Biotechnol. Prog. 1995, 1 1, 294-298
294
Development of Metabolically Engineered Saccharomyces cerevisiae Cells for the Production of Lactic Acid Danilo Porro,* Luca Brambilla, Bianca Maria Ranzi, Enzo Martegani, and Lilia Alberghina Dipartimento di Fisiologia e Biochimica Generali, Sezione Biochimica Comparata, Universita di Milano, Via Celoria 26, 20133, Milano, Italy Interesting challenges from metabolically engineered Saccharomyces cerevisiae cells arise from the opportunity to obtain yeast strains useful for the production of chemical(s). In this paper, we describe the accumulation of lactic acid in the culture medium of growing, engineered yeast cells expressing a mammalian lactate dehydrogenase gene (LDH-A). High and reproducible productions (20g L ) and productivities (up to 11 g/wh) of lactic acid have been obtained by modulating the physiological growth conditions. Since yeast cells are acid tolerant and survive a t very low pH values, the production of lactate can be avoided. In perspective, the approaches described could be useful for the production of lactic acid, outflanking the problems related to the synthesis from bacteria cells. In fact, during industrial productions, there is a n inhibitory effect on the metabolic activities of the growing bacteria (i.e., Lactobacillus spp.) caused by the acid produced and by the low pH value. Thus, strategies to prevent the lowering of pH are conventional operations. These processes allow the production of lactate(s) and require the purification of the acid from its salt. The biotechnological implications of this study are also discussed.
Introduction Saccharomyces cerevisiae is one of the organisms of choice in industrial microbiology, and it is widely used for the production of biomass and ethanol, as well as recombinant proteins, pharmaceutical agents, and vaccines (Fiechter et al., 1981; Romanos et al., 1992; Bucholz, 1993). The expression of specific heterologous activity in S. cerevisiae allows one to modify its pathway(s1 and, consequently, allows the development of new strains useful, for example, for the recovery of energy from renewable sources (Inlow et al., 1988; Wong et al., 1988; Kotter et al., 1990; Adam and Polaina, 1991; Porro et al., 1992a,b; Bailey et al., 1993; Compagno et al., 1993; Ramakrishnan and Hartley, 1993; Murooka and Imanaka, 1994). In addition, S. cerevisiae is an acid tolerant microorganism and can survive at very low pH values. This could be an advantage for the production of organic acid(s) through modification of the yeast metabolism. The applications of lactic acid and its derivatives encompass many fields of industrial activities Le., chemistry, cosmetic, and pharmacy), as well as important aspects of food manufacture and use [for a detailed review, see Benninga (1990)l. During lactic acid production, there is an inhibitory effect caused by the acid produced on the metabolic activities of the producing cells (i.e., Lactobacillus spp.; Hongo et al, 1986; Benninga, 1990). Besides the presence of lactic acid, lowering of the pH value also inhibits cell growth (Hongo et al., 1986). Therefore, the addition of CaC03, NaOH, or NHIOH to prevent lowering of the pH is a conventional operation in industrial processes (Butcha, 1983; Benninga, 1990). These processes allow the production of lactate(s), by maintaining the pH at a constant value (about 5 ) . However, as the solubility of lactate is low, at high concentrations of this product the medium tends to solidify, complicating the fermentation behavior and the subsequent isolation procedures (Butcha, 1983). Finally,
* Author to whom correspondence should be addressed. Telephone: 39 2 70644801.Fax: 39 2 70632811.
++
++
additional operations are required to recover free lactic acid from its salt; this is an easy, but expensive process (Benninga, 1990). All these problems could be overcome if lactic acid could be produced by microorganisms able to grow and survive at low pH values, thus avoiding the production of lactate(s1. In this paper, we show that the expression of a bovine muscle lactate dehydrogenasegene (LDH-A)(Ishiguro et al., 1990) allows the modification of the glycolytic flux in yeast with the production of relevant amounts of lactic acid. Since yeast cells are quite resistant to low pH values, no addition of bases is required and a satisfactory production of lactic acid was observed under controlled fermentation conditions. The strategies described could be useful for biotechnological applications, although for the moment the productivity of lactic acid obtained is lower than that with bacterial systems.
