JOURNAL OF

BIOSCIENCE AND BIOENGINEERWG

Vol. 89, No. 5, 426-430. 2000

Effect of Ethanol on the Production of Carboxypeptidase Y Using the GAL10 Promoter in a Saccharomyces cerevisiae gal80 Mutant
YOICHIRO SHIBA,* CHIHO ONO, FUMIO FUKUI, AND HIROJI YOSHIKAWA Lead Discovery Research Laboratories, Sankyo Co. Ltd., 389-4 Aza-Ohburugi, Shimokawa, Izumimachi, Iwaki, Fukushima 971-8183, Japan
Received 19 November 1999/Accepted 1 February 2000

In the course of studying carboxypeptidase Y (CPY) production, we found that the expression level of the gene, which is under the control of the GAL10 promoter, increased in a Sacchuromyces cerevisiaegal80 mutant grown in a medium containing ethanol as the sole carbon source. In the cultivation of the gaZ80 mutant KS58-2D/pCY303 carrying a multlcopy plasmld, which contains the PRCZ gene fused to the GAL10 promoter, CPY production continued after the consumption of galactose. In this phase, the cells utilized ethanol as the carbon source. To increase the CPY production level, we examined the effect of carbon source feeding in a fed-batch culture. The production level in the fed-batch culture using ethanol was l.ffold higher than that in a batch culture and l.dfold higher than that in a fed-batch culture using galactose. By $-deletion analysis of the GAL20 promoter, the region between - 256 and - 232 was found to be important for the promoter activity in the gal80 mutant growing in the presence of ethanol. cerevisiae,ethanol, gal80 mutant, GAL10 promoter] [Key words: carboxypeptidase Y, Saccharomyces Carboxypeptidase Y (CPY), encoded by the PRCI gene, is a Bl-kDa glycoprotein localized in the vacuoles of the yeast Saccharomyces cerevisiae (1, 2). CPY is a useful enzyme for determining C-terminal sequences of peptides and proteins (3). In addition to its peptidase activity, CPY catalyzes aminolysis of C-terminal peptide esters in the presence of suitable nucleophiles, resulting in the elongation of a peptide chain or amidation of a peptide (4). The enzyme has therefore attracted the attention of industrial chemists as a possible candidate for the production of biologically active proteins such as calcitonin (5). We previously reported on the transformation of the S. cerevisiae ssll mutant KS58-2D, a soluble vacuolar protease-missorting mutant, with multicopy plasmid pCY303 which contains the PRCl gene fused to an inducible GAL10 promoter (6). The transformant KS582D/pCY303 efficiently secreted active CPY into the culture medium using galactose for cell growth and induction of the GALZO promoter. Furthermore, through the reassessment of medium composition and optimization of culture conditions, we succeeded in the secretion of active CPY up to a concentration of over 400mg/f (7), which was lo-fold higher than that reported previously (8). During these studies, we found that the production of CPY continued even after galactose was fully consumed. In this phase, the cells utilized ethanol as the carbon source. In this study, we found that the expression level of the gene, which is under the control of the GAL10 promoter, increased in an S. cerevisiae gal80 mutant grown in a medium containing ethanol as the sole carbon source. To increase the production level of CPY by the gal80 mutant KS58-2D/pCY303, we applied an ethanol feeding strategy. We also investigated the region of the GAL10 promoter associated with the expression caused by ethanol in the gal80 mutant. MATERIALS AND METHODS

