Gene, 123 (1993) 131-136

0 1993 Elsevier Science Publishers B.V. All rights reserved. 0378-l 119/93/$06.00 131

GENE 06843

TATA box mutations in the Schizosaccharomyces pombe nmtl promoter affect transcription efficiency but not the transcription start point or thiamine repressibility

(Recombinant DNA; transcriptional regulation; expression vectors; fission yeast)

Gabriele Basi”, Elisabeth Schmidb and Kinsey Maundrellb

a European Molecular Biology Laboratory, D-6900 Heidelberg, Germany. Tel. (49-6221) 387357; and bGlaxo Institute, I228 Geneva, Switzerland

Received by J.K.C. Knowles: 14 April 1992; Accepted: 16 June 1992; Received at publishers: 11 September 1992

SUMMARY

The nmtl gene of Schizosaccharomyces pombe is highly expressed and subject to transcriptional repression by thiamine.

The nmtl promoter, in common with other strong promoters in this organism, contains a canonical sequence element,

5’-ATATATAAA, located 25 bp upstream from the transcription start point (tsp). We have made stepwise deletions of

the TATA box and quantitated the effects of the mutations by assaying the expression of the chloramphenicol acetyl-

transferase (CAT)-encoding gene (cat) cloned downstream. Our results demonstrate that progressive truncation of the

TATA box results in a concomitant decrease in promoter strength as judged both by the loss of CAT activity in cell

extracts and by a reduction in the steady-state level of cat mRNA. Both the induced level and the residual, repressed

level of expression observed in the presence of thiamine are similarly down-regulated. On the other hand, even in the

most extreme mutant, the tsp is unaffected, suggesting that other elements in the nmtl promoter are important in

determination of the tsp. The properties of the modified promoters have made them useful for extending the range of

the PREP inducible expression vectors.

INTRODUCTION

Functional analysis of genes transcribed by RNA poly-

merase II has revealed a number of important regulatory

elements upstream from the tsp. A typical mammalian

class-II promoter includes enhancers and UPE involved

in regulating the efficiency and specificity of transcription,

Correspondence to: Dr. K. Maundrell, Glaxo Institute, 14 Chemin des

Aulx, 1228 Geneva, Switzerland. Tel. (41-22) 709 66 66; Fax (41-22) 794

69 65.

Abbreviations: acCm, acetylated Cm; bp, base pair(s); BSA, bovine

serum albumin; CAT, Cm acetyltransferase; car, gene encoding CAT;

Cm, chloramphenicol; kb, kilobase or 1000 bp; nt, nucleotide(s); oligo,

oligodeoxyribonucleotide; S., Saccharomyces; Sz., Schizosaccharomyces; tsp, transcription start point(s); UPE, upstream promoter element(s); wt,

wild type.

together with the TATA element which provides the re-

cognition site for the transcription factor TFIID and

which is important both in regulating the rate of tran-

scription and in specifying the tsp, typically located

25-30 bp further downstream (Benoist et al., 1980; Dynan

and Tijan, 1985; Myers et al., 1986). Interestingly, a group

of promoters exists in which the TATA box appears not

to specify the tsp but only to control the efficiency of

transcription (Hen et al., 1982; Dierks et al., 1983; Wu

et al., 1987; Jones et al., 1988). Yet a third group of class-

II genes contains TATA-less promoters in which the spec-

ificity of the transcription initiation is regulated by other

elements either upstream from (Blake et al., 1990) or at

(Smale and Baltimore, 1989) the tsp. Studies on transcriptional regulation in S. cerevisiae

have revealed considerable similarity in promoter struc-

ture both in the proximal elements and somewhat less

132

expectedly in the UPE (reviewed in Struhl, 1987; Guar-

ente and Bermingham-McDonogh, 1992). In S. cerevisiae, however, transcription does not initiate at a strictly de-

fined distance from the TATA box but rather within a

‘window’ ranging from 40 to 120 bp downstream, and

additional initiator elements have been shown to be im-

portant for accurate tsp determination (Chen and Struhl,

1985; Hahn et al., 1985; McNeil and Smith, 1985; Nagawa

and Fink, 1985). Recent evidence has shown that even in

the presence of these elements, correct spacing from the

TATA box is required for efficent transcription initiation

(Healy and Zitomer, 1990; Furter-Graves and Hall, 1990).