Materials and Methods Yeast Strains and Growth Conditions. S. cerevisiue strains GRF18 (Mat a, leu2, his3) and YSH 5.127-17 (Apdcl, Apdc5, Apdc6; kindly provided by S. Hohmann) were used in this work. Yeast strains were transformed according to the procedure described by Ito et al. (1983). Batch cultures were run by shaking at 30 "C in YEP [1%(w/v) yeast extract and 2% (wh) peptone1 or in minimal medium containing 0.67% (wh) Difco yeast nitrogen base (YNB) without amino acids, supplemented with 50 pg/mL of the required supplements. Glucose (GLU) or galactose (GAL) was used as the carbon source (see text). Pregrowth in fed-batch fermentatiods) was performed in a 2 L aerated, stirred-tank bioreactor equipped with temperature, agitation, air flow rate, dissolved oxygen, pH, and ethanol controllers (Porro et al., 1991). Composition of the mineral medium and control of the bioprocesses were as previously described (Porro et al., 1991). During pregrowth, the bioreactor was aerated with a flow of 2 Umin and the pH value was controlled at 4.950, while the production of lactic acid (see text) was carried
8756-7938/95/3011-0294$09.00/00 1995 American Chemical Society and American Institute of Chemical Engineers
295
Biotechnol. Prog., 1995, Vol. 11, No. 3
pEMBLyex4
pLDH12 m x x w
LDH-A
UASTATA GAL CYC1
1 Kb
,
PLAT1
pLAT3D . .-
GAL
A
0
n X
-5 .I C
pLAT2 ADH1 >-
Figure 1. Constructions of the expression cassettes used in this work. Details of the constructions and references of plasmids may be found in the text. The arrow indicates the location and direction of transcription of the LDH-A cDNA gene.
out without aeration and pH control. Foam was suppressed by the occasional addition of drops of diluted and sterilized antifoam [poly(propylene glycol) 20001. Plasmids. Standard DNA manipulations were performed according to Sambrook et al. (1989). Plasmid PLAT1 was constructed by inserting the 1.75 kbp XbaIHindIII cDNA-LDH-A (encoding the bovine lactate dehydrogenase) from plasmid pLDH12 (Ishiguro et al., 1990) (kindly provided by N. Ishiguro) in a n XbaIHindIII-cut pEMBLyex4 plasmid, downstream from the UASGAJI'ATACYC~ promoter sequences (0.6 kbp). pEMBLyex4 is a derivative of pEMBLyex2 (Baldari et al., 19871, in which the XbaI site in the 2p region was eliminated. The expression cassette obtained, UASGAL/ TAThYcl-cDNA (LDH-A),was not completely repressed by glucose or induced by galactose-containing medium, but high and comparable levels of expression have been observed using both carbon sources. Such a lack of transcriptional control has been ascribed to the long sequence (0.15 kbp) upstream of the first ATG codon in the cDNA (LDH-A) gene (Ishiguro et al., 1990). In fact, full repression by glucose and induction by galactose have been obtained by deleting the whole TATACYCIsequence (0.25 kbp): plasmid pLAT3D. To obtain plasmid pLAT3D, we cut the plasmid PLAT1 with the restriction enzymes XhoI and SstI, made the ends blunt with Klenow, and religated. Constitutive expression of lactate dehydrogenase has also been obtained by cloning the same 1.75 kbp XbaIHindIII cDNA from plasmid pLDH12 in a XbaI-HindIIIcut pVT-U plasmid (Vernet et al., 1987) under the control of the yeast promoter alcohol dehydrogenase (ADH1). The construction of the expression cassettes is summarized in Figure 1. Determination of Cell Number. Small samples of the cultures were sonicated and, after appropriate dilution with Isoton (Coulter Electronics, Harpenden, England), were counted with a Coulter Counter ZBI equipped with a 100 pm orifice (Porro et al., 1991). Determinations of Glucose, Ethanol, Galactose, Lactic Acid, Lactate Dehydrogenase Activity, and Total Cell Protein Content. Glucose and ethanol
determinations were carried out as previously described (Porro et al., 1991). Galactose and lactic acid were determined using Boehringer (Mannheim, Germany) kits (no. 176303 and no. 139084). LDH activity was determined using a Sigma (St. Louis, MO) kit (DG1340-K), while total cell protein determinations were carried out using a Bio-Rad (Bruxelles, Belgium) kit (500-0006).
