Lu et al. — In vivo expression of bifunctional ACSO fusion enzyme

Bot. Bull. Acad. Sin. (1999) 40: 107-114

(Invited Paper)

Expression of a novel ethylene-producing bifunctional fusion enzyme in yeast

Bing Wen Lu, Bing Yu, and Ning Li¹

Department of Biology, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

(Received April 28, 1998; Accepted May 19, 1998)

Abstract. The 1-aminocyclopropane-1-carboxylic acid (ACC) dependent ethylene biosynthetic pathway of higher plants was reconstituted in yeast (Saccharomyces cerevisiae). The ACC-dependent ethylene biosynthesis in yeast was catalyzed by a novel bifunctional ACC synthase-ACC oxidase (ACSO) fusion enzyme (Ning Li, Xiang Ning Jiang, Guo Ping Cai and Shang Fa Yang [1996] The Journal of Biological Chemistry 271: 25738_25741). This fusion enzyme ACSO which was further fused to glutathione S-transferase is capable of converting yeast endogenous S-adenosyl-L-methionine (AdoMet) to ethylene. The molecular weight of the fusion enzyme, GST-ACSO, expressed in yeast, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), was 110 kDa. The ethylene production rate of the yeast cell containing GST-ACSO fusion enzyme was determined to be 21.4 pmol hr^-1 per 10⁸ cells at 8 h of galactose induction. The level of ACC, produced as an intermediate during the sequential reactions from AdoMet to ethylene, was found to increase gradually after galactose induction. Because ACSO is capable of producing ethylene from the ubiquitous and prevalent AdoMet in the living eukaryotic cell and the method commonly used to measure ethylene is simple, fast, and extremely sensitive (0.03 parts per billion), we anticipate this bifunctional fusion enzyme to be useful in the near future for research in molecular biology, developmental biology, fermentation, and genetic engineering.

Keywords: ACC oxidase; ACC synthase; ACSO; Bifunctional; Ethylene; Fusion enzyme; Metabolic engineering; Reporter gene; Yeast.

Abbreviations: aa, amino acid; ACC, 1-aminocyclopropane-1-carboxylic acid; AdoMet, S-adenosyl-L-methionine; PCR, polymerase chain reaction; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.

Introduction

The simple organic molecule, ethylene, has a number of complex biological functions. It regulates the growth and development of higher plants in many ways. These effects of ethylene on higher plants include promoting fruit ripening, flower and leaf senescence, organ abscissions, inhibiting seedling elongation, stimulating root initiation, promoting initiation of the flowering of pineapples, and increasing latex flow in rubber trees (Abeles et al., 1992; Lieberman, 1979; Yang and Hoffman, 1984). Some of ethylene's effects are less desirable, such as when the objective is to prolong the shelf-life of climacteric fruit and flowers. Others are more desirable, such as when the objectives are to enhance the controlled abscission of flowers or young fruits (referred to as thinning), to increase cotton ball opening, to sustain latex flow, to stimulate adventitious root initiation and to promote the female flower differentiation in Cucurbitaceae and the flowering of pineapple plants. To manipulate the extent of the ethylene effect on these agricultural and horticultural crops, many registered

chemical compounds such as aminoethoxyvinyl glycine (AVG) and silver thiosulfate (STS, Ag(S₂O₃)₂^_3 ), inhibitors of ethylene biosynthesis and action, respectively, and ethephon, the "liquid ethylene," have been developed (Maynard and Swan, 1963; Owens et al., 1971). Exogenous application of these chemicals onto crops of commercial value has been shown to increase their productivity and quality (Abeles et al., 1992). However, elucidation of the ethylene biosynthesis pathway in higher plants (Adams and Yang, 1979) and recent successful cloning of the key genes involved in the ethylene biosynthesis offer a good opportunity to either enhance or repress ethylene production endogenously via genetic engineering (Hamilton et al., 1990; Nakajima et al., 1990; Sato and Theologis, 1989; Van der Straeten et al., 1990).

Ethylene is synthesized through the "ACC-dependent" pathway in higher plants. In this pathway, the precursor, S-adenosyl-L-methionine (AdoMet), a ubiquitous and prevalent compound involved in many biological reactions including polyamine biosynthesis and biological methylation of polysaccharides, polypeptides, and nucleic acids (Tabor and Tabor, 1984), is first converted to 1-aminocyclopropane-1-carboxylic acid (ACC) and finally to ethylene (Adams and Yang, 1979). Continued production of ethylene via this pathway is sustained by replen