We are interested in transcriptional regulation in fis-

sion yeast. In this organism, the canonical TATA se-

quence is present in highly expressed promoters, and as

in metazoans, the tsp normally occurs within a tightly

defined region 25-30 bp downstream (Russell, 1983;

1985). The result of deleting the TATA box from the adh promoter suggests a functional role of this element in Sz.

polnbe (Furter-Graves and Hall, 1990).

We have recently described a thiamine-repressible gene,

nmtl, which in cells grown in the absence of thiamine is

among the most highly expressed genes. We showed that

the nmtl promoter contains the sequence 5’-ATATA-

TAAA located 25 bp upstream from the tsp. This is the

first well characterized transcriptionally regulated system

to be described in fission yeast, and it thus provides an

opportunity to study the factors which act in cis and trans to regulate transcription of Sz. pombe genes. In the pre-

sent work we have studied the role of the nmtl TATA box

and the effects of mutating this sequence by monitoring

the expression of the bacterial cut reporter gene cloned

downstream. We report that the TATA element is cru-

cially important in setting the level of transcription but

does not affect the thiamine repressibility nor the tsp.

EXPERIMENTAL AND DISCUSSION

(a) TATA box mutations affect transcription efficiency of

PREP-cat

The pREPl-cat plasmid used in these studies consists

of the bacterial cat gene fused to the thiamine-repressible

nmtl promoter via genetically engineered Ndel sites at

the ATG start codon (Maundrell, 1989). The TATA box

of this conslruct was mutated by site-directed mutagene-

sis (Kunkel, 1985) to give the series of mutants described

in Fig. 1. Each of these derivatives was used to transform

fission yeast, and the effect of TATA box modification on

CAT activity was determined in extracts of cells grown

either in the presence or absence of thiamine. For com-

parison, we also assayed the level of CAT activity when

cat was fused to the strong adh promoter of Sz. pombe or

I wt ATATATAAAGGAAGAGGAATCCTGGCATATCATCAATTGAA

T6 ATATAA................................ T5 ATATTAAA................................ T4 ATAAA................................ T89 ATA................................ T81 AT................................

Fig. 1. Structure of the nmtl TATA box mutants. The sequence of the

wt nmtl promoter from the TATA box to the region surrounding the

tsp is shown at the top (Maundrell, 1989). The designations and struc-

tures of five TATA box mutants used in the present study are detailed

below. The TATA box in each case is shown in bold-face type, and nt

identical to the wt sequence are indicated by dots. The arrow indicates

the wt tsp located 27 bp downstream of the TATA box (Maundrell,

1989).

to the SV40 promoter, which also functions well in this

organism. The results obtained using 1 pg protein extract

per reaction are shown in Fig. 2. They demonstrate

clearly that progressive deletion of the TATA box results

in a concomitant reduction in CAT activity under induc-

ing conditions as well as in the residual activity in cells

grown under repressing conditions. However, accurate

quantitation of these data was not possible because of the

extremes in activity which result from assaying a constant

amount of extract. We therefore repeated the analysis

using variable amounts of extract chosen to achieve a

uniform level of activity in which approx. 30% of the

radiolabelled substrate was acetylated (data not shown).