Results and Discussion LDH-A Expressionin Yeast Cells and Lactic Acid Accumulation. Pyruvate is the end product of glycolysis; it can be further metabolized either by the pyruvate dehydrogenase complex (PDH, EC 1.2.4.1) to acetylcoenzyme A or by pyruvate decarboxylase (PDC, EC 4.1.1.1) to acetaldehyde and subsequently to ethanol (Fiechter et al., 1981; Gancedo and Serrano, 1989). In the yeast S. cereuisiae, most of the pyruvate is channeled through the PDC reaction (Gancedo and Serrano, 1989). The expression of a muscle bovine lactate dehydrogenase gene (LDH-A) in S. cerevisiae cells introduces a new and alternative pathway for NAD+ regeneration, allowing direct reduction of the intracellular pyruvate to lactic acid. Data reported in Figure 2A,B show LDH-A expression and lactic acid production from metabolically engineered GRFlB[pLATl] cells growing on GLU-YNB. The expression of bovine LDH-A (EC 1.1.1.27) is partially repressed by glucose since the highest production of heterologous activity (1.5 unitslmg of total cell proteins) was observed in the stationary phase of growth (Figure 2A). Under such conditions, the highest accumulation of lactic acid (1.1g/L) in the growth medium was obtained during the late exponential phase of growth (Figure 2A). Since yeast cells can efficiently grow and survive at low pH values (Figure 2B), the production of lactate(s) can be avoided (lactic acid pKa = 3.78; Holbrook et al., 1975). However, as PDC (pyruvate decarboxylase(s))and ADH (alcohol deydrogenase(s1,EC 1.1.1.1)activities were present, ethanol was also produced (up to 6 g/L; Figure 2B), strongly lowering the yield of lactic acid (8-lo%, grams of lactic acid produced per gram of carbon source used). Higher yields could be obtained by avoiding the
296
Biotecbnol. frog., 1995, Vol. 11, No. 3
-e
A
25
I
-
20
i l
I :
\
w w m E
15
z
10
3
cl el
w V
a
0
-
i l
0
9
18
36
27
0
TIME ( H O U R S )
16
32
5
-
. -
48
64
B e
80
TIME ( H O U R S )
h
w
0
v
4
B
U
10'
i l
m
6
\
p:
w
m 1 3
10'
10'
4
z el
i l
w
2
V
V 0
10'
107
2
0 0
9
18
27
36
0
TIME ( H O U R S ) Figure 2. Heterologous LDH-A expression and lactic acid accumulation during batch growth of metabolically engineered GRF18[pLAT1] yeast cells growing on YNB-GLU-based medium. (A) Cell number/mL (01, lactic dehydrogenase activity (B) (LDH, 0, unitdmg of total cell proteins), lactic acid (W; Cell number/mL (O), glucose (GLU, 0, g/lOO mL), ethanol (EtOH, 0 , g/L), PH (HI.
a).