¹Corresponding author. Fax: 852-2358-1559; E-mail: boningli@ust.hk

Botanical Bulletin of Academia Sinica, Vol. 40, 1999

ishing the AdoMet via the methionine cycle at the expense of ATPs (Yang and Hoffman, 1984). The two carbons of ethylene are ultimately derived from the ribose portion of ATP. The two key enzymes catalyzing the formation of ethylene from AdoMet are ACC synthase (S-adenosyl-L-methionine methylthioadenosine-lyase, EC 4.4.1.14) and ACC oxidase (Boller et al., 1979; Dong et al., 1992; Ververidis and John, 1991; Yu et al., 1979). Genes encoding the two key enzymes have recently been cloned (Hamilton et al., 1990; Nakajima et al., 1990; Sato and Theologis, 1989; Van der Straeten et al., 1990) and expressed in heterologous systems as functional enzymes (Li et al., 1992; Li and Mattoo, 1994; Nakajima et al., 1990; Sato and Theologis, 1989; White et al., 1994; Zhang et al., 1995). Several experiments utilizing antisense ACC synthase and ACC oxidase RNA techniques designed to repress ethylene biosynthesis in fruit and flowers have successfully delayed fruit-ripening and flower-fading in these transgenic plants (Hamilton et al., 1990; Oeller et al., 1991). These successful approaches using transgenic plants have set examples for further experiments with the antisense or sense ACC synthase and ACC oxidase transgenes in transgenic plants. Since recent molecular biological studies on the two genes have revealed that the expression of both genes is regulated developmentally and environmentally (Kende, 1993), it is, therefore, possible that ACC synthase and ACC oxidase gene may not co-express in a concerted fashion at a specific developmental stage or in a specific cell-type of interest. Fusion of ACC synthase and ACC oxidase into a bifunctional enzyme will facilitate the genetic engineering of ethylene biosynthesis in plants and other living organisms such as yeast.

Genetic fusion as a means of creating bi- or multi-functionality has commonly been used to study some of the fundamental problems in biology (Champe and Benzer, 1962; Silhavy and Beckwith, 1985; Yourno et al., 1970). When this genetic fusion approach was adopted in metabolic engineering, the superiority of the bifunctional fusion enzyme over two separate enzymes catalyzing the same multi-step sequential reaction was clearly demonstrated in the hybrid enzymes (Bulow et al., 1985; Ljungerantz et al., 1989; Lindbladh et al., 1992). When E. coli b-glucuronidase (gus), a commonly used scorable marker (Jefferson et al., 1987), was fused to some efficient selectable markers such as neomycin phosphotransferase (npt II) or metallothionein protein, the resulting fusion proteins became ideal scorable as well as selectable artificial bifunctional markers (Datla et al., 1991; Elmayan and Tepfer, 1994). In the case of genetic engineering of ethylene biosynthesis, an artificial bifunctional enzyme (ACSO) was created by in-frame fusion of ACC synthase and ACC oxidase (Li et al., 1996b). This bifunctional fusion enzyme has two active sites juxtaposed in close proximity and catalyzes two sequential steps of ACC-dependent ethylene biosynthesis. It produced both ACC and ACC-dependent ethylene in the in vitro assay system (Li et al., 1996b). In this manuscript,

we report the in vivo expression of the bifunctional enzyme ACSO in a model genetic system, yeast. We have clearly demonstrated that the GST-ACSO multi-functional enzyme is capable of generating ethylene via an ACC-dependent pathway in yeast cells. The ACC-dependent ethylene biosynthesis pathway of higher plants was successfully reconstituted in a heterologous eukaryotic system. This success enables the microbes to produce ethylene via three different pathways: 2-keto-4-methythiobutyric acid, 2-oxoglutarate (Fukuda and Ogawa, 1991) and 1-aminocyclopropane-1-carboxylic acid (this report).

Materials and Methods

Vectors, Reagents and Bacterial and Yeast Strains

The bacterial strain and vector used to overexpress the fusion enzyme were BL21(DE3) and pET14b (Novagen, Madison, WI, USA), respectively. The Escherichia coli strain INVaF' was used for subcloning and construction of the recombinant plasmids and was purchased from Invitrogen BV (The Netherlands). The plasmid vector pCR^TM II (Invitrogen BV, The Netherlands) was used to clone polymerase chain reaction (PCR) products. The yeast strain used to express the fusion protein was S. cerevisiae BJ5459 (MATa, ura3-52, trp1, lys2-801, leu2D1, his3D200, pep4::HIS3, prb1D1, 66R can1 GAL) kindly provided by Dr. D. K. Banfield (Department of Biology, HKUST, Hong Kong). The GST expression vector, pEG(KG) abbreviated as pEG (Mitchell et al., 1993), was made from a high-copy, galactose-inducible shuttle vector as previously described (Baldari et al., 1987). Goat anti-GST antibodies were purchased from Pharmacia. ACC and rabbit anti-goat IgG antibodies were bought from Sigma.

Yeast Transformation

Saccharomyces cerevisiae BJ5459 was transformed with pEG vector or pEG-ACSO recombinant expression plasmid through the lithium acetate transformation method followed by selection on a synthetic medium lacking uracil (Hill et al., 1991). Yeast transformants were confirmed by restriction enzyme digestion analysis of plasmid DNA extracted from the transformed yeasts.

