T. Hoshino, Department of Applied Biological Chemistry, Faculty of Agriculture, Niigata University, Ikarashi 2-8050, Nishi-ku, Niigata 950-2181, Japan
Fax: +81 25 262 6854
Tel: +81 25 262 6637
E-mail: hoshitsu@agr. niigata-u.ac.jp
β-Amyrin, a natural triterpene, is widely distributed in the plant kingdom, and its pentacyclic skeleton is produced by oxidosqualene cyclase (OSC). OSC enzymes are classified as membrane proteins, and they catalyze the polycyclization reaction of (3S)-2,3-oxidosqualene to yield nearly 150 different cyclic triterpene skeletons. To date, no report has described the successful purification and characterization of plant β-amyrin synthase. The β-amyrin synthase from Euphorbia tirucalli (EtAS) was expressed as a polyhistidine-tagged protein in Saccharomyces cerevisiae GIL77, which lacks the lanosterol synthase gene. The expression yield, determined by western blotting analysis, was 5–7 mg. By Ni2+–nitrilotriacetic acid affinity column chromatography and careful selection of the proper imidazole concentration during the purification processes of washing and elution, a single band was successfully obtained on SDS/PAGE. We then tested the effects of four detergents on the enzyme activity. Supplementation with Triton X-100 at a concentration of 0.05% yielded the highest activity. The optimal pH and temperature were 7.0 and 30 °C, respectively. The kinetic parameters, Km and kcat, were determined to be 33.8 ± 0.53 μm and 46.4 ± 0.68 min−1, respectively. To the best of our knowledge, there are no reports describing both Km and kcat for OSCs except for two examples of rat and bovine lanosterol synthases. The β-amyrin synthase purified in this study showed a significantly higher catalytic efficiency (kcat/Km) (~ 103-fold) than those of the two reported lanosterol synthases. Gel-filtration HPLC indicated that the OSC exists as a monomer, and the eluted OSC retained its activity. Furthermore, the inhibition constants Ki and IC50 and types of inhibition by iminosqualene, Ro48-8071 and U18666A were determined, and indicated that iminosqualene and Ro48-8071 are potent inhibitors. Additionally, this is the first report of the kinetic data of the mutated enzymes targeted for the DCTAE(485–489) motif, which is a putative initiation site for the polycyclization reaction. No activity of the D485N variant and significantly decreased activity of the C564A variant were found, definitively demonstrating that the acidic carboxyl residue Asp485 serves as a proton donor to initiate the polycyclization reaction, and that Cys564 is involved in hydrogen bond formation with the carboxyl residue Asp458 to enhance the acidity. The CD spectrum is the first to be reported for OSCs, and the CD spectra of the wild-type and the mutated EtASs were almost the same, indicating that the protein architecture was not altered by these mutations.
Naturally occurring cyclic triterpenes are biosynthesized by an enzymatic reaction with squalene or (3S)-2,3-oxidosqualene (1) as the substrate. The structural diversity of cyclic triterpenes is remarkable; ~ 150 additional triterpenoid scaffolds have been identified to date . Lanosterol and cycloartenol can serve as precursors for membrane sterol and steroid hormones. Many genes encoding squalene-hopene cyclases (SHCs) [2, 3] and oxidosqualene cyclases (OSCs) [4, 5] have been identified, and investigations of OSCs from Arabidopis thaliana have provided substantial insights into plant OSCs. OSCs are generally classified into the following three categories: accurately monofunctional OSCs (giving a single or dominant product) [5, 6], multifunctional OSCs (producing a variety of triterpene skeletons, none of which are dominant) [5, 6], and a seco-type triterpene synthase OSC [C–C bond cleavage(s), i.e. Grob fragmentation]. Only three examples of seco-type triterpene synthases have been identified to date: marneral synthase (At5 g42600, PEN5) from A. thaliana , seco-β-amyrin synthase or seco-α-amyrin synthase (At1 g78500, PEN6) , and achilleol B synthase from Oryza sativa .