synthesis of ethanol. Lower productions of ethanol could be achieved by reducing the pyruvate decarboxylase activity. In addition to being induced by glucose (Hohmann, 1991; Hohmann and Cedeberg, 19901, the expression of the PDC gene(s) seems to be under autoreg ulation at the transcriptional level (Hohmann and Cedeberg, 1990). Furthermore, it has previously been reported that deletion from yeast cells of the P D C l , PDC5, and PDC6 genes results in very poor growth on glucose, with a strong reduction in the overall PDC activity (Hohmann, 1991). The plasmid PLAT1 therefore was introduced into YSH 5.127-17 host cells (a deleted p d c l , pdc5, and pdc6 yeast strain); interestingly, transformed cells were isolated and propagated simply through selection for the ability to grow quickly on glucose. In fact, as a consequence of the expression of cloned LDH activity, a strong increase in the growth rate was observed (Td= 2.2 h versus 28 h for the untransformed cells). Such behavior could be explained by an increase in the metabolic flux related to the regeneration of NAD+ by means of the cloned LDH-A activity. However, in comparison with the GRF18[pLATl] transformed cells, a very low activity of LDH and a corresponding low production of lactic acid were observed (data not shown). Since the K, of the LDH-A for pyruvate is 2 x M (Holbrook et al., 1975), versus 5 x M for the PDC endogenous enzyme(s) (Hohmann and Cedeberg, 19901,
16
32
48
64
80
TIME ( H O U R S ) Figure 3. Lactic acid accumulation during batch growth of metabolically engineered GRF18[pLAT1] yeast cells growing on a mixed 0.2% (w/v) GLU/2% (w/v) GAL-based medium. (A) Cell number/mL (a),glucose (GLU, 0, g/L), galactose (GAL, 0 , g/L). (B) Cell number/mL (O), lactic dehydrogenase activity (LDH, 0, units/mg of total cell proteins), ethanol (EtOH, 0 , g/10 L), lactic acid (W, g/L).
a low concentration of intracellular pyruvate should result in the production of lactic acid, reducing the accumulation rate of ethanol. Theoretically, this attempt could be performed by modulating the intracellular pyruvate level, for instance, using nutrients supporting different glycolytic flow rates; galactose represents a good candidate. Furthermore, PDC synthesis is weak during growth on such a carbon source (Hohmann, 1991). Recombinant GRF18[pLATll yeast cells grew poorly when inoculated directly on galactose-based medium. For this reason, transformed GRF18[pLATll yeast cells were grown on a mixed 0.2% (w/v) glucose/2% (w/v)galactosebased medium (Figure 3A,B). The strategy described allowed fast production of biomass with low accumulation of ethanol during growth on glucose and then bioconversion of galactose in lactic acid, with yields as high as 5070% (grams of lactic acid produced per gram of carbon source used). Improvement in Lactic Acid Production and Productivity. From a biotechnological point of view, the data described in the preceding section are not completely satisfactory due to the low production (1-4 g/L) and low productivity (40-80 mg/L/h) of lactic acid. Figures 2A and 3B show that highest levels of LDH activity have been observed in the stationary phase of growth. On the basis of the assumption that higher LDH-A activity should result in higher production/
Biotechnol. Pro& 1995, Vol. 11, No. 3 70 1
60
60
50
50
h
20
70
15
1
16
A
40
0 v
40
12
w
5 L?
30
5
30
8
0
0
20
20
10
10
0 -30-20-10
0
10 20 30 4 0 50
4
0 -2.50
-0.50
TIME (HOURS) Figure 4. Lactic acid accumulation from metabolically engineered GRFlS[pLATl] yeast cells during a two-stage batch culture. GRF18[pLAT1] yeast cells were pregrown until the stationary phase under the same condition shown in Figure 2. At time T = 0, glucose was added (62 gL, the initial concentration) and lactic acid production (H, gL), and ethanol (0,g L ) and glucose (0,g L ) behavior were followed.