Quantitation of S. cerevisiae Cell Number

The optical density at 600 nm (OD₆₀₀) of each sample of a yeast culture dilution series was first measured. The corresponding cell density of each sample was then quantified by plating out an aliquot of yeast culture on to Yeast extract-Peptone-Dextrose (YPD) plates followed by cell counting after incubation at 30°C for two days. Thus, a correlative standard curve between cell density and the optical density was established. The cell density of the transformed yeast culture used to study the ACSO fusion enzyme expression was measured according to this standard curve throughout the experiment.

Lu et al. — In vivo expression of bifunctional ACSO fusion enzyme

Expression of and in vivo Assay for the ACSO Fusion Enzyme

Yeast cells hosting the pEG-ACSO plasmid or the parental vector were streaked onto solid complete dropout minimal medium without uracil and grown at 30°C for 48_72 h. Single colonies were picked and inoculated into 4 ml of YPD_dextrose medium (yeast extract 10 g/liter, peptone 20 g/liter, dextrose 20 g/liter, pH 5.5) at 30°C for one day with constant shaking at 250 rpm. Two mililiters of the culture were inoculated into 50 ml of YPD_dextrose medium and cultured until the OD₆₀₀ of yeast culture reached 1.3_1.5. The cells were spun down at 4,000 rpm for 4 min. The YPD medium was decanted, and the cell pelletes were washed with 5 ml of YPD medium without dextrose. The cells was resuspended either in YPD_galactose medium (yeast extract 10 g/liter, peptone 20 g/liter, galactose 20 g/liter, pH 5.5) for induction of the expression of the fusion protein or in YPD_dextrose medium again for the non-induction control. The cells were returned to a 30°C shaker. Aliquots of 4 ml of the culture were removed every 4 h and placed into a sterile 10 ml-test tube. Iron(II) sulfate-7-hydrate and sodium ascorbate were added to a final concentration of 100 µM and 30 mM, respectively. The tube was sealed with a rubber serum stopper. Two ml of gas removed from the head-space was replaced with 2 ml of pure CO₂ to a final CO₂ concentration of 20%. The test tubes containing the culture were put back into the 30°C shaker for another 4 h before 2 ml of the gas inside the tube was removed and assayed by gas chromatography. A half mililiter of the culture was centrifuged at 12,000 rpm for 2 min, and 500 µl of the medium was assayed for ACC content as described previously (Lizada and Yang, 1979).

Affinity Purification of GST-ACSO Fusion Enzyme

Yeast cells over-expressing the GST-ACSO fusion enzyme were broken by sonication followed by centrifugation at 12,000 g for 30 min to remove cell debris. The protein extract was first filtered through a 0.22 µm filter before being applied to glutathione agarose column. The protein bound to the agarose column was washed once with 1 × PBS buffer to remove the non-specifically bound proteins. The GST-ACSO fusion enzyme was eluted off with an elution buffer containing 50 mM Tris/HCl, pH 8.0, 10 mM reduced glutathione. The eluates were concentrated on Centricon-30 (Amicon Ltd, USA). The concentrated eluate and un-concentrated eluate samples were then analysed by SDS-PAGE.

SDS-PAGE and Western Blot Analysis

SDS polyacrylamide gel was prepared as previously described (Laemmli, 1970). Yeast protein samples were prepared by suspending the cells in ACO extraction buffer (Li et al., 1996b), followed by sonication. Cell debris was removed by centrifugation. The protein concentration was determined by the method of Bradford (Bradford, 1976) using bovine serum albumin (BSA) as a standard. The proteins fractionated by SDS-PAGE were visualized by

Coomassie Brilliant Blue-250 staining. To make a Western blot of proteins separated on SDS-PAGE, the protein samples were first transferred onto nitrocellulose membrane by electro-blotting (Sambrook et al., 1989). The membrane was then air-dried followed by non-fat milk blocking (5% non-fat milk in 1 × PBS). The non-fat milk pretreated blot was incubated with a 1:100 dilution goat anti-GST primary antibody (Pharmacia) for one hour at room temperature and washed three times with washing buffer containing 20 mM Tris/HCl, pH 7.5, 500 mM NaCl, 0.2% Triton X-100, and 0.05% Tween 20. The antigen-primary antibody complex on the membrane was treated with a 1:2000 dilution secondary antibody, horseradish peroxidase-conjugated rabbit anti-goat IgG, for one hour at room temperature. After four washings with 1 × PBS buffer, GST-ACSO fusion enzyme was visualized with ECL substrate (Bronstein et al., 1994) according to the manufacture's protocol (Amersham).