β-Amyrin (2) is widely distributed in the plant kingdom, and its scaffold is constructed by the OSC enzyme, which folds 1 in a chair–chair–chair–boat–boat conformation in the reaction cavity (Fig. 1) . Proton attack on the epoxide ring initiates a C–C bond-forming reaction with regiospecificity and stereospecificity to give a 6/6/6/5-fused tetracyclic dammarenyl cation, followed by ring expansion to give a 6/6/6/6-fused tetracyclic baccharenyl cation. Further cyclization gives a 6/6/6/6/5-fused pentacyclic lupenyl cation, which undergoes ring expansion to give a 6/6/6/6/6-fused pentacyclic oleanyl cation. The subsequent reactions of hydride shift and deprotonation in an antiperiplanar fashion give the β-amyrin skeleton. Many β-amyrin synthase genes have been identified from dicots – Panax ginseng (PNY) , Pisum sativum (PSY) , Glycyrrhiza glabra (GgbAs1) , Medicago truncatula , tomato , and Gentianastraminea Maxim  – and from a monocot, Avena strigosa , and so on. A variety of other plant OSCs have also been cloned and characterized .
Very little is known of the enzymatic properties of OSCs, owing to the difficulties that are frequently encountered in purifying membrane proteins. To date, among the many known OSCs, only two lanosterol synthases, from human  and bovine liver , have been completely purified. There is no report describing the complete purification and the enzyme characterization of purified plant OSCs. Recently, Kajikawa et al. cloned the gene encoding β-amyrin synthase (EtAS; EC 18.104.22.168) from Euphorbia tirucalli . We achieved the first complete purification of this plant OSC, which retained remarkably high activity. Here, we report the detailed characterization of the OSC, including the kinetic parameters for 1 and for three potent inhibitors. Moreover, the CD spectrum of native EtAS is also reported here for the first time. The DCTAE(485–489) motif, which is highly conserved among OSCs, has been presumed to be responsible for the proton attack on the epoxide ring, but no kinetic data for the site-directed mutated OSCs have been reported. Comparison of the CD spectra between the wild type and the mutants allowed us to determine whether alteration of the protein architecture occurred because of the mutations. We also report here the kinetic data for the mutated EtASs targeted for the motif.
Results and Discussion
Expression of β-amyrin synthase in Saccharomyces cerevisiae and the cyclization products 3 and 4
The pPICZ A plasmid harboring the EtAS coding sequence  was used as the template for PCR amplification of the gene. The amplified gene was ligated into the yeast expression vector pYES2, and then transformed into S. cerevisiae GIL77 lacking the lanosterol synthase gene. The grown yeast cells were collected by centrifugation, and then subjected to heat treatment with 15% KOH/MeOH at 80 °C. Hexane was added to the MeOH solution, and the hexane extract was analyzed with a GC instrument (Fig. 2). As reported by Kajikawa et al. , we also detected β-amyrin by GC/MS. In addition to β-amyrin (2), very small amounts of other two triterpenes, 3 and 4 (m/z 426, M+), were found, although Kajikawa et al. did not refer to the production of 3 and 4. The two products were isolated as follows. The hexane extract was first fractionated by SiO2 column chromatography [hexane/tetrahydrofuran (THF) = 100 : 1] to collect the triterpene fraction, which was acetylated with Ac2O/pyridine. The triterpene acetate mixtures were subjected to repeated separation by normal-phase HPLC with a mixture of hexane and THF (100 : 0.15), leading to the isolation of pure β-amyrin (2). However, separation of the two minor triterpene acetates was unsuccessful, despite the use of RP-HPLC (C18). Complete separation was attained by using a chiral column (CHIRALPACK IB, Daicel; hexane/THF = 100 : 0.05). The former peak (3) on GC was identified as butyrospermol [eupha-7(8),24-diene-3β-ol] and the latter peak (4) was tirucall-7(8),24-diene-3β-ol by the detailed NMR analyses, including two-dimensional NMR spectra (Figs S1 and S2). The structural difference between 3 and 4 is found only in the stereochemistry at C20. The stereochemistry was determined to be 20R for 3, but 20S for 4, on the basis of the reported chemical shifts of Me-21 in 1H-NMR . The detailed NMR analyses of 3 and 4 are described in Figs S1 and S2. The 2/3/4 product ratio was determined to be 96.6 : 1.6 : 1.8 by GC analysis. Formation of the tetracyclic 3 and 4, albeit in negligible amounts, suggests that the polycyclization reaction stopped at the premature cyclization stage. Folding of the chair–chair–chair–chair conformation could give 3, whereas folding of the chair–chair–chair–boat conformation could give 4, according to the reaction scheme of Fig. 3. Formation of the tetracyclic compounds 3 and 4 indicates that the β-amyrin synthase is somewhat loosely packed around D-ring and E-ring formation sites [1-6], but the amounts of 3 and 4 produced were negligible, indicating that this amyrin synthase is a monofunctional OSC.