By comparing the amounts of extract required, we were

able to quantitate the effect of TATA box mutations on

promoter activity (Fig. 3). All values are normalised to

the activity of the wt promoter in the presence of thia-

mine. In Fig. 3A this is arbitrarily defined as 1 unit; in

+ -+ -+ -+ -+ -+ -+ -+ - P-P--_-_

WT T6 T.5 T4 T89 T81 adh SV40 Fig. 2. Promoter strength of the TATA box mutants as determined by

CAT activity. The Sz. pomhe strain leuf-32 transformed with the

pREPI-cat plasmid or a derivative plasmid mutated in the TATA box

(see Fig.1) was grown overnight from a single colony in minimal me-

dium (Moreno et al., 1991) containing 4 PM thiamine. Saturated cultures

were washed twice in minimal medium, used to re-inoculate minimal

medium (-) or minimal medium containing 4 PM thiamine (+) and

cultured for a further 24 h to a cell density of 5x106/ml. Two other

constructs, pARTl-cat containing the adh promoter (McLeod and

Beach, 1987) and SVE-CAT [pREPI-CAT containing the SV40 pro-

moter from pSG5 (Green et al., 1988) in place of the nmtl promoter]

were used for comparison. Protein extraction and CAT assays were

performed as described by Jones et al. (1988), blocking the reaction after

10 min. Protein concentrations were determined (Bradford. 1976) using

BSA as a standard, and 1 pg total protein was used for each assay.

133

80 - > c .z 5 m 60-

k 0

J 40- .X= (d H

20 -

wl T5 T4 T89 T81 adh SV40 wt

(-1 YJ t-1 (-) (3 (-) (+I-) (+/-I (+I

B

100

80 >

.C 2 5 cu 60

L- 0

wl T6 T5 T4 T89 T81

(+) (+) (+I (+) (+) (+)

Fig. 3. Quantitation of promoter strength based on CAT activity. Vary-

ing amounts of the protein extracts described in Fig. 2 were assayed in

order to obtain about 30% conversion of substrate in each case. The

exact % of acCm was then determined by scintillation counting. (A)

CAT activity of the TATA box mutants grown in the absence of thiamine

(-). For comparison, the activities are shown for the Sz. pombe adh and

SV40 promoters for which identical results were obtained in the pres-

ence or absence (+/-) of thiamine. Also shown for comparison is the

activity of the wt nmtl promoter determined in cells grown in the pres-

ence of 4 PM thiamine (+). (B) Activity of the TATA box mutants grown

in the presence of 4 ,uM thiamine (+). CAT activity is indicated in

arbitrary units normalized to that of the wt promoter determined in

cells cultured under repressing conditions (4 ,uM thiamine). Note that

in order to accomodate the wide range of promoter activities, the scale

in B is expanded loo-fold compared with the scale in A (compare the

last column in A with the first column in B).

Fig. 3B the scale is expanded lOO-fold (compare the last

column in A with the first column in B).

The first conclusion we can draw from the results in

Fig. 3 is that the wt nmtl promoter under inducing condi-

tions of thiamine deprivation is 6-7 times more active

than the adh promoter and about 30 times more active

that the SV40 promoter. As we showed previously, the

presence of thiamine in the medium results in a substan-

tial down-regulation of the nmtl promoter, estimated here

to be around SO-fold, even though thiamine per se has no

general effect on transcription levels, as evidenced by the

fact that CAT activity in the adh and SV40 transformants

remains unaffected.

The minor modifications to the nmtl TATA box (T6

and T.5; leaving ATATAA and ATATTAAA, respectively)

have little effect on the induced levels of promoter activity

but reduce the repressed level of expression two- and

threefold, respectively. More extensive truncation, which

leaves the sequence ATAAA (T4), reduces the induced

level of expression approx. sixfold and the repressed level

about 15fold, while the most extensive deletion, T81,

which leaves only AT of the original sequence and thus

essentially eliminates the element altogether, results in an

go-fold reduction in the induced activity (to about the

same level as the repressed wt promoter) and a further

250-fold reduction in activity in cells cultured in the pres-

ence of thiamine.