* Corresponding author. 426

The S. cerevisiae strains used Strains and plasmids in this study were KS58-2D (MATa ssll leu2 his3 ura3 galgO), KK4 (MATa leu2 ura3 trpl his1 or his3 gaBO), PEP4 (MATa leu2 pep4) and A2-l-IA (MATa mall gal2 leu2). KS58-2D is a soluble vacuolar protease-missorting mutant derived from a cross between KSSl-3C and KSM-169 (9, 10). Escherichia coli JMlO9 (11) was used as the host for the propagation and manipulation of plasmids. Construction of the multicopy plasmids, pCY303 and pGAL-LUC, in which the PRCI gene (pCY303) and luciferase gene (pGAL-LUC), respectively, were inserted between the GAL10 promoter and the 2-,um plasmid FLP terminator, was described previously (6, 7, 12). A completely synthetic Medium and cultivation (CS) medium containing a vitamin mixture (2-10,000 pg/l), an amino acid mixture (20-4OOmg/Z), trace elements (40-500 pg/l) and salts (0.1-7.5 g/T) was prepared, whose composition has been described previously (7) and further supplemented with 2% glucose and 0.5% skim milk. CS medium was used for the preculture of the yeasts. Preculture in a 200 ml Erlenmeyer flask containing 20 ml of CS medium was performed at 26C for 3 d on a rotary shaker (210 rpm). CS-2 medium containing histidine (final concentration of 60 mg/l) and the components of CS medium without 2% glucose was prepared with the composition described previously (7). CS2 medium supplemented with 2% glucose (Glu medium), 2% galactose (Gal medium), 2% ethanol (Eth medium), 2% glycerol (Gly medium), 2% raffinose (Raf medium), 2% lactate (Lac medium) or 2% glucose in addition to 2% galactose (GluGal medium) was used for the production culture. After inoculating 1 ml of the preculture broth into 100 ml of fresh production culture medium in a 500 ml Erlenmeyer flask, yeast cells were cultivated at 28C for 1 d (Glu medium), 2d (Gal medium) or 5 to 6d (Eth, Gly, Raf, Lac and GluGal medium). For the cultivation using a 30-l fermentor, a second preculture in GluGal medium was carried out at 28C for 3 d. The

VOL. 89. 2000

EFFECT OF ETHANOL ON GALlO PROMOTER

427

production culture was conducted with a 1% inoculum in a 30-I fermentor containing 20 1 of GluGal medium supplementedwith O.l2g/l histidine at 28C, 0.5 vvm of aeration and 0.05 MPa of internal pressure for 7-10 d. DO was maintained at about 5 ppm by varying the agitation speed. Ethanol feeding was begun when ethanol in the culture broth was consumed. The concentration of ethanol in the feed solution was 50% (w/v). The methods used for preparaGenetic methods tion and manipulation of DNA and RNA, and for the transformation with lithium acetate were described by Rose et al. (13). The methods used for the transformation of E. coli and nucleotide sequencedetermination by the dideoxy-chain termination method were describedby Sambrook et al. (11). Preparation of cell-free extract To measurethe intracellular enzymatic activity, cell-free extracts were prepared as follows. After washing with water and 100mM Tris-HCI buffer (pH 7.5), the cells were resuspendedin the same buffer and disrupted for 3 m in with glass beads by a cell homogenizer(B. Braun MelsungenAG, Germany). The suspensionwas centrifuged at 15,OOOrpm for 10min at 4C and the obtained supernatant was used for the enzymeassays. Analytical methods Concentrationsof biomass,glucose, galactose and ethanol were measured as previously described (7). CPY activity was measured photometrically by a kinetic method as previously described (6, 7). Luciferase activity was measured spectrophotometrically by a luminometer detection method using a PicaGene kit (Toyo Ink Co. Ltd., Tokyo) (12). The assay for UDP-galactose 4-epimerasewas performed as described by Wilson et al. (14). Total protein was measured with the BioRad (Richmond, CA., USA) protein assay. Bovine serum albumin was used as a reference. Twenty m icrograms of total Northern blot analysis RNA was electrophoresedon 1.5% agarosegel containing formaldehyde. Following its transfer to a HybondN+ nylon membrane(Amersham, Buckinghamshire,UK) by capillary action, the blot was hybridized with the 1.3kb SalI-BgfII fragment containing a part of the PRCZ gene or the l.O-kb ElindIII-XhoI fragment containing a part of the ACT1 gene. The probes were labeled with fluorescein ll-dUTP, and a Fluorescein Gene Imagesdioxetane detection module was used for the detection according to the suppliersinstructions (Amersham).
Construction of deletions in the GAL10 promoter

TABLE 1. Effect of carbon sources on the production of CPY Strain KS58-2D KS58-2D/pCY303 Glu ND ND Gal ND 11.5 CPY (mg/ml) Eth Glv ND ND 8.9 2.4 Raf ND 1.5 Lac ND 1.3

