium lactate (2%) was substituted for glucose and the pH was adjusted to 6.3. Two times YM5L had the same composition as YM5L except that the concentration of each component was doubled. UDP-glueose dehydrogenase was purified according to Wilson (1965). Other chemicals and indicator enzymes were as described previously (Segawa and Fukasawa 1979; Fukasawa et al. 1980). Irradiation by UV. Log-phase cells grown in 500 ml YM5L at 30 °C overnight were harvested, washed with H20, and suspended in H20 at a cell concentration giving an absorbance of A660ran = 1.0. The cell suspension was divided into 25 ml portions placed in glass petri dishes (9 cm in diameter), and irradiated by UV for the times indicated, with stirring, at a distance of 29 em from a Toshiba GL-15 (15 Watt) germicidal lamp. (The dose rate at the surface of cell suspension was 28 erg/mm2/s.) Each sample of the irradiated cell suspension was divided into 23 ml and 1 ml portions. All operations hereafter until the addition of eycloheximide were carried out in the dark to avoid the photo reactivation. Induction of gal Enzymes. Twenty-three milliliters of irradiated ceU suspension were centrifuged. The cells were suspended in 50 ml of YM5L containing 0.5% galactose in 200 ml Erlenmyer's flasks arid immediately incubated at 30 °C with aeration. At 30 and 60 rain of incubation duplicate 5 ml portions were transferred to duplicate centrifuge tubes containing 0.5 ml of 0.1% cycloheximide and chilled on ice. Cells were pelleted and resuspended in 1 ml of lysis buffer containing 0.1 M sodium phosphate (pH 7.5), 0.25% 2-mercaptoethanol and 0.1 mg of Zymolyase 60,000 (Kirin Brewery) per ml (Zymolyase was dissolved in advance in 2 M sorbitol at a concentration of 3 mg/ml and diluted in the buffer) and incubated at 25 °C for 1 h, Cell debris were removed by centrifugation and the enzyme activities in the supernatant fraction were determined as described below. Determination of Enzyme Activities. Kinase was assayed according to Nogi et al. (1977). Transferase (Segawa and Fukasawa 1979) and epirnerase (Fukasawa et al. 1980) were assayed by the two-step methods described previously. Labeling and Extraction o f Ribosomal RNA. One milliliter of irradiated cell suspension was combined with 1 mi of 2 x YM5L containing 4% glucose and incubated at 30 °C. After 10 rain, 10 #C of [5-aH]-uracil (25 C/m tool, Radinchemical Centre) was added and incubation was continued for another 50 rain. Cold uracil was added to the culture at a concentration of 1 mg/ml and the culture was further incubated for 30 min to allow the processing of precursor RNA to mature rRNA molemales. Incubation was terminated upon the addition of cyclohexirnide (0.1 mg/ml) and the cells were kept frozen at -80 °C

T. Segawa and T. Fukasawa: Transcriptional Units of Yeast Galactose Genes

225

Sedimentation Analysis of rRNA. RNA was dissolved in 0.1 ml of sedimentation buffer containing 0.02 M Tris/HC1 (pH 7.4), 0.1 M NaC1, 5 mM EDTA and 0.5% sodium dodeeyl sulfate and layered on the top of a 5%-30% linear sucrose gradient in the sedimentation buffer. After centrffngation for 12 h at 24,000 rpm at 15 °C in a Beckman SW50.1 rotor, fractions were collected directly onto 3MM filters (25 mm in diameter, Whatman) from the bottom of the gradient. Filters were washed with 5% TCA and ethanol/ether (1:1). The radioactivity was determined in a Beckman liquid scintillation counter.

2

(:3 X

i

I

Results

Effect of UV Irradiation on Induced Synthesis of gal Enzymes. The synthesis of gal enzymes in S. cerevisiae

2

GALIO > GAL7. The product of the GAL4 gene, positively regulating the expression of gal genes (Mortimer and Hawthorne 1969), has recently been suggested to be synthesized constitutively, and also to be metabolically stable (Matsumoto et al. 1978; Petlman and Hopper 1979). Therefore, UV lesions in GAL4 would not influence the observed UV sensitivity of gal enzyme synthesis in the present experiments.