Purification of the recombinant β-amyrin synthase (pYES2/CT–EtAS)
The EtAS gene was ligated into the pYES2/CT vector, in order to purify the EtAS with an Ni2+–nitrilotriacetic acid column. The construct pYES2/CT–EtAS was transformed into S. cerevisiae GIL77. This strain also produced the two minor triterpenes, 3 and 4, in addition to a large quantity of 2. The product ratio was the same as that of pYES2–EtAS. The β-amyrin synthase is expressed as polyhistidine-tag fusion protein (His6, C-terminus), which enabled complete purification in a single step by use of an Ni2+–nitrilotriacetic acid column (Fig. 4A). The yeast cells from a 0.5-L culture were suspended in 30 mL of binding buffer (20 mm Tris/HCl, pH 7.9, containing 10 mm imidazole, 300 mm NaCl, and 0.1% Triton X-100), and then disrupted with glass beads (60 g). The supernatants obtained by centrifugation were pooled after repeated extractions, and then loaded onto the Ni2+–nitrilotriacetic acid resin column. The bound protein was washed with the buffer containing 40 mm imidazole, and finally eluted with the same buffer solution, which also contained 250 mm imidazole. As shown in Fig. 4A (lane 5), the eluted β-amyrin synthase resulted in a single band on 10% SDS/PAGE. The molecular mass estimated from the SDS/PAGE gel mobility was 77–79 kDa (Fig. S3), and the apparent mass was smaller than the calculated one (91.1 kDa). A significant difference between the calculated and apparent molecular masses is frequently encountered; for example, for fission yeast proteins, the observed difference was reported to be at the level of 10–30% . We also found that the molecular mass of the Rv3378c enzyme from Mycobacterium tuberculosis estimated with SDS/PAGE was significantly lower (~ 20%) than the calculated one , but MALDI-TOF MS analyses of the proteolytic digestion fragments showed that the expressed protein was actually of full length . A gel filtration experiment showed that the actual size of this triterpene synthase was 94.8 kDa (Fig. 4B), indicating that EtAS exists as a monomer, as does human lanosterol synthase . In contrast, bovine liver lanosterol synthase was reported to form a dimeric structure . The amount of EtAS expressed by pYES2/CT was estimated by western blotting using an SDS/PAGE gel. The amount of EtAS expressed from 1 L of culture was estimated to be 5.5–6.7 mg. The purified protein was obtained in ~ 50% yield. To measure the CD spectrum of EtAS, Triton X-100 was replaced by Brij35, because Triton X-100 and Tween-80 have benzene rings and double bonds in the molecules that electronically absorb in the UV region and interfere with the CD spectrum, whereas Brij35 has no such functional groups. The bound EtAS on the Ni2+–nitrilotriacetic acid column, described above, was washed with a buffer (80 mL) containing a mixture of 0.01% Brij35 and 80 mm imidazole, and then eluted with a buffer (4 mL) including 250 mm imidazole (Fig. 4A, lane 6). After dialysis, the CD spectrum was measured (Fig. 5), and showed that the enzyme architecture was not altered at temperatures < 30 °C. The enzymatic activity was significantly lower in Brij35 detergent (Fig. 6A). However, substantial molecular ellipticity was observed, suggesting that a higher-order protein structure was maintained in Brij35 solution. Furthermore, the CD spectrum was quite similar to that of SHC, and was superimposable on that measured in a different type of detergent, β-octylglucoside (β-OG) (Fig. S4A).