We conclude therefore that progressive deletion of the

nmtl TATA box is paralleled by a loss of CAT activity in

the cell and that the induced and residual uninduced

levels are both affected. It is curious that all the TATA

box mutations we tested result in a two- to threefold

greater down-regulation of the uninduced level of expres-

sion than of the induced level, but the significance of this

observation is not yet clear.

That these differences in CAT activity are indeed due

to an effect on the steady-state levels of cat mRNA is

indicated by the results of Northern blotting (Fig. 4). In

this experiment fission yeast was transformed with

pREPl-cat, T4-cat and TSl-cat, and total RNA was puri-

fied from cultures grown in the presence or absence of

thiamine. Hybridization to detect cat mRNA indicated a

close correspondence between CAT activity and mRNA

abundance. The effect of disrupting the TATA box is

therefore to elicit a general reduction in promoter

strength without affecting the thiamine-mediated tran-

scriptional repression.

(b) The location tsp is by TATA

mutations

The end of cut transcript each of RNA

preparations by Northern in Fig. was

mapped primer extension. oligo primer

134

Fig. 4. Northern blot analysis of the CAT mRNA levels in pREPl-cat,

pREP41-cat and pREPRl-cat transformants. Transformants were

grown as described in Fig. 2, and RNA was prepared, electrophoresed

through 1.2% agarose and analysed by Northern blotting as described

previously (Maundrell, 1989). Total RNA (10 pg) from cells transformed

with pREPl-cat (lanes 1,2), pREP41-cat (lanes 3,4) and pREPSl-cat

(lanes 5.6) was probed using a 1.6-kb HindIII-BnmHI fragment contain-

ing the cat gene (Laimins et al., 1982) radiolabelled in vitro by random

priming (Feinberg and Vogelstein, 1984). In lanes 1, 3 and 5 cells were

grown in the absence of thiamine; in lanes 2, 4 and 6 cells were grown

in the presence of 4 ELM thiamine.

mentary to the sequence between nt + 41 and + 58 in the

cat coding region (see legend to Fig. 5) was annealed and

extended with reverse transcriptase as described pre-

viously (Grimm et al., 1988), and the reaction products

were analysed by direct comparison to the sequence of

the nmtl-cat fusion sequenced using the same primer

(Fig. 5). The intensities of the primer extension products

are as expected from the results obtained from Northern

blotting, and importantly, the tsp in all cases, even with

the extreme T81 mutation (see lane S), corresponds to

that previously determined as the tsp for the wt nmtl gene

(lane 1; Maundrell, 1989).

The retention of the wt tsp in the T81 mutant is intri-

guing since the sequence ATAAT, which conceivably

could serve as a substitute TATA element, occurs only

30 bp further upstream. Since transcription in Sz. pombe

normally starts within 25-30 bp of the TATA box, we

might have expected the cat mRNA to begin a corre-

sponding distance upstream in this mutant. The fact that

this does not happen suggests that additional elements

are also involved in defining the tsp in this promoter. In

this regard, it is of interest that sequence comparison

GATCl 23456 1’ 2’ 3’ 4’ 5’ 6’

Fig. 5. Primer extension analysis to determine the tsp of pREPI-cat.

pREP41-cat and pREPXl-cat. Total RNA preparations described in

Fig. 4 were used for primer extension as described in Grimm et al. (1988)

using the oligo SCCCAATGGCATCGTAAAG complementary to the

cat sequence from nt +41 to +58. Lanes 1 and 2. RNA from cells

transformed with pREPl-cat; lanes 3 and 4, RNA from cells trans-

formed with pREP41-cat; lanes 5 and 6. RNA from cells transformed

with pREP81-cat. Cells analysed in lanes I, 3 and 5 were cultured in

minimal medium lacking thiamine (Moreno et al., 1991); cells analysed

in lanes 2,4 and 6 were cultured in the same medium containing 4 PM

thiamine. Lanes 1’ to 6’ show a longer exposure of lanes 1 to 6. The

products of the reverse transcriptase reaction were analysed by electro-

phoresis on a 6% polyacrylamide sequencing gel. The sequence of

pREPI-cat using the primer described above was run in parallel. The

position of the TATA element is bracketed. The arrowhead indicates

the position of the tsp.