Yeast cells were cultivated for 1 din Glu medium, 2 din Gal medium and 5 d in Eth, Gly, Raf or Lac medium. The culture supernatants were used for measuring CPY activities. The amount of CPY was calculated according to the standard activity curve obtained for authentic CPY. ND, Not detected.

that cells were able to use ethanol as the carbon source after galactose was completely consumed and produce CPY continuously (7). To investigate whether or not ethanol is a suitable carbon source for the production of CPY, we cultivated the strains KSSS-2Dand KS58-2D/ pCY303 in several different media. The results are summarized in Table 1. In the caseof KS58-2D, CPY was not detectedin any of the culture supernatants.On the other hand, the production of CPY was observedwhen KS582D/pCY303 was cultivated using any medium except for Glu medium and the production level in Gal or Eth medium was higher than that in Gly, Raf and Lac media. We also examinedthe CPY mRNA level under these conditions. As shown in Fig. 1, a large amount of CPY mRNA was observed when either galactose or ethanol was used as the carbon source. Furthermore, we measured the activity of UDP-galactose4-epimerase, which is encodedby the GALZO gene (Table 2). In the cultivation of KS58-2D, we could not find any differences in the specific activity of UDP-galactose 4-epimerasewhether galactose or ethanol was used as the carbon source. These results show that ethanol causesthe high expression of the PRCZ gene under the control of the GALZO promoter in KS58-2D/pCY303, resulting in the production of CPY.
Application of ethanol feeding for CPY production

The deletions in the GALZO promoter were constructed as follows. A 2.7-kb BarnHI-Hind111 fragment containing the GAL10 promoter and the PRCl gene from pCY303 was ligated into the BamHI-Hind111 gap of pHSG398 (TAKARA Shuzo Co. Ltd., Kyoto). Unidirectional 5 deletions in the GALZO promoter in the resultant plasmid were performed using the Exo III nesteddeletion kit, Erase-a-Base System(PromegaCo., Madison, USA). After the constructs were sequenced,they were digested with EcoRI and HindIII. The resultant fragments containing the deleted GALZO promoters and the PRCZ gene replaced the 2.7-kb BarnHI-Hind111 fragment region of pCY303, which originally contained the full length of the GALZO promoter and the PRCZ gene.
RESULTS AND DISCUSSION Effect of carbon sources on the production of CPY

We previously reported that KS58-2D/pCY303 was a useful strain for the secretory production of CPY (6, 7). To increase the productivity of CPY, we investigated the effect of the addition of ethanol on the cultivation of KS58-2D/pCY303. The results are shown in Table 3. When KS58-2D/pCY303 was cultivated for 2 d in GluGal medium, the cells used up glucose and galactose (data not shown). When galactose was added to the 2d culture broth and cultivation was extended for four more days, the production level of CPY was about 70%

PRCl

During the cultivation of KS58-2D/pCY303, we found

FIG. 1. Effect of carbon sourceson PRCI transcription under the control of the GAL10 promoter. Total RNA was prepared from yeast cells grown in Glu medium (lane l), Gal medium (lane 2) or Eth medium (lane 3). Twenty micrograms of total RNA was loaded onto a formaldehyde-agarose gel and Northern blot analysis was performed, as described in Materials and Methods.

428

SHIBA ET AL.

J. BIOSCI.BIOENG.,

TABLE

2. Comparison of UDP-galactose 4-epimerase activity using various carbon sources UDP-galactose 4-epimerase (U/ma) 0.017 0.478 0.539
2.5

Glu Gal Eth

20

15

The strain KS58-2D was cultivated for 1 d in Glu medium, 2 din Gal medium and 5 d in Eth medium. After cell extracts were prepared as described in Materials and Methods, UDP-galactose 4-epimerase activities were measured. The activities are expressed as units/mg of protein.