Optimal catalytic conditions and kinetic parameters
Figure 6 shows the determination of optimal incubation conditions. The enzymatic activities with respect to the detergent concentrations are shown in Fig. 6A. The following four different detergents were utilized: Triton X-100, Tween-80, β-OG, and Brij35. The highest activity was seen for Triton X-100, at a concentration of 0.05% (w/v). The activity in Tween-80 was nearly equal to that in Triton X-100. In the case of β-OG, reasonably high activities were observed at concentrations less than the critical micelle concentration (CMC, 0.6%), but the activities gradually decreased at higher concentrations. In Brij35, the activities were significantly low, as shown in Fig. 6A. Brij35 (0.01%) was replaced with Triton X-100 (0.05%) by use of a His-tag column, and the activity was then assayed. No decrease in the enzymatic activity was found (Fig. S4B). Thus, there was no protein denaturation by Brij35. No effect of dithiothreitol on the enzymatic activity was observed in the concentration range of 0–10 mm (Fig. S5). In contrast, in the case of the cycloartenol synthase from Ochromonas malhamensis, supplementation with 5 mm dithiothreitol gave an approximately two-fold enhancement of the activity . The optimal temperature and pH for the catalysis were 30 °C and 7.0, respectively (Fig. 6B,C). The kinetic parameters relating to enzymatic activity were measured under optimal catalytic conditions. Figure 6D shows that the enzymatic activities were linearly increased for up to 30 min after the incubation. Kinetic parameters were obtained by estimating the amounts of β-amyrin produced after incubation for 20 min. The specific activity of the OSC was estimated to be 352 ± 11.8 nmol·min−1·mg−1. The Michaelis–Menten plot is shown in Fig. 7. Steady-state parameters were determined by fitting the curve to v = Vmax[S]/(Km + [S]), with the solver program of Microsoft Excel. The values of Km and kcat for (3S)-2,3-oxidosqualene (1) were determined to be 33.8 ± 0.53 μm and 46.4 ± 0.68 min−1, respectively, and kcat/Km was 1.37 ± 0.21 μm−1·min−1 (Table 1). Table 1 lists the kinetic data of other OSCs [19, 25-28] and SHCs [27, 29, 30] that have been reported to date. The difference in Km values between the OSCs was not significant, but the kcat and specific activity of EtAS were appreciably higher than those of the other three OSCs, indicating that the enzymatic activity was retained during the purification process. The kcat and specific activity of SHCs are remarkably higher than those of OSCs, but the reported kinetic data differed appreciably between the three research groups [27, 29, 30]. This may have been caused by the use of different assay methods and different detergents . It is difficult to determine whether the substrate inhibition occurs at higher concentrations, because oxidosqualene substrate 1 was not dissolved at concentrations > 112.7 μm. However, slight inhibition was observed at concentrations > 98.6 μm (Fig. 7), suggesting that substrate inhibition for EtAS had occurred. SHC, one of the triterpene cyclases, also showed substrate inhibition in our experiment (unpublished result). No change in EtAS activity was observed for at least 2 months when it was stored at – 80 °C.