productivity of lactic acid, attempts to increase the total activity of the LDH-A enzyme during the exponential phase of batch growth (i.e., corresponding to a high availability of carbon source to be bioconverted) have been performed. A constitutive (ADH1) and an inducible (UASGAL) promoter sequence were tested (i.e, plasmids pLAT2 and pLATSD, respectively). However, better heterologous activities as well as better lactic acid productiodproductivity were not achieved (data not shown). To increase the volumetric production of lactic acid, we therefore used a two-stage batch culture. Transformed yeast cells were grown initially under the same conditions shown in Figure 2; in the stationary phase of growth (i.e., corresponding to the highest activity of LDH-A), additional glucose was added to the flask (62 &, final concentration). In order to further channel the intracellular pyruvate through the PDC (Gancedo and Serrano, 1989; Fiechter et al., 1981) and/or LDH reactions, lactic acid production was carried out without aeration and under reduced agitation (Figure 4). Under such conditions, interesting productions of lactic acid were obtained (12.5 g/L), but still with unsatisfactory productivity (285 mg/L/h). To achieve higher cellular density and, thus, to increase the productivity, the pregrowth in batch culture was replaced by a computer-controlled fed-batch process (Figure 5). During pregrowth, the addition of fresh nutrients was settled up according to the ethanol concentration behavior (Porro et al., 1991). It has previously been shown that such system control allows one to obtain recombinant biomass productions as high as 100 g of dry weightJL (Alberghina et a l . , 1991). In order to reduce the PDC activity, the preculture was allowed to grow on the ethanol produced (Schmitt and Zimmermann, 1982; Entian and Zimmermann, 1980))before the readdition of glucose (time T = 0; 62 g/L was the initial glucose concentration). Also, in this case, lactic acid production was carried out without aeration and under reduced agitation. In the first hour, production was similar to that shown in Figure 4, but both higher productivity (11 g/L/h) and yield (50-60%) of lactic acid were observed. The low production of ethanol observed during the first hour could be explained by considering that yeast cells reach the highest PDC activity only 3 h after glucose addition (Schmitt and Zimmermann, 1982). Step additions of lower amounts of glucose allowed us to achieve higher accumulations of lactic acid (Figure 6).
0 3.50
1.50
TIME (HOURS) Figure 5. Lactic acid accumulation from metabolically engineered GRF18[pLATl]yeast cells during a two-stage fed-batch culture. GRF18[pLAT1] yeast cells were pregrown to high cell density (3.2 x lo9 cells/mL) in computer-controlled fed-batch culture. During this first phase, ethanol has been used as the parameter controllingthe addition of fresh mineral medium (see text). At time T = 0, glucose was added (62 gL, the initial concentration). After the addition of glucose, the cell number concentration did not change over time. Lactic acid production (m,gL), and ethanol (0,gL) and glucose (0,g L ) behavior were followed. 70 60
. 20
50
. 15 . 10 20
.5
10 0-
1
0
1
2
3
4
5
TIME (HOURS) Figure 6. Lactic acid accumulation from metabolically engi-
neered GRF18[pIATl]yeast cells during a multistage fed-batch culture. GRF18[pLAT1] yeast cells were pregrown to high cell density (2.1 x lo9 celldmL) in computer-controlled fed-batch culture. At times T = 0 and T = 1, glucose was added at a concentration of 15 g/L, while at times T = 2 and T = 3, glucose was pulsed to concentrations of 30 and 40 gL, respectively. The cell number concentration did not change over time. Lactic acid production (H, gL), and ethanol (0,g/L) and glucose (0,g L ) behavior were followed.
Conclusions Among the major sugar fermentations, ethanol by yeast and lactate by Lactobacillus spp. are the most important. The difference between the two productions is related to the enzyme reducing the intracellular pyruvate. In this work, we have shown that the expression of the bovine LDH-A gene in s. cerevisiae cells introduces a new and alternative pathway for regenerating intracellular NAD+,leading to the production of lactic acid by means of a reduction in intracellular pyruvate. Lactic acid is a metabolite widely used in food, pharmaceutical, and cosmetic technologies (i.e., preservation of food and as an additive in pharmaceutical and cosmetic agents). Furthermore, today there is growing interest in the production of such an organic acid to be used directly for the synthesis of plastic, biodegradable material(s).