between nmtl and a second thiamine-repressible gene,

nmt2, has revealed an identical 1 1-bp element surround-

ing the tsp which may be involved in defining the location

of this site (A. Manetti, M. Rosetto and K.M., manuscript

in preparation). In some mammalian promoters, se-

quences around the tsp have been shown to co-operate

with the TATA element in defining this point (Concino

et al., 1984).

(c) Construction of pREP41 and pREP81 expression vectors

The repressibility of the nmtl promoter was exploited

previously to generate the PREP series of regulated ex-

pression vectors for Sz. pombe (Maundrell, 1989; 1992,

preceding article). The quantitative analysis presented

above demonstrates that the wt nmtl promoter is ex-

tremely active in this construct, induced expression levels

being about six-fold higher than levels obtained with the

Sz. pombe adh promoter. While this high level of expres-

sion can be useful, circumstances do arise in which an

135

excess of gene product in the cell can produce undesirable,

non-physiological effects (Gould et al., 1991). Moreover,

as we describe elsewhere, it is not possible to control the

level of expression from the nmtl promoter by changing

the thiamine concentration in the medium (Tommasino

and Maundrell, 199 1). In this respect, the TATA box mut-

ations described above are useful in that they provide a

means of attenuating transcription efficiency without

affecting other properties of the promoter, namely thia-

mine repressibility or the tsp. We have therefore con-

structed two new families of vectors based on the T4 and

T81 mutated promoters. These mutations have been in-

corporated into pREP1, generating pREP41 and

pREP81, and into pREP2, generating pREP42 and

pREP82 (Table I). The extended range of vectors thus

allows a greater degree of control over the steady-state

level of mRNA in the cell. Moreover, they provide a con-

venient way of analysing the in vivo effects of gene-

product dosage on cell physiology.

In addition to downregulating the induced levels of

expression, the modified promoters have the further prop-

erty of reducing the residual expression often observed

under repressing conditions. Compared with the wt pro-

moter, the T4 and T81 mutations are, respectively, 16-

fold and 250-fold less active when the cells are cultured

in the presence of thiamine (Table I). This more complete

shutdown of expression is especially useful in suppressing

the phenotypic effects of gene products which are active

at low levels in the cell and has proved particularly useful

in the isolation and analysis of dominant lethal mutations

in which even very low level expression is toxic (Ducom-

mun et al., 1991).

TABLE I

Characteristics of the pREP1, pREP41 and pREP81 vectors and their

relative promoter strengths under inducing and repressing conditions

Vector” TATA box Relative promoter activityb

~ thiamine + thiamine

pREPI

pREP41

pREP8 1

ATATATAAA 80 1

ATAAA 12 0.06

AT 1 0.004

“pREP41 and pREP81 are identical to pREPI described previously

(Maundrell, 1989; 1992, preceding article) except for differences in the

TATA box as indicated. All carry the S. cereuisiae LEU2 gene which

complements the Sz. pombe leul-32 allele. The corresponding deriva-

tives of pREP2 (Maundrell, 1992, preceding article) designated pREP42

and pREP82 have also been constructed. The latter carry the SZ. pombe ura4+ marker in place in LEU2.

bRelative promoter activity was measured as CAT activity (see Fig. 3).

Data are normalized to the activity of the wt promoter determined in

cells cultured in the presence of thiamine (= 1). For comparison, the

relative activities of the adh and SV40 promoters are 12 and 2.5, respec-

tively, when assayed either in the presence or absence of thiamine.