III

s I (I

of that in the control culture (with no addition of galactose or ethanol). The addition of galactose caused a high accumulation of ethanol in the medium, resulting in the reduction of cell mass and CPY production, as previously described (7). In the case where ethanol was added to a 4-d culture broth of GluGal medium, the cells consumed ethanol for growth and the production level of CPY increased to l.Zfold higher than that in the control culture, indicating the effectiveness of ethanol feeding for CPY production. On the basis of these results, a number of fed-batch fermentations with continuous feeding of ethanol were conducted in a 30-l fermentor. Figure 2 shows the time course of CPY secretion in the fed-batch culture. After ethanol was consumed, a 50% ethanol solution was fed continuously at a constant rate of 0.5%/d. With continued ethanol feeding, CPY production increased to reach a final concentration of 512 mg/f. When the feed rate was increased to more than 0.5%/d, ethanol accumulated at high levels, resulting in the reduction of CPY production and cell mass, as reported by Sohn et al. (15) and Park et al. (16). The production level in the fed-batch culture using ethanol was 1.3-fold higher than that in a batch culture and 1.6fold higher than that in a fed-batch culture using galactose (7). These results indicate the effectiveness of ethanol feeding for improvement of CPY production in KS5&2D/pCY303. We also suppose that the secretory production system (10) with an optimum ethanol feeding
TABLE Addition None 3. Effect of the addition of carbon sources on the production of CPY Time (4 2 4 6 2 4 6 4 6 CPY @x/ml) 49.5 112.4 108.7 51.2 80.2 73.6 110.3 139.4 Cell mass (g-dry cells/f) 2.67 3.81 3.81 2.72 3.54 3.62 3.75 4.32 Ethanol (%I 1.18 0.17 ND 1.15 1.65 1.48 0.18 (1.20) 0.33

Time

(h)

FIG. 2. Time courses of CPY-producing culture with continuous feeding of ethanol in a 30-1fermentor. The arrow indicates the time at which feeding was begun. Symbols: 0, secreted CPY concentration (mg/l); 0, cell concentration (g-dry cells/l); A, glucose concentration (g/f); A, galactose concentration (g/n; I , ethanol concentration (g/f).

Galactose

strategy using KS58-2D can be applied for high-level production of heterologous proteins. Effect of g&O mutation on the gene expression under the control of the GAL10 promoter The strain KS582D is a gal80 mutant derived from the strain KK4 (9). Torchia et al. reported that the GAL10 gene was highly expressed in the gal80 disruptant grown in medium containing noninducing and nonrepressing sugars, such as glycerol and lactate (17). Yocum et al. also reported the expression of the GALI-lacZ fusion gene in the gal80 mutant grown in the presence of glycerol plus ethanol (18). The GAL1 and GAL10 genes are adjacent to each other and are transcribed divergently from a shared 365bp control region, UASo (18, 19). To clarify why KS582D/pCY303 produced CPY during the consumption of ethanol, we investigated the effect of the gal80 mutation on the gene expression under the control of the GALlO promoter. Yeast strains harboring pGAL-LUC, in which the luciferase gene is expressed under the control of the GAL10 promoter, were cultivated in Eth, Gly, Raf or Lac medium. The results are shown in Table 4. Remarkably, higher luciferase activity was detected in the gal80 mutants compared to the GAL80 wild-type strains in Eth medium. In the gal80 mutant KS58-2D/pGAL-LUC, luciferase activity was detected in any culture. However, the activity in Eth medium was much higher than that in other media. We consider that these results support the effectiveness of ethanol as a feeding substrate for the
TABLE 4. Comparison of the expression of the luciferase gene under the control of the GAL10 promoter between gal80 mutant and GAL80 wild-type strain Strain Genotype Medium Luciferase activity W-U) 4 7 248 288 45 29 21

Ethanol

After the strain KS58-2D/pCY303 was cultivated for 2 d in GluGal medium, galactose was added to the medium to adjust the concentration at 1% and cultivation was extended for four more days. Similarly, after the strain KS58-2D/pCY303 was cultivated for 4d in GluGal medium, ethanol was added to the medium and cultivation was extended for two more days. When neither galactose nor ethanol was added to the medium (none: control culture), the cells were cultivated for 6d. The culture supernatants were used for measuring CPY activities and ethanol concentrations. The amount of CPY was calculated according to the standard activity curve obtained for authentic CPY. The value in parenthesis shows the concentration after the addition of ethanol. ND, Not detected.