Results

Construction of ACSO Expression Plasmid

The ACSO fusion fragment was excised out with the restriction enzymes Sac I and Nhe I from pETACSO recombinant plasmid encoding a hybrid enzyme made of a soybean ACC synthase and a tomato ACC oxidase and possessing both ACC synthase and ACC oxidase activities (Li et al., 1996b). Soybean ACC synthase cDNA (GMCACCS1) and tomato ACC oxidase cDNA (LEETHYBR or pCR13) were isolated by PCR as previously reported (Li et al., 1996b). The parent vector pEG was digested with Xba I and Sal I. Because the cohesive ends generated by Xba I and Nhe I restriction digestions are compatible, the ACSO fusion gene was linked into the pEG vector by adding a Sal I-Sac I adapter in ligation mixture. The ACSO fusion was fused to the C-terminus of glutathione S-transferase (Figure 1). The fusion protein is expressed from a CYC1 promoter under the control of a galactose-inducible upstream activator sequence (Guarente and Ptashne, 1981). Orientation of the insert with respect to the promoter was determined by restriction enzyme mapping. A recombinant plasmid with the insert constructed in the sense orientation was named pEGACSO (Figure 1). All steps were carried out by standard methods (Sambrook et al., 1989). For the purpose of conciseness, the yeast hosting the pEGACSO plasmid will henceforth be called pEGACSO yeast, while the yeast hosting the parental vector, pEG(KG) will be called pEG yeast.

Expression of the ACSO Hybrid Enzyme

Both pEGACSO and pEG yeast were grown in a synthetic medium missing uracil and supplemented with 2% raffinose. The expression of GST-ACSO fusion hybrid enzyme was induced with 4% galactose when cell density reached 0.8_1.0 × 10⁸ cells per mililiter. Aliquots of yeast cells at different induction times were harvested followed by sonication. The crude yeast protein extracts

Botanical Bulletin of Academia Sinica, Vol. 40, 1999

with the in vitro assay for the bifunctional activity of this fusion enzyme, the ACC oxidase activity of the ACSO fusion enzyme was quite unstable, probably as a result of structural interference between ACC synthase and ACC oxidase. In contrast, during our recent experiment, we found that a direct fusion of GST to a banana ACC oxidase generated a fusion enzyme with a specific activity of 200 nmol mg^-1 h^-1 after one step purification through GST column (unpublished results). That was 33.3 fold more enzyme specific activity than previously reported for ACSO enzyme (Li et al., 1996b), suggesting that the structural interference between GST and ACC oxidase is minimal. We, therefore, plan to use GST protein or the C-terminal part of GST protein as a linker and insert it in between ACC synthase and ACC oxidase in making the new version of ACSO fusion enzyme.

Figure 1. Map of the pEGACSO recombinant plasmid. Gal1-10UAS, galactose-inducible upstream activator sequence; CYC1 (Hahn et al., 1985), yeast iso-1-cytochrome c gene promoter; GST, glutathione S-transferase gene of Schistosoma japonicum (Smith et al., 1987); ACS, soybean ACC synthase; ACO tomato ACC oxidase; 2 µ, 2 µ origin of yeast 2 µ plasmid which allows propagation in yeast; URA3, orotidine-5-phosphate decarboxylase gene encoding a key enzyme in uracil biosynthesis.

from both induced and uninduced pEGACSO as well as the control pEG yeast cells were analyzed by SDS-PAGE. The expression of GST-ACSO fusion enzyme in yeast under the induction of galactose was visualized via Western blot analysis using anti-GST antibody (Figure 2A). The calorimetric reaction failed to detect the GST-ACSO fusion enzyme in the pEG yeast (Figure 2A: lane 1), the galactose-induced pEG yeast (Figure 2A: lane 2), or the uninduced pEGACSO yeast samples (Figure 2A: lane 3). A 110 kDa protein was detected in all three pEGACSO yeast samples induced for various time periods (Figure 2A: lane 4_6). The expression level of the GST-ACSO fusion enzyme was approximately the same in all three samples at different time intervals upon galactose-induction (Figure 2A, lane 4_6). It was estimated that the level of the expressed fusion enzyme was lower than 0.1% of the total protein in the extract.

To verify whether the GST-ACSO fusion enzyme retained both ACC synthase and ACC oxidase acitivties in vitro, the crude fusion enzyme extract was passed through the glutathione agarose column followed by 10 mM glutathione elution. SDS-PAGE analysis of the eluate showed a major protein band of 110 kDa, which matched very well with the predicted size of the GST-ACSO fusion enzyme (Figure 2B: lane, 1 and 2). The fusion enzyme purified through glutathione agarose column was assayed for both ACC synthase and ACC oxidase activity. It was found that only ACC synthase activity was maintained through the glutathione agarose column purification. The specific activity of ACC synthase was determined to be 0.6 µmol hr^-1 mg^-1. The ACC oxidase activity was not detected from the GST-column purified fusion enzyme. This may have resulted from the inactivation of ACC oxidase activity during the purification process. Based on our previous experience

Figure 2. Expression of GST-ACSO fusion protein. A: Western blot analysis of GST-ACSO fusion protein. Goat anti-GST antibody was used as the primary antibody. Horseradish peroxidase conjugates rabbit anti-goat IgG antibody was used as the secondary antibody. The immuno-complex was detected by using the ECL substrates (Amersham). Lane 1, pEG yeast, not induced; lane 2, pEG yeast, galactose-induced; lane 3, pEGACSO yeast, not induced; lane 4_6, pEGACSO yeast induced by galactose and collected at 4, 8 and 12 h after induction, respectively. B: SDS-PAGE analysis of the purified GST-ACSO fusion protein. 12% SDS polyacrylamide gel was used to resolve the proteins followed by Coomassie Brilliant Blue-250 staining. Lane M, protein markers; lane 1, concentrated eluate; lane 2, unconcentrated elutate; lane 3, crude extract; lane 4, a flow-through sample.