ACKNOWLEDGEMENTS

The authors would like to express thanks to Dr. Marco

Bazzicalupo and Dr. Giulio Draetta for collaboration

and for supplying laboratory space, equipment and mate-

rials; to Dr. Dirk Bohmann for comments on the manu-

script; and to Christopher Hebert and Magali Leemann-

Husler for help in preparing the figures. G.B. would also

like to thank Concetta Schipani for encouragement dur-

ing the initial phase of this work.

REFERENCES

Benoist, C., O’Hare, K., Breathnach, R. and Chambon, P.: The oval-

bumin gene-sequence of putative control region. Nucleic Acids

Res. 8 (1980) 1277142.

Blake, M.C., Jambou, R.C., Swick, A.G., Kahn, J.W. and Azizkhan, J.C.:

Transcriptional initiation is controlled by upstream CC-box inter-

actions in a TATA-less promoter. Mol. Cell. Biol. 10 (1990) 6632-

6641.

Bradford, M.: A rapid and sensitive method for the quantitation of

microgram quantities of protein utilizing the principle of protein

dye binding. Anal. Biochem. 7 (1976) 248-254.

Chen. W. and Struhl, K.: Yeast mRNA initiation sites are determined

primarily by specific sequences, not by the distance from the TATA

element. EMBO J. 4 (1985) 3273-3280.

Concino, M.F., Lee, R.F., Merryweather, J.P. and Weinmann, R.: The

adenovirus major late promoter TATA box and initiation site are

both necessary for transcription in vitro. Nucleic Acids Res. 12

(1984) 7423-7433.

Dierks, P., Van Ooyen, A., Cochran, M.D., Dobkin, C., Reiser, J. and

Weissmann, C.: Three regions upstream from the cap site are re-

quired for efficient and accurate transcription of the rabbit p-globin

gene in mouse 3T6 cells. Cell 32 (1983) 6955706.

Ducommun, B., Brambilla, P., Felix, M-A., Franza R.B., Karsenti, E.

and Draetta, G.: cdc2 phosphorylation is required for its interaction

with cyclin. EMBO J. 10 (1991) 331 l-3319.

Dynan,W.S. and Tijan, R.: Control of eukaryotic messenger RNA syn-

thesis by sequence specific DNA-binding proteins. Nature 316 (1985)

774478 1.

Feinberg, A.P. and Vogelstein, B.: A technique for radiolabelling DNA

restriction nuclease fragments to high specific activity. Anal. Bi-

ochem. 137 (1984) 2666267.

Furter-Graves, E.M. and Hall, B.: DNA sequence elements required for

transcription initiation of the Schizosaccharomyces pombe ADH

gene in Saccharomyces cerevisiue. Mol. Gen. Genet. 223 (1990)

4077416.

Gould, K.S., Moreno, S., Owen, D.J., Sazer, S. and Nurse, P.: Phosphor-

ylation at thr i6’ is required for Schizosaccharomyces pombe. ~34~~~’ function. EMBO J. 10 (1991) 329773309.

Green, S., Isserman, I. and Sheer, E.: A versatile in vitro eukaryotic

expression vector for protein engineering. Nucleic Acids Res. 16

(1988) 369.

Grimm, C., Kohli, J., Murray, J. and Maundrell, K.: Genetic engineering

of Schizosaccharomyces pombe: a system for gene disruption and

replacement using the urn4 gene as a selectable marker. Mol. Gen.

Genet. 215 (1988) 81-86.

Guarente, L. and Bermingham-McDonogh, 0.: Conservation and evo-

lution of transcriptional mechanisms in eukaryotes. Trends Genet.

8 (1992) 27732.

136

Hahn, S., Hoar, E.T. and Guarente, L.: Each of the three ‘TATA-ele-

ments’ specifies a subset of the transcription initiation sites at the

CYCI promoter of Saccharomyces cerevisiae. Proc. Nat1 Acad. Sci.

USA 82 (1985) 856228566.