Table 1. Kinetic data for EtAS, lanosterol cyclases and SHCs that have been reported to date. NR, not reported
Kinetics of inhibition of EtAS by iminosqualene (5), Ro48-8071 (6), and U18666A (7)
Figure 8 shows the structure of known OSC inhibitors (5–7). The inhibitory effects of these three compounds on the activities of EtAS were examined (Table 2; Fig. 9). Inhibition constants were determined with Dixon plots . The (3R,S)-racemic mixture of iminosqualene (5) was chemically synthesized according to the reported method [32-34]. The synthetic method for 5 is described in Fig. S6. Compound 5 strongly inhibited the activity of EtAS: IC50 of 30.9 nm, and Ki of 13.4 nm. Corey et al. first reported the potent inhibition of hog liver lanosterol cyclase by 5 . Duriatti et al. reported that 5 also inhibited pea seedling β-amyrin synthase (IC50 of 0.3 μm; Ki of 0.2 μm) and rat liver lanosterol synthase (IC50 of 0.3 μm; Ki of 0.4 μm), and that the inhibition of both enzymes occurred in a noncompetitive fashion , which differed from our result; Fig. 9A indicates that 5 exhibited competitive and potent inhibition of EtAS (IC50 of 30.9 nm; Ki of 13.4 nm), but less potent inhibition of pea seedling β-amyrin synthase and rat liver lanosterol synthase (Table 2). No other report has described the inhibition of OSCs by 5. The nitrogen atom of 5 is protonated under physiological pH to yield an aziridium ion, which mimics the protonated oxirane ring (transition-state structure). Thus, the inhibition appeared to be competitive but not noncompetitive.
Table 2. Comparison of the inhibitory activities against the OSCs by 5–7 and inhibition types. Ki values were determined by Dixon plots . The Ki value of 5 was calculated as the [3R,S]-racemic mixture, because it is likely that both the R-configuration and and the S-configuration behave as the inhibitors. NR, not reported
Table 3. The enzymatic activities of the D485, C486 and C564 variants. ND: the enzyme activity was not detected and was completely lost. The enzyme amounts of the C486A and C564A variants used for obtaining the kinetic data were 5 μg (22 nm) and 10 μg (44 nm), respectively. These kinetic data were determined by incubation at 30 °C. Other incubation conditions were the same as those for the wild type (see Experimental procedures)
Specific activity (nmol·min−1·mg−1)
Activity relative to the specific activity (%)
33.8 ± 0.53
46.4 ± 0.68
1.37 ± 0.21
352 ± 11.8
34.8 ± 3.4
20.4 ± 1.0
0.59 ± 0.03
163 ± 5.7
51.1 ± 4.1
(0.77 ± 3.6) × 10−3
(0.015 ± 6.9) × 10−5
5.79 ± 0.085
Ro48-8071 (6) was also a potent inhibitor (IC50 of 10.7 nm; Ki of 7.0 nm). As shown in Table 2, 6 is a potent inhibitor of the lanosterol synthases from bovines , rats [27, 35], and humans [18, 36, 37], and of the SHC from Alicyclobacillus acidocaldarius [27, 30]. The IC50 values for EtAS (10.7 μm) were determined as shown in Fig. S7, and they were nearly the same as those for the lanosterol synthases and SHC. Lineweaver–Burk plots (Fig. 9B) indicated that the inhibition was mixed noncompetitive. Abe et al. concluded that this inhibitor appeared to be a noncompetitive inhibitor of rat lanosterol synthase, although their experimental data in Lineweaver–Burk plots clearly indicate mixed noncompetitive inhibition . This mode of inhibition indicates that there is a second binding site in addition to the active site. However, X-ray analyses of cocrystals of 6 with human OSC  and with SHC  revealed that this inhibitor is located only in the reaction cavity, suggesting that 6 should behave as a competitive inhibitor. Fluorescence titration experiments showed a 1 : 1 binding ratio for human OSC and 6 , further supporting competitive inhibition. These findings for human OSC and SHC imply that the plant EtAS also undergoes competitive inhibition by 6, i.e. that a secondary binding site is not involved. To explain this inconsistency between the kinetic and the X-ray data, Ruf et al.  and Lenhart et al.  suggested that the biphasic state needed for the enzymatic activity measurements could have influenced the kinetics as a result of altered accessibility of the substrate and inhibitor in mixed micelles. X-ray analysis of the complex of 6 with human lanosterol synthase revealed the following findings : (a) Asp455 is hydrogen-bonded to the nitrogen atom of the amine group; (b) the fluorophenyl group is stacked between the side chains of Phe696 and His232; and (c) the terminal bromophenyl group interacts with Trp192, Trp230, and Phe521. These Asp, Trp and Phe residues are highly conserved among EtAS, bovine OSC, rat OSC, and SHCs (Table S1). These highly conserved residues would have significantly contributed to the potent inhibition of various triterpene cyclases by 6.