Lu et al. — In vivo expression of bifunctional ACSO fusion enzyme

In vivo Production of ACC-Dependent Ethylene

Ethylene production in both pEG yeast and pEGACSO yeast was examined first. The data showed that a basal level of ethylene production (less than 0.1 nl per 10⁸ cells per hour) existed in yeast under non-induction conditions. To examine whether the basal level of ethylene produced by yeast is ACC-dependent as well as fusion enzyme-dependent, ACC was added into the galactose-induced pEG yeast to different final concentrations. The ethylene production levels of these yeast samples were found to be similar to that of the control (Figure 3). Incubation of the galactose-induced pEG yeast with the used growth medium in which the pEGACSO yeast was grown indicated that the high level of ethylene is not associated with pEG yeast (Figure 3). In contrast, ethylene production from the pEGACSO yeast was dramatically higher than that of pEG yeast plus ACC (Figure 3). This data demonstrated that the observed high level of ethylene production in pEGACSO yeast is GST-ACSO fusion enzyme-dependent as well as ACC-dependent. Further investigating the levels and induction kinetics of ethylene and ACC production in the pEGACSO yeast showed that ACC production continued to increase 12 h after galactose-induction (Figure 4) whereas ethylene production peaked 8 h after galactose-induction and dropped one-third by 12 h after induction, presumably, due to the instability of ACC oxidase of the fusion enzyme and the inactivation of ACC oxidase during the conversion of ACC to ethylene. In contrast to the results observed in pEGACSO yeast, both the pEG yeast and the uninduced pEGACSO yeast produced only small amounts of basal level ethylene over a period of 12 h (Figure 4).

A

B

Figure 4. ACC and ethylene production by the pEGACSO yeast and the pEG yeast. A: The time course of ACC production by both pEGACSO and pEG yeast cells. For each sample, aliquot of cell culture was examined for cell density and then the cell culture was spun at 12,000 rpm for 2 min. Five hundred µl of the growth medium was assayed for ACC content; B: The time course of ethylene production by both pEGACSO and pEG yeast cells. For each sample, aliquots of 4 ml of the cell culture were removed every 4 h into a sterile 10 ml-test tube. Iron (II) sulfate-7-hydrate (100 µM), sodium ascorbate (30 mM) and carbon dioxide (20%) were added. The mixture was incubated at 30°C for 4 h before 2 ml of the gas in the head-space of the test tube was removed and assayed by gas chromatography. Data were presented as means ± SD of triplicate measurements.

Treatment of A11 Tomato Fruit with the Gas Produced by pEGACSO Yeast

The A11 tomato fruit is an ethylene biosynthesis deficient fruit originally produced via over-expression of a complementary ACC synthase RNA, which somehow abolished the level of endogenous ACC synthase translation. The A11 tomato fruits can not ripen in the absence of exogenous ethylene. We placed the mature A11 tomato fruits together with the galactose-induced pEGACSO yeast (400 ml of cells) in a 8.8-liter airtight plastic container for nine days. It was found that ethylene produced by the pEGACSO yeast accumulated to 0.35 ppm during this time and enhanced A11 tomato fruit ripening (data not shown here).

Figure 3. Ethylene produced by the pEGACSO yeast and the pEG yeast with exogenous ACC. Column 1, pEG yeast induced by galactose; column 2, pEG yeast induced by galactose plus 0.5 mM of ACC; column 3, pEG yeast induced by galactose plus 1.0 mM of ACC; column 4, incubation of galactose-induced pEG yeast with the used growth medium in which the pEGACSO yeast was grown; column 5, pEGACSO induced by galactose. Data were presented as means ± SD of triplicate measurements.

Botanical Bulletin of Academia Sinica, Vol. 40, 1999

Discussion

Addition or/and removal of a polypeptide to both the N-terminus and the C-terminus of ACC synthase has been shown to alter ACC synthase activity rather than inactivate it (Li and Mattoo, 1994; Li et al., 1996a and 1996b). Fusion of ACC synthase and glutathione S-transferase to the N-terminus of ACC oxidase also preserves the activity of ACC oxidase (Li et al., 1996b, unpublished results). Therefore, we believed that fusion of GST, ACS and ACO in the given order should have little effect on the bifunctionality of the fusion enzyme even though the fusion might change the kinetics of both enzymes to some extent compared to those of each individual enzyme. The actual alteration in enzymatic kinetics of the fusion enzyme remained to be investigated.