Absolute Sizes of Transcriptional Units of gal Genes. To calculate the absolute size of transcriptional unit

226

T. Segawa and T. Fukasawa: Transcriptional Units of Yeast Galactose Genes

0 100

UV Dose (sec) 16 24 32

8 •

,

o

,

.

,

40 ,

Fig. 3. Effect of UV (ultraviolet light) irradiation on the synthesis of gal enzymes and rRNAs. Increase of enzyme activities between 30 rain and 60 min after the start of induction by galaetose were determined as described in Materials and Methods. Radioactivities under the peaks of 18S and 25S rRNA were also determined as described in Materials and Methods. Values of 31-Icounting were corrected by subtracting the spillover of 14Ccounts and divided by the values of 14C.counts" The ratios of 3H/14C were normalized to the values of unirradiated sample as unity. Values of each enzyme or of rRNA at the indicated dose obtained in a typical experiment were plotted in logarithmic scale against the UV dose. Lines were drawn by the method of least-square omitting the values at the dose of 0 s (in the case of enzyme activities) or 20 s (in the case of rRNAs). (o) kinase, ( i ) epimerase, (A) transferase, () 25S rRNA, (*) 18S rRNA

A

r-

I0

Table l(a). Relative rates of synthesis Experiment

25 S rRNA

Kinase

Epimerase

Transferase

1

1.00 (-0.0914)

0.338 (-0.0309)

0.327 (-0.0299)

0.210 (-0.0192)

2

1.00 (-0.0746)

0.408 (-0.0304)

0.346 (-0.0258)

0.269 (-0.0201)

3

1.00 (-0.0742)

0.442 (-0.0328)

0.411 (-0.0305)

0.265 (-0.0197)

4

1.00 (-0.0638)

0.555 (-0.0354)

0.513 (-0.0327)

0.353 (-0.0225)

5

1.00 (-0.0709)

0.653 (-0.0463)

0.436 (-0.0309)

0.353 (-0.0250)

mean -+ S.D.

1.00

0.479 +- 0.125

0.407 + 0.075

0.290 -+ 0.062

Slopes of straight lines (s - 1 ) obtained as described for Fig. 2 are given in parentheses Table l(b). Molecular weight of products from gal genes

Primary transcript a

Poly(A)-RNA

Coding sequences b

Number of amino acid residues per molecule or subunit (MW)c

1.3 -+ 0.35

0.76 d 0.6 e

0.50

520 (58,000) f

GALIO (epimerase)

1.1 -+ 0.21

-

0.76

790 (91,500)g

GAL7 (transferase)

0.81 +- 0.17

0.61 d

0.34

350 (43,000) h

Gene (enzyme)

GALl (kinase)

a b c d e f g h

RNA (MW x 10 - 6 )

See the text Deduced from the size of the respective enzyme molecule Kinase has no subunit. Transferase and epirnerase consist of two identical subunits respectively Hopper and Rowe (1978) Schell and Wilson (1979) Schell and Wilson (1977) Fukasawaet a t (1980) Segawa and Fukasawa (1979)

T. Segawaand T. Fukasawa: Transcriptional Units of Yeast Galactose Genes a.

d.

~ i

,~

4O

ID

g , 20

~

'o 0 ~

b. ~

.

. . . . . $

e.

~

Q5

~

'

/ n k)

>,l( >

E

U

t..)

r6 ,

i

T

~

,

0

c. ~,

,

~

l

~

,

"

f.

4,

q.)

I

1.5 ID O5 10

20 Fraction

10

0

Number

Fig. 4. Sedimentation patterns of RNA labeled after the irradiation by UV (ultraviolet light). Cells of strain ST1-3A were grown, exposed to UV for (a) 0 s, (b) 4 s, (c) 8 s, (d) 12 s, (e) 16 s, and (f) 20 s, and labeled with [3HI-uracilat 5/~C/ml in YM5Lcontaining 2% glucose during 10 and 60 rain of incubation, chased by addition of cold uracil (100 ~g/ml). RNA was extracted and analyzed in the sucrose density gradient as described in Materials and Methods. Values of 3H-count were corrected by subtracting the spiltover (41.3%) of 14C.count. Arrowsindicate the positions of 25S rRNA (left) and 18S rRNA (right). (o) 3 H . a c t i v i t y , ( e ) 14C_aetivity