Biotechnol. Prog., 1995, Vol. 11, No. 3
298
The aim of the experiments described here has been the improvement of lactic acid production and productivity by metabolically engineered yeast cells. For such a purpose, different physiological approaches have been developed. The strategies depicted cannot compete with the current productions of lactic acid from bacterial cells. For example, some of the most developed fermentation techniques (i.e., membrane recycle bioreactor, gelatination and enzyme thinning of the carbon source, and electrodialysis fermentations to give a few examples) allow lactate(s) production of 100 g/L or productivity of 60-80 glLJh using Lactobacillus spp. cells (Cheng et al., 1991; Mehaia and Cheryan, 1986; Benninga, 1990). However, productions and productivities obtained from metabolically engineered S. cereuisiae are sufficiently high and reproducible to find, in perspective, interesting applications. In fact, it was anticipated that the manufacture of lactic acid from bacterial cells is, in terms of fermentation technology and downstream processing [for a detailed review, see Benninga (1990)], a complicated matter involving many disciplines. The production of lactic acid from engineered yeast cells might solve some problems related to the production from bacterial cells. Such problems include (i)inhibitory effects caused by the produced acid, (ii) strategies to prevent the lowering of pH, and (iii) purification procedures. Finally, as recently showed by Dequin and Barre (19941, yeast cells able to produce both ethanol and lactic acid could be useful during alcohol fermentations when acidification is required (Le., during brewing, cider making, baking, and oenology).
Acknowledgment We thank Prof. N. Ishiguro and Dr. S. Hohmann for providing the LDH-A gene and YSH 5.127-17C yeast strain, respectively. Research was supported by the National Research Council of Italy, Special Project RAISA, subproject 4. Literature Cited
Adam, A. C.; Polaina, J. Construction of a Saccharomyces cerevisiae strain able to ferment cellobiose. Curr. Genet. 1991, 20,5-8. Alberghina, L.; Porro, D.; Martegani, E.; Ranzi, B. M. Efficient production of recombinant DNA proteins in Saccharomyces cerevisiae by controlled high-cell-density fermentation. Biotechnol. Appl. Biochem. 1991,14,82-89. Bailey, M. J.; Siika-Aho, M.; Valkeajarvi, A.; Penttila, M. E. Hydrolytic properties of two cellulases of Trichoderma reesei expressed in yeast. Biotechnol. Appl. Biotechnol. 1993,17, 65-76. Baldari, C.; Murray, J. A. H.; Ghiara, P.; Cesareni, G.; Galeotti, C. L. A novel leader peptide which allows secretion of a fragment of human interleukin lp in Saccharomyces cerevisiae. EMBO J. 1987,6, 229-234. Benninga, H. A history of lactic acid making; Kluwer Academic Publishers: Dordrecht, The NetherlandsJ990. Buchta, K. Lactic acid. In Biotechnology; Dellweg, H., Ed.; Verlag Chemie: Weinheim, Federal Republic of Germany, 1983; Vol. 3, pp 409-417. Buckholz, R. G. Yeast systems for expression of heterologous gene products. Curr. Opin. Biotechnol. 1993,4,538-542. Cheng, P.; Mueller, R. E.; Jaeger, S.; Bajpai, R.; Iannotti, E. L. Lactic acid production from enzyme-thinned corn starch using Lactobacillus amylovorus. J . Znd. Microbiol. 1991,7,2734. Compagno, C.; Tura, A.; Ranzi, B. M.; Martegani, E. Bioconversion of lactose/whey to fructose diphosphate with recombinant Saccharomyces cerevisiae cells. Biotechnol. Bioeng. 1993,42,398-400. Dequin, S.; Barre, P. Mixed lactic acid-alcoholic fermentation by Saccharomyces cerevisiae expressing the Lactobacillus casei L(+)-LDH. BiolTechnology 1994,12,173-177.