The in-vitro assay conditions for ACC synthase had been established long before those for ACC oxidase had (Boller et al., 1979; Dong et al., 1992; Ververidis and John, 1991; Yu et al., 1979). Carbon dioxide was reported to stimulate ACC oxidase activity both in-vivo and in-vitro (Dong et al., 1992; Misutani et al., 1995; Yang and Hoffman, 1984), and the pH-optima for ACC oxidase in the presence of 20% CO₂ was found to be 6.7 (Misutani et al., 1995). In our experiment with pEGACSO yeast, although we did not study systematically the activation effect of CO₂ on the ethylene production by pEGACSO yeast, when assaying for ACC oxidase activity or ethylene production by pEGACSO yeast, the co-factor ferrous iron, activator CO₂, and the co-substrate ascorbic acid were all added to the reaction mixture. The pH used in this reaction mixture was set at 6.7 in order to make the reaction more efficient, and we assumed that the cellular pH value would change accordingly under the given conditions. This reaction buffer has proven to be successful (Figures 3 and 4) although it might not be an optimal reaction condition for pEGACSO yeast.

It has been reported that some strains of S. cerevisiae can produce ethylene using methionine as a precursor (Thomas and Spencer, 1977). However, microbes, including S. cerevisiae, synthesize ethylene through pathways different from that of higher plants (Fukuda et al., 1984; Fukuda and Ogawa, 1991). The fact that pEG yeast can not utilize exogenous ACC to produce ethylene demonstrated that the significantly higher ethylene production by the pEGACSO yeast was not derived from an ACC-independent pathway. Moreover, in view of the possible shortage of fossil resources in the future, ethylene, the flamable gas, could become a substitute for the widely used town-gas. Reconstitution of Yang-cycle-dependent ethylene production in an ethylene-generating microbe, in which the "ACC-independent" ethylene production pathway operates, is expected to enhance the ethylene production rate of the transgenic microbe. Thus the ethylene conversion efficiency of this microbe will be enhanced, assuming the consumption of the nutrient supply by this organism is unchanged.

At the climacteric stage of apple fruit ripening, the ethylene production rate of apple fruit tissue could reach

15 nmol per hour per gram of fresh tissue (Yip et al., 1991). In case of ACSO fusion enzyme (Li et al., 1996b), the amount of ethylene produced (5.97 nmol h^-1 mg^-1) in one hour by one mg of fusion enzyme from in vitro assay mix could be used to trigger the ripening of 1327 gram of tomato fruit, provided that 0.1 ppm of ethylene is sufficient to do so (Oeller et al., 1991). If the pEGACSO fusion yeast is applied, 10-ml of pEGACSO yeast cells (about 10⁷ cells per ml) could produce about 12 nl ethylene in one day, which would be sufficient to trigger the ripening of 120 grams of A11 tomato fruit. In this experiment, ethylene produced by pEGACSO yeast did indeed enhance A11 tomato fruit ripening (data not shown). Considering that the ACC synthase level in apple fruit is about 0.0021% of total cellular protein (Dong et al., 1991), and whereas the ACC oxidase level in the same fruit is about 0.56% (Dong et al., 1992), we predict that enhancement of the cellular level of GST-ACSO fusion enzyme or modification of the enzyme kinetics of the fusion enzyme in yeast cells will increase the ethylene production rate of pEGACSO yeast so that fewer pEGACSO yeast cells could be used to induce more fruit ripening.

This ethylene-generating fusion enzyme is also anticipated to be useful for both theoretical and applied research. The ACSO fusion gene is fully expected to be used in study of the physiological role ethylene plays in various types of plant development, such as flowering in transgenic plants. Apart from its benefits to plant research, of obvious commercial importance is the use of ACSO as a phytohormone-producing enzyme in plant biotechnology in place of an ethylene-releasing compound such as ethephon (Maynard and Swan, 1963). Ethylene could also be used as a new class of gaseous indicator or reporter for monitoring the gene expression level in a heterologous model system because of the simple (simply draw the air from the sample container), fast (less than two minutes) and extremely sensitive (0.03 nl/Liter by laser-based photoacoustic detector, Woltering et al., 1988) detection methods available in the measurement of ethylene.

Acknowledgments. This research was supported by grants from the Croucher Foundation (CF94/95.SC20), the Research Grant Council of Hong Kong (HKUST649/96M), the Biotechnology Research Institute (BRI-96-III-3) and UGC Research Infrastructure Grant (RI 95/96.SC08) awarded to Dr. Ning LI.

Literature Cited

Abeles, F.B., P.W. Morgan, and M.E. Saltveit. Jr. 1992. Ethylene in Plant Biology, 2nd edn., Academic Press, Inc., San Diego, CA, USA.

Adams, D.O. and S.F. Yang. 1979. Ethylene biosynthesis: identification of 1-aminocyclopropane-1-carboxylic acid as an intermediate in the conversion of methionine to ethylene. Proc. Natl. Acad. Sci. USA 76: 170_174.

Baldari, C., J.A.H. Murray, P. Ghiara, G. Cesareni, and C.L. Geleotti. 1987. A novel leader peptide which allows efficient secretion of a fragment of human interleukin 1 beta in Sac

Lu et al. — In vivo expression of bifunctional ACSO fusion enzyme

Li, N. and A.K. Mattoo. 1994. Deletion of the carboxyl-terminal region of 1-aminocyclopropane-1-carboxylic acid synthase, a key protein in the biosynthesis of ethylene, results in catalytically hyperactive, monomeric enzyme. J. Biol. Chem. 269: 6908_6917.