Eth GAL80 PEP4/pGAL-LUC Eth GAL80 AZ-l-lA/pGAL-LUC Eth gal80 KK4/pGAL-LUC Eth gal80 KS58-2D/pGAL-LUC KS58-2D/pGAL-LUC gal80 GUY gal80 Raf KS58-2D/pGAL-LUC gal80 Lac KS58-2D/pGAL-LUC Yeast cells were cultivated for 5 d in different media. After cell extracts were prepared as described in Materials and Methods, luciferase activities were measured.

VOL. 89. 2000

AAAAT CATCCCTTCG CTGATTAATT ACCCCACAAA TAAGGCTAAA AAACTAATCG CATTATCATC CTATGGTTGT TAATTTGATT CGTTCATTTG AAGGTTTGTG GGGCCAGGTT ACTGCCAATT TTTCCTCTTC ATAACCATAA AAGCTAGTAT TGTAGAATCT TTATTGTTCG GAGCAGTGCG GCGCGAGGCA CATCTGCGTT TCAGGAACGC GACCGGTGAA GACGAGGACG CACG~AGGAG~AG ClTa 1 CIXJ..G 2 -461 -421 -361 -301 -241 -181 -121 -61 -1

EFFECT OF ETHANOL ON GALlO PROMOTER

429

F TCAATATAGC AATGAGCACT TAAGCGTATT ACTGAAAGTT CCAAAGAGAA GGTTTTTTTA GGCTAAGATA ATGGGGCTCT TTACATTTCC ACAACATATA AGTAACATTA GATATGGATA TGTATATGGA TATGTATATG GTGGTAATGC CATGTAATAT GATTATTAAA CTTCTTTGCG TCCATCCAAA AAAAAAGTAA GAATTTTTGA

FIG. 3. DNA sequence of the GAL10 promoter region in the PRCZ expression plasmid pCY303. Nucleotide - 1 corresponds to the position 154 in the DNA sequence of the GALI-GAL10 divergent promoter region reported by Yocum et al. (18). Arrows denote deletion endpoint positions of -256 and -232. Gal4-protein-binding regions 2 and 3 are indicated by dotted lines. Negative control element GAL4 is underlined. Activating element GAE, is indicated by bracket. The open box and hatched box indicate the Gcrl-protein and the Rebl-protein-binding region, respectively.

g&80 mutant KS%2D. Torchia et al. (17) reported that only Gal80p-dependent inhibition of Gal4p prevents GALlO transcription in glycerol. Therefore, disruption of GAL80 results in an extremely high level of the gene expression. Our results suggest that the high level of the gene expression, which is under the control of the GALZO promoter, caused by ethanol is due to the gal80 mutation. Deletion analysis of the GAL10 promoter The sequence of the 505bp fragment of the GAL10 promoter region in the CPY expression plasmid, pCY303, is shown in Fig. 3. To find which region in the promoter is involved in the expression caused by ethanol in gal80 mutant, we constructed various S-truncated versions of the GALZO promoter. The multicopy plasmids containing the deleted GAL10 promoters with the PRCI gene were used to transform KS58-2D, and the obtained transformants were cultured in Glu, Gal or Eth medium (Fig. 4). The transformant with the 5 -endpoint 505-bp upstream of the translational starting site, produced about lOmg/l of CPY when cultivated in Gal or Eth medium. Additional 5 -deletions had no significant effect on the expression of the PRCI gene until the region between -423 and -379 was deleted. In the case of galactose,
C P Y (mgN) au Gal Eth