227

in unirradiated cells was plotted in a logarithmic scale against the UV dose, straight lines were obtained (Fig. 3). No RNA sedimenting faster than 25S was observed in the cells irradiated by UV at any dose, indicating that all rRNA precursors were processed to mature RNA (Fig. 4). We assume that 25S rRNA is cut exclusively from the primary transcript of the full length. Based on the slope for 25S rRNA, the relative slopes for kinase, epimerase, and transferase were estimated to be -0.479, -0.409, and -0.290, respectively (Table l(a)). Using the value of 2.8 x 106 as the size of rRNA transcript (Nikolaev et al. 1979) we calculated the molecular weight of the primary transcript for kinase (GALl), epimerase (GALIO), and transferase (GAL7) to be 1.3 x 106, 1.1 x 106, and 0.81 x 106, respectively. A summary of the results is shown in Table l(b), together with the published sizes of gal enzymes and also poly(A)-gal mRNA. Note that the sizes of the primary transcripts of the gal genes, especially for GALl and GAL 7 were considerably larger than those of the coding sequences. Bassel and Mortimer (1971) determined the order of the gal genes by the genetic analysis to be centromereGAL7-GALIO-GAL1. Most recently, St. John and Davis have suggested, in the transfer hybridization experiment with separated DNA strands of gal genes cloned in k phage as probes, that GAL 7 and GALI O are transcribed toward the centromere, and GALl in the other direction (1979 and personal communication). In other words, GALl is transcribed as a monocistronic transcript. In the light of the present experiments, we now know that GAL7 and GALIO are also transcribed monocistronically, because the transcript of GAL 7 was smaller than that of GALIO.

Discussion

for each gal gene we measured the sensitivity of synthesis of 25S rRNA as a standard. In S. cerevisiae, rRNA is transcribed as a large precursor consisting of three rRNA species and spacer sequences. The molecular weight of the primary transcript of rRNA has been reported to be 2.5 to 2.8 x 106 (35S to 37S) (Udem and Warner 1972; Nath and Bollon 1977; Nikolaev et al. 1979), in which mature rRNA sequences are located in the order of 5'spacer-18S-58S-25S-spacer-3' (Nath and Bollon 1977), Assuming that the processing to 25S rRNA takes place after the completion of transcription, the sensitivity of 25S rRNA synthesis to UV irradiation should represent the sensitivity of synthesis of its 37S primary transcript. Thus RNA in irradiated cells was labeled with [3H]-uracil in the dark, extracted, and analyzed in a sucrose density gradient (Fig. 4). When the ratio of the relative amount of [3H]-rRNA in irradiated cells to that of [14C]-rRNA

The UV mapping technique in our experiments is based on several assumptions: (1) In yeast premature termination of transcription by RNA polymerase II at the site of UV lesion occurs as efficiently as that of transcription by RNA polymerase I. This has been shown to be the case in mammalian cells (Hackett and Sauerbier 1975; Giorno and Sauerbier 1976). (2) The initiation frequency of transcription by the two RNA polymerases does not change after UV irradiation as shown in mammalian cells (Hackett and Sauerbier 1975). (3) UV irradiation has no effect on the stability of enzyme proteins. In a preliminary experiment cells grown in a medium containing galactose were irradiated by UV as described in Materials and Methods and transferred to a medium containing glucose to repress de novo synthesis of gal enzymes. The levels of the gal enzymes were constant for at least 1 h, suggesting that UV irradiation had neither direct nor indirect effects on the stability of the

228

T. Segawa and T. Fukasawa: Transcriptional Units of Yeast Galactose Genes

enzymes in our experimental conditions (data not shown). (4) F o u r bases are distributed uniformly throughout the gene in question. Recently many genes have been cloned from S. cerevisiae and the primary structure of some genes including non-coding sequences have been elucidated. Some o f the gene have been demonstrated to contain pyrimidine-rich regions upstream from the starting point o f the coding sequences which are suggested to be necessary for transcription (Holland and Holland 1979; Smith et al. 1979; Montgomery et al., 1980). If this is also true for gal genes, and not for r R N A genes, the size o f the primary transcripts from gal genes could be overestimated b y this UV mapping technique. Despite these assumptions, we believe that the present experiment gave approximate sizes o f the primary transcripts o f gal genes in S. cerevisiae. These results therefore would be helpful in experiments designed to obtain further insight into the mechanism o f expression o f gal genes in this organism, for instance, in identification of the primary transcripts of these genes.