Li, N., S. Huxtable, S.F. Yang, and S.D. Kung. 1996a. Effects of N-terminal deletions on 1-aminocyclopropane-1-carboxylate synthase activity. FEBS Lett. 378: 286_290.

Li, N., X.N. Jiang, G.P. Cai, and S.F. Yang. 1996b. A novel bifunctional fusion enzyme catalyzing ethylene synthesis via 1-aminocyclopropane-1-carboxylic acid. J. Biol. Chem. 271: 25738_25741.

Li, N., Z. Wiesman, D. Liu, and A.K. Mattoo. 1992. A functional tomato ACC synthase expressed in Escherichia coli demonstrates suicidal inactivation by its substrate S-adenosylmethionine. FEBS Lett. 306: 103_107.

Lieberman, M. 1979. Biosynthesis and action of ethylene in growth, development, and senescence. Annu. Rev. Plant Physiol. 30: 533_591.

Lindbladh, C., M. Persson, L. Bülow, and K. Mosbach. 1992. Characterization of a recombinant bifunctional enzyme, galactose dehydrogenase/bacterial luciferase, displaying an improved bioluminescence in a three-enzyme system. Eur. J. Biochem. 204: 241_247.

Lizada, M.C.C. and S.F. Yang. 1979. A simple and sensitive assay for 1-aminocyclopropane-1-carboxylic acid. Anal. Biochem. 100: 140_145.

Ljungerantz, P., H. Carlsson, M. Mansson, P. Buckel, K. Mosbach, and L. Bulow. 1989. Construction of an artificial bifunctional enzyme, beta-galactosidase/galactose dehydrogenase, exhibiting efficient galactose channeling. Biochemistry 28: 8786_8792.

Maynard, J.A. and J.M. Swan. 1963. Organophosphorus compounds, I. 2-chloroalkylphosphonic acids as phosphorylating agents. Aust. J. Chem. 16: 596_608.

Misutani, F., J.G. Dong, and S.F. Yang. 1995. Effect of pH on CO₂-activated 1-aminocyclopropane-1-carboxylate oxidase activity from apple fruit. Phytochemistry 39: 751_755.

Mitchell, D.A., T.K. Marshall, and R.J. Deschenes. 1993. Vectors for the inducible overexpression of glutathione S-transferase fusion proteins in yeast. Yeast 9: 715_722.

Nakajima, N., H. Mori, K. Yamazaki, and H. Imaseki. 1990. Molecular cloning and sequence of a complementary DNA encoding 1-aminocyclopropane-1-carboxylate synthase induced by tissue wounding. Plant Cell Physiol. 31: 1021_1029.

Oeller, P.W., M.W. Lu, L.P. Taylor, D.A. Pike, and A. Theologis. 1991. Reversible inhibition of tomato fruit senescence by antisense RNA. Science 254: 437_439.

Owens, L.D., M. Lieberman, and A. Kunishi. 1971. Inhibition of ethylene production by rhizobitoxin. Plant Physiol. 48: 1_4.

Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press.

Sato, T. and A. Theologis. 1989. Cloning the messenger encoding 1-aminocyclopropane-1-carboxylate synthase, the key enzyme for ethylene biosynthesis in plants. Proc. Natl. Acad. Sci. USA 86: 6621_6625.

Silhavy, T.J. and J.R. Beckwith. 1985. Uses of lac fusions for the study of biological problems. Microbiol. Rev. 49: 398_418.

Smith, D.B., K.M. Davern, P.G. Board, W.U. Tiu, E.G. Garcia,

charomyces cerevisiae. EMBO J. 6: 229_243.

Boller, T., R.C. Herner, and H. Kende. 1979. Assay for and enzymatic formation of an ethylene precursor, 1-aminocyclopropane-1-carboxylic acid. Planta 145: 293_303.

Bradford, M.M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248_254.

Bronstein, I, J. Fortin, P.E. Stanley, G.S.A.B. Stewart, and L.J. Kricka. 1994. Chemiluminescent and bioluminescent reporter gene asssays. Analyt. Biochem. 219: 169_181.

Bulow, L., P. Ljungcrantz, and K. Mosbach. 1985. Preparation of a soluble bifunctional enzyme by gene fusion. Bio/technology 3: 821_823.

Bulow, L. and K. Mosbach. 1991. Multienzyme systems obtained by gene fusion. Trends Biotechnol. 9: 226_231.

Champe, S.P. and S. Benzer. 1962. An active cistron fragment. J. Mol. Biol. 4: 288_292.

Datla, R.S.S., J.K. Hammerlindl, L.E. Pelcher, W.L. Crosby, and G. Selvaraj. 1991. A bifunctional fusion between beta-glucuronidase and neomycin phosphotransferase: a broad-spectrum marker enzyme for plants. Gene 101: 239_246.