deletion from -379 to -256 decreased the production level of CPY to 3.0mg/Z. With regard to ethanol, removal of this region did not affect the expression of the PRCI gene, suggesting that the expression mechanism by ethanol may be different from that induced by galactose. In both cases, CPY was not detected in the culture supernatant when the region between -256 and -232 was further deleted. Northern blot analysis also showed the same results for CPY production caused by ethanol using these deletion mutants (data not shown). West et al. (20) reported a deletion analysis of the GAL10 promoter using GAL80 and GAL4c strain grown in glycerol plus ethanol. The GAL20 promoter activity gradually decreased with deletion from the full length to -259, and completely disappeared when the region between -259 and -240 was deleted. Our results show that the GAL10 promoter activity in gal80 mutants cultivated using ethanol was not affected by deletion up to -256, and completely disappeared by deletion from -256 to -232. As shown in Fig. 3, the region between -256 and -232 includes a part of the negative control element GAL02 and the Gal4p/galactose-independent activating element GAEr which corresponds to the binding sites for the general regulatory proteins, Rebl protein and Gcrl protein (21-25). This region also overlaps with a part of the Gal4p-binding site 3 and the entire region of the Gal4p-binding site 2 (26). Our results suggest that this region is important for the GAL20 promoter activity in the gal80 mutant growing in the presence of ethanol. It would be interesting to study how this transcriptional system is regulated and to consider application of the region including the sequence from -256 to -232 for further improvement of CPY production. Further molecular and biological analysis is required to clarify the expression mechanism caused by ethanol.
REFERENCES 1. Valls, L.A., Hunter, C. P., Rothman, J. H., and Stevens, T. H.: Protein sorting in yeast: the localization determinant of yeast vacuolar carboxypeptidase Y resides in the propeptide. Cell, 48, 887-897 (1987). 2. Hasilik, A. and Tanner, W.: Biosynthesis of the vacuolar yeast glycoprotein carboxypeptidase Y: conversion of precursor into the enzyme. Eur. J. Biochem., 85, 599-608 (1978). 3. Breddam, K. and Ottesen, M.: Determination of C-terminal sequences by digestion with serine carboxypeptidase: the influence of enzyme specificity. Carlsberg Res. Commun., 52, 55-63 (1987). 4. Widmer, F. and Johansen, J. T.: Enzymatic peptide synthesis. Carboxypeptidase Y catalyzed formation of peptide bonds. Carlsberg Res. Commun., 44, 37-46 (1979). 5. Tamaoki, H., Kawagochi, J., and Satoh, K.: C-terminal amidation of synthetic model substrates of human calcitonin by carboxypeptidase Y. Ann. Rep. Sankyo Res. Labo., 38, 73-79 (1986). 6. Ichikawa, K., Shiba, Y., Jigami, Y., and Serizawa, N.: Secretion and overproduction of carboxypeptidase Y by a Succharomyces cerevisiae ssll mutant strain. Biosci. Biotech. Biochem., 57, 1686-1690 (1993). 7. Shiba, Y., Fukui, F., Ichikawa, K., Serizawa, N., and Yoshikawa, H.: Process development for high level secretory production of carboxypeptidase Y by Saccharomyces cerevisiae. Appl. Microbial. Biotechnol., 50, 34-41 (1998). 8. Nielsen, T. L., Holmberg, S., and Petersen. J. G. L.: Regulated overproduction and secretion of yeast carboxypeptidase Y. Appl. Microbial. Biotechnol., 33, 307-312 (1990). 9. Suzuki, K., Ichikawa, K., and Jigami, Y.: Yeast mutants with enhanced ability to secrete lysozyme: isolation and identifica-

-423 I -379 I

ND ND N!J ND

11.0 10.9 3.0 ND

9.2 9.6 9.5 ND

FIG. 4. Deletion in the GALlO promoter and its effect on the expression of the PRCI gene. The strain KS58-2D was transformed with multicopy plasmids which contained the PRCI gene under the control of the deleted GAL10 promoters. The transformants were cultivated for 1 d in Glu medium, 2 d in Gal medium and 5 d in Eth medium. CPY activities in the culture supernatants were measured. The amount of CPY was calculated according to the standard activity curve obtained for authentic CPY. The numbers at the left ends of the boxes represent the length (in bp) of the GALlO promoter %flanking DNA present in each plasmid. ND, Not detected.