Acknowledgments. We thank Dr. S. Nakai for supplying the radl mutant and Dr. T. St, John for communicating the data before publication. Our thanks are also due to Dr. P. V. Venkov for his advice in the analysis of rRNA. This work was supported in part by a grant (No. 410712) from the Ministry of Education, Culture and Science of Japan.

References Bassel J, Mortimer B (1971) J Bacteriol 108:179-183 Buttin G (1963) J Mol Biol 7:183-205 Carlson J O, Pettijohn D E (1979) J Mol Biol 132:141-161 Fukasawa T, Obonai K, Segawa T, Nogi Y (1980) J Biol Chem 255:2705-2707 Giorno R, Sauerbier W (1976) Cell 9:775-783 Hackett P B, Sauerbier W (1925) J Mol Bio191:235-256

Hackett P B, Traub P, GaUwith D (1978) J Mol Biol 126:619635 Hartwell L H (1967) J Baeteriol 93:1162-1670 Holland J P, Holland M J (1979) J Biol Chem 254:9839-9845 Hopper J E, Broach J R, Rowe L B (1978) Proc Nat Acad Sei USA 75:2878-2882 Hopper J E, Rowe L B (1978) J Biol Chem 253:7566-7569 Kalekar H M (1958) Adv Enzymol 20:111-134 Matsumoto K, Toh-e A, Oshirna Y (1978) J Baeteriol 134: 446-457 Michalke H, Bremer H (1969) J Mol Biol 41:1-23 Montgomery D L, Leung D W, Smith M, Shalit P, Faye G, Hall B D (1980) Proc Nat Acad Sei USA 77:541-545 Mortimer R K, Hawthorne D C (1969) Yeast genetics, In: Rose A H, Harrison J S (eds) The yeasts, vol 1. New York, Academic Press, p 385 Nakai S, Matsumoto S (1967) Mutat Res 4:129-136 Nath K, Bollon A P (1977) J BiN Chem 252:6562-6571 Nikolaev N, Georgiev O I, Venkov P V, Hadjiolov A A (1979) J Mol Biol 127:297-308 Nogi Y, Matsumoto K, Toh-e A, Oshima Y (1977) Mol Gen Genet 152:137-144 Perlman D, Hopper J E (1979) Cell 16:89-95 Prakash L (1975) J Mol Biol 98:781-795 Saedler H, Gullon A, Fiethen L, Starlinger P (1968) Mol Gen Genet 102:79-88 Sauerbier W, Millette R L, Haekett P B (1970) Biochim Biophys Acta 209:368-386 Schell M A, Wilson D B (1977) J Biol Chem 252:1162-1166 Schell M A, Wilson D B (1979) J Biol Chem 254:3531-3536 Segawa T, Fukasawa T (1979) J Biol Chem 254:10707-10709 Smith M, Leung D W, Gillam S, Astell C R, Montgomery D L, Halt B D (1979) Cell 16:753-761 St John T P, Davis R W (1979) Cell 16:443-452 Udem S A, Warner J R (1972) J Mol Biol 65:227-242 Urtrau P, Wheatcroft R, Cox B S (1971) Mot Gen Genet 113: 359-362 Waters R, Moustacchi E (1974) Biochim Biophys Acta 353: 407-419 Wilson D (1965) Anal Biochem 10:472-478 Communicated b y J.-M. Wiame Received August 6, 1980

Transcriptional units of GAL genes in Saccharomyces cerevisiae determined by ultraviolet light mapping.

The size of the transcriptional unit of the structural genes for three galactose-metabolizing enzymes which form a cluster on chromosome II in Sacchar...
557KB Sizes 0 Downloads 0 Views