Dong, J.G., J.C. Fernández-Maculet, and S.F. Yang. 1992. Purification and characterization of 1-aminocyclopropane-1-carboxylate oxidase from ripe apple fruit. Proc. Natl. Acad. Sci. USA 89: 9789_9793.

Dong, J.G., W.T. Kim, W.K. Yip, G.A. Thompson, L. Li, A.B. Bennett, and S.F. Yang. 1991. Cloning of a CDNA encoding 1-aminocyclopropane-1-carboxylate synthase and expression of its mRNA in ripening apple fruit. Planta 185: 38_45.

Elmayan, T. and M. Tepfer. 1994. Synthesis of a bifunctional metallothionein/beta-glucuronidase fusion protein in transgenic tobacco plants as a means of reducing leaf cadmium levels. Plant J. 6: 433_440.

Fukuda, H., T. Fujii, and T. Ogawa. 1984. Microbial production of C₂-hydrocarbons, ethane, ethylene and acetylene. Agric. Biol. Chem. 48: 1363_1365.

Fukuda, H. and T. Ogawa. 1991. Microbial ethylene production. In A.K. Mattoo and J.C. Suttle (eds.), The Plant Hormone Ethylene, Boca Raton, Florida: CRC Press, pp. 279_314.

Guarente, L. and M. Ptashne. 1981. Fusion of Escherichia coli lac Z to the cytochrome c gene of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 78: 2199_2203.

Hamilton, A.J., G.W. Lycett, and D. Grierson. 1990. Antisense gene that inhibits synthesis of the hormone ethylene in transgenic plants. Nature 346: 284_287.

Hahn, S., E.T. Hoar, and L. Guarente. 1985. Each of three "TATA elements" specifies a subset of the transcription initiation sites at the CYC-promoter of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 82: 8562_8566.

Hill, J., K.A.I.G. Donald, and D.E. Griffiths. 1991. DMSO-enhanced whole cell yeast transformation. Nuc. Acid Res. 19: 5791.

Jefferson, R.A., T.A. Kavanagh, and M.W. Bevan. 1987. GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EBMO J. 6: 3901_3907.

Kende, H. 1993. Ethylene biosynthesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44: 283_307.

Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680_685.

Botanical Bulletin of Academia Sinica, Vol. 40, 1999

and G.F. Mitchell. 1987. Mr 26,000 antigen of Schistosoma japonicum recognized by resistant WEHI 129/J mice is a parasite glutathione S-transferase. Proc. Natl. Acad. Sci. USA 84: 6541_6541.

Tabor, C.W. and H. Tabor. 1984. Methionine adenosyltransferase (S-adenosylmethionine synthetase) and S-adenosylmethionine decarboxylase. Adv. Enzymol. 56: 251_282.

Thomas, K.C. and M. Spencer. 1977. L-Methionine as an ethylene precursor in Saccharomyces cerevisiae. Can. J. Microl. 23: 1669_1674.

Van der Straeten, D., L. Van Wiemeersch, H.M. Goodman, and M. Van Montagu. 1990. Cloning and sequence of two different cDNAs encoding 1-aminocyclopeopane-1-carboxylate synthase in tomato. Proc. Natl. Acad. Sci. USA 87: 4859_4863.

Ververidis, P. and P. John. 1991. Complete recovery in vitro of ethylene-forming enzyme activity. Phytochemistry 30: 725_727.

White, M.F., J. Vasquez, S.F. Yang, and J.F. Kirsch. 1994. Expression of apple 1-aminocyclopropane-1-carboxylate synthase in Escherichia coli: Kinetic characterization of wild-type and active-site mutant forms. Proc. Natl. Acad. Sci. USA 91: 12428_12432.

Woltering, E.J., F. Harren, and H.A.M. Boerringter. 1988. Use of a laser-driven photoacoustic detection system for measurement of ethylene prodcution in Cymbidium flowers. Plant Physiol. 88: 506_510.

Yang, S.F. and N.E. Hoffman. 1984. Ethylene biosynthesis and its regulation in higher plants. Annu. Rev. Plant Physiol. 35: 155_189.

Yip, W.K., J.G. Dong, and S.F. Yang. 1991. Purification and characteriation of 1-aminocyclopropane-1-carboxylate synthase from apple fruits. Plant Physiol. 95: 251_257.

Yourno, J., T. Kohno, and J.R. Roth. 1970. Enzyme evolution: generation of a bifunctional enzyme by fusion of adjacent genes. Nature 228: 820_824.

Yu, Y.B., A.D. Adams, and S.F. Yang. 1979. 1-aminocyclopropane-1-carboxylic acid synthase, a key enzyme in ethylene biosynthesis. Atch Biochem. Biophys. 198: 280_286.

Zhang, Z., C.J. Schofield, J.E. Baldwin, P. Thomas, and P. John. 1995. Expression, purification and characterization of 1-aminocyclopropane-1-carboxylate oxidase from tomato in Escherichia coli. Biochem. J. 307: 77_85.