Entian, K.-D.; Zimmermann, F. K. Glycolytic enzymes and intermediates in carbon catabolite repression mutants of Saccharomyces cerevisiae. Mol. Gen. Genet. 1980,177, 345350. Fiechter, A.; Fuhrmann, G. F.; Kappeli, 0. Regulation of glucose metabolism in growing yeast cells. Adv. Microb. Physiol. 1981, 22,2031-2043. Gancedo, C.; Serrano, R. Energy yielding metabolism, 2nd ed.; Rose, A. H., Harrison, J. S., Eds.; Academic Press. Inc.: New York, 1989; Vol. 3, pp 205-259. Hohmann, S. Characterization ofPDC6, a third structural gene for pyruvate decarboxylase in Saccharomyces cerevisiae. J . Bacteriol. 1991,173, 7963-7969. Hohmann, S.; Cedeberg, H. Autoregulation may control the expression of yeast pyruvate decarboxylase structural genes PDCl and PDC5. Eur. J . Biochem. 1990,188,615-621. Holbrook, J. J.; Liljas, A.; Steindel, S. J.; Rossmann, M. G. Lactate dehydrogenase. In The Enzymes; Boyer, P. D., Ed.; Academic Press: New York, 1975; Vol. 11, pp 191-292. Hongo, M.; Nomura, Y.; Iwahara, M. Novel methods of lactic production by electrodialysis fermentation. Appl. Environ. Microbiol. 1986,52,314-319. Inlow, D.; McRae, J.;Ben-Basset, A. Fermentation of corn starch to ethanol with genetically engineered yeast. Biotechnol. Bioeng. 1988,32,227-234. Ishiguro, N.; Osame, S.; Kagiya, R.; Ichijo, S.; Shinagawa, M. Primary structure of bovine lactate dehydrogenase-A isozyme and its synthesis in Escherichia coli. Gene 1990,91, 281285. Ito, H.; Fukuda, K.; Murata, K.; Kimura, A. Transformation of intact cells treated with alkali cations. J. Bacteriol. 1983,153, 163-168. Kotter, P.; Amore, R.; Hollemberg, C. P.; Ciriacy, M. Isolation and characterization of the Pichia stipitis xylitol dehydrogenase gene, XYL2, and construction of a xylose-utilizing Saccharomyces cerevisiae transformant. Curr. Genet. 1990, 18, 493-500. Mehaia, M. A.; Cheryan, M. Lactic acid from acid whey permeate in a membrane recycle bioreactor. Enzyme Microb. Technol. 1986,8, 289-292. Murooka, Y.; Imanaka, T. In Recombinant microbes for industrial and agricultural applications; Marcel Dekker Inc.: New York, 1994; Section 5, pp 723-852. Porro, D.; Martegani, E.; Tura, A.; Ranzi, B. M. Development of a pH controlled fed-batch system for budding yeast. Res. Microbiol. 1991,142, 535-539. Porro, D.; Martegani, E.; Ranzi, B. M.; Alberghina, L. Lactose/ whey utilization and ethanol production by transformed Saccharomyces cerevisiae cells. Biotechnol. Bioeng. 1992a,39, 799-805. Porro, D.; Martegani, E.; Ranzi, B. M.; Alberghina, L. Development of high cell density cultures of engineered Saccharomyces cerevisiae cells able to grow on lactose. Biotechnol. Lett. 1992b,14,1085-1088. Ramakrishnan, S.; Hartley, B. S. Fermentation of lactose by yeast cells secreting recombinant fungal lacatse. Appl. Environ. Microbiol. 1993,59, 4230-4235. Romanos, M. A.; Scorer, C. A.; Clare, J. J. Foreign gene expression in yeast: a review. Yeast 1992,8,423-488. Sambrook, J.;Fritsch, E. F.; Maniatis, T. Molecular Cloning. A Laboratory Manual, 2nd ed,; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY,1989. Schmitt, H. D.; Zimmermann, F. K. Genetic analysis of the pyruvate decarboxylase reaction in yeast glycolysis. J. Bacteriol. 1982,151, 1146-1152. Vernet, T.; Dignard, D.; Thomas, D. Y. A family of yeast expression vectors containing the phage fl intergenic region. Gene 1987,52,225-233. Wong, W. K. R.; Curry, C.; Parekh, R. S.; Parek, S. R.; Wayman, M.; Davies, R. W.; Kilburn, D. G.; Skipper, N. Wood hydrolysis by Cellulomonas fimi endoglucanase coexpressed as secreted enzymes in Saccharomyces cerevisiae. Bio /Technology 1988, 6, 713-717. Accepted November 18, 1994.@ BP940103J Abstract published in Advance ACS Abstracts, February 1, 1995. @