430

SHIBA ET AL. tion of a protease-deficient mutant. Mol. Gen. Genet., 219, 5864 (1989). Shiha, Y., Ichikawa, K., Serizawa, N., and Yoshikawa, H.: Extracellular processing of carboxypeptidase Y secreted by a Succharomyces cerevisiae ssll mutant strain. J. Ferment. Bioeng., 86, 545-549 (1998). Sambrook, J., Fritsch, E. F., and Maniatis, T.: Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Lab., Cold Spring Harbor, New York (1989). Ichikawa, K., Shiba, Y., Yamazaki, M., and Serizawa, N.: Thiamine increases expression of yeast gene. Biosci. Biotech. Biochem., 61, 1221-1224 (1997). Rose, M. D., Winston, F., and Hieter, P.: Methods in yeast genetics: a laboratory course manual. Cold Spring Harbor Lab., Cold Spring Harbor, New York (1990). Wilson, D. B. and Hogness, D. S.: Galactokinase and uridine diphosphogalactose 4-epimerase form Escherichiu coli. Meth. Enzymol., 8, 229-240 (1966). Sohn, J. H., Choi, E. S., Chung, B.H., Youn, D. J., Seo, J. H., and Rhea, S. K.: Process development for the production of recombinant hirudin in Saccharomyces cerevisiae: from upstream to downstream. Process Biochem., 30, 653-660 (1995). Park, Y. S., Shiba, S., Iijima, S., and Kobayashi, T.: Comparison of three different promoter systems for secretory n-amylase production in fed-batch cultures of recombinant Succharomyces cerevisiae. Biotechnol. Bioeng., 41, 854-861 (1993). Torchia, T. E., Hamilton, R. W., Cano, C. L., and Hopper, J. E.: Disruption of regulatory gene GAL80 in Saccharomyces cerevisiae: effects on carbon-controlled regulation of the galactose/melibiose pathway genes. Mol. Cell. Biol., 4, 1521-1527 (1984). Yocum, R. R., Haniey, S., West, R., Jr., and Ptashne, M.: Use of lucZ fusions to delimit regulatory elements of the induci-

J. BIOSCI. BIOENG., ble GALl-GAL10 promoter in Saccharomyces cerevisiae. Mol. Cell. Biol., 4, 1985-1998 (1984). Johnston, M.: A model fungal gene regulatory mechanism: the GAL genes of Saccharomyces cerevisiae. Microbial. Rev., 51, 458-476 (1987). West, R., Jr., Yocum, R. R., and Ptashne, M.: Succharomyces cerevisiue GALl-GAL10 divergent promoter region: location and function of the upstream activating sequence USAc. Mol. Cell. Biol., 4, 2467-2478 (1984). Chasman, D. I., Lue, N. F., Buchman, A. R., LaPointe, J. W., Larch, Y., and Kornberg, R. D.: A yeast protein that influences the chromatin structure of UASG and functions as a powerful auxiliary gene activator. Genes Dev., 4, 503-514 (1990). Baker, H. V.: GCRl of Saccharomyces cerevisiae encodes a DNA-binding protein whose binding is abolished by mutations in the CTTCC sequence motif. Proc. Natl. Acad. Sci. USA, 88,9443-9447 (1991). Finley, R. L., Jr. and West, R. W., Jr.: Differential repression of GAL4 and adjacent transcription activators by operators in the yeaSt GAL upstream activating sequence. Mol. Cell. Biol., 9, 4282-4290 (1989). Finley, R. L., Jr., Chen, S., Ma, J., Byrne, P., and West, R. W., Jr.: Opposing regulatory functions of positive and negative elements in UASo control transcription of the yeast GAL genes. Mol. Cell. Biol., 10, 5663-5670 (1990). Chen, S., West, R. W., Jr., Ma, J., Johnson, S. L., Gans, H., and Woldehawariat, G.: TSFl to TSF6, required for silencing the Saccharomyces cerevisiae GAL genes, are global regulatory genes. Genetics, 134, 701-716 (1993). Giniger, E., Varnum, S. M., and Ptashne, M.: Specific DNA binding of GAL4; a positive regulatory protein of yeast. Cell, 40, 767-774 (1985).

10.

19. 20.

11. 12. 13. 14. 15.

21.

22.

23.

24.

16.

25.

17.

26.

18.