Healy, A.M. and Zitomer, RX: A sequence that directs transcriptional

initiation in yeast. Curr. Genet. 18 (1990) 1055109.

Hen, R.: Sassone-Corsi, P., Corden, J.; Gaub; M.P. and Chambon, P.:

Sequences upstream from the TATA-box are required in vivo and

in vitro for efficient transcription from the adenovirus serotype 2

major late promoter. Proc. Nat1 Acad. Sci. USA 79 (1982) 7132-

7136.

Jones, K.A., Luciw, P.A. and Duchange, N.: Structural arrangements of

transcription control domains within the 5’ untranslated leader re-

gions of the HIV-l and HIV-2 promoters. Genes Dev. 2 (1988) 1101-

1114.

Jones, R.H., Moreno, S.. Nurse, P. and Jones, N.C.: Expression of the

SV40 promoter in fission yeast: identification and characterization

of an AP-I like factor. Cell 53 (1988) 5699667.

Kunkel, T.A.: Rapid and efficient site specific mutagenesis without phe-

notypic selection. Proc. Nat1 Acad. Sci. USA 82 (1985) 488492.

Laimins, L.A., Khoury, G., German; C., Howard, B. and Gruss, P.:

Host-specific activation of transcription by tandem repeats from

simian virus 40 and Moloney murine sarcoma virus. Proc. Nat1

Acad. Sci. USA 79 (1982) 6453-6457.

Maundrell, K.: nmtf of fission yeast: a highly expressed gene completely

repressed by thiamine. J. Biol. Chem 265 (1989) 10857710864.

Maundrell. K.: Thiamine-repressible expression vectors PREP and

pRIP for fission yeast. Gene 123 (1993) 127-130.

McLeod, M. and Beach, D.: The product of the mei3- gene, expressed

under control of the mating-type locus, induces meiosis and sporula-

tion in fission yeast. EMBO J. 4 (1987) 3531-3538.

McNeil, J.B. and Smith, M.: Saccharomyces cereuisiae CYCl mRNA 5’

end positioning: analysis by in vitro mutagenesis using synthetic

duplexes with random mismatch base pairs. Mol. Cell. Biol. 5 (1985)

354553551.

Moreno, S., Klar, A. and Nurse; P.: Molecular genetic analysis of fission

yeast, Schizosaccharomyces pombe. Methods Enzymol. 194 (199 1)

7955823.

Myers, R.M., Tylly, K. and Maniatis, T.: Fine structure genetic analysis

of the B-globin promoter. Science 232 (1986) 6133618.

Nagawa, F. and Fink, G.: The relationship between the TATA sequence

and the transcription initiation site at the HIS4 gene of Snccharo-

myces cerevisiae. Proc. Natl. Acad. Sci. USA 82 (1985) 8557-8561.

Russell, P.R.: Evolutionary divergence of the mRNA transcription initi-

ation mechanism in yeast. Nature 301 (1983) 167-169.

Russell, P.: Transcription of the triose-phosphate-isomerase gene of

Schizosnccharomyces pombe initiates from a start point different to

that in Snccharanzyces cereuisiae. Gene 40 (1985) 125- 130.

Smale, S.T. and Baltimore, D.: The ‘initiator’ as a transcriptional control

element. Cell 57 (1989) 1033113.

Struhl, K.: Promoters, activator proteins, and the mechanism of tran-

scriptional activation in yeast. Cell 49 (1987) 2955297.

Tommasino. M. and Maundrell, K.: Uptake of thiamine by Schizos-

acchnronzyces pombe and its effect as a transcriptional regulator of

thiamine-sensitive genes. Curr. Genet. 20 (1991) 63366.

Wu: L.D.. Rossner, SE., Schmidt, M.C. and Berk, A.: TATA-box impli-

cated in transcriptional activation of a simple adenovirus 2 pro-

moter. Nature 236 (1987) 512-515.