It has been reported that U18666A (7) inhibits rat liver lanosterol synthase and pea seedling β-amyrin synthase at IC50 levels of 0.8 and 0.25 μm, respectively, but this inhibition type has not been reported previously . The IC50 and Ki for EtAS were determined to be 0.14 μm and 102 nm, respectively. The inhibition appeared to be noncompetitive (Fig. 9C), so 7 might bind to the outer surface area of EtAS, leading to a loss of enzymatic activity because of the altered protein architecture. The inhibitory activity of 7 was relatively moderate as compared with those of 5 and 6 (Table 2).
Enzymatic activities of site-directed mutated EtAS targeted for the DCTAE motif
The DCTAE motif and the DXDD motif are highly conserved in eukaryotic OSCs and prokaryotic SHCs, respectively (Fig. S8). We have previously reported the enzyme kinetic data for the mutated SHCs targeted for the DXDD motif, and have demonstrated that the DXDD motif plays a crucial role in the initiation of the polycyclization reaction by a proton attack on the terminal double bond of the squalene molecule . It has been predicted that the DCTAE motif corresponding to the DXDD motif works to initiate the polycyclization reaction. The attack of the proton, which is released from the acidic Asp, on the terminal epoxide ring of 1 triggers a cascade of the ring-forming reaction [36, 38]. However, kinetic studies on the site-directed mutated OSCs targeted for the DCTAE motif have not been reported thus far. Corey et al. proposed that Asp456 of the DCTAE motif of S. cerevisiae lanosterol synthase is responsible for initiating the cyclization, on the basis of the result of the viability assay of the D456N variant , but no kinetic data have been provided. On the basis of X-ray crystallographic analysis of human lanosterol synthase, Thoma et al. proposed that the acidity of Asp455 of the DCTAE motif is enhanced by the hydrogen bond formed with Cys456 and Cys533 , but no kinetic evidence for the role of the Cys residues has been reported. To verify the functions of these three residues of EtAS, four variants (D485N, D485E, C486A, and C564A) were constructed. Figure 10 shows the relative activities versus incubation temperatures. The optimum temperatures of the mutants were slightly changed, but were little different from that of the wild type. The enzymatic activities of the D485N and D485E variants were completely lost at all temperatures examined. The activity of the C486A variant was somewhat retained (~ 50% of the wild type), whereas that of the C564A variant was almost completely lost. The kinetic data are shown in Table 3. The absence of activity of the D485N variant indicates that the acidic carboxylic residue Asp serves as a proton donor to initiate the polycyclization reaction, but the Glu variant also showed no activity, despite the acidic residue, indicating that the extra CH2 moiety attached to Asp had interfered with the access of the carboxyl group to the epoxide ring. Thus, an appropriate distance of the carboxyl moiety from the epoxide ring must be present for cyclization to be initiated. Cys564 is more strongly hydrogen-bonded with the carboxyl residue Asp458 than with Cys486, because the C564A variant had significantly decreased activity (1.6% of the wild type), whereas the Cys486-mutated EtAS retained ~ 50% activity (Fig. 10; Table 3). The CD spectra of the four mutated EtASs were the same as that of the wild type (Fig. 11), indicating that the protein architecture was not altered or was only slightly altered by these mutations.
Estimation of in vivo enzymatic activities of mutants
The expression of mutated OSCs is frequently found to be absent or significantly decreased. To analyze in vivo the functions of the site-directed variants, it is crucial to determine both the amounts of the expressed OSCs and the quantities of the triterpenes produced by the mutants; the relative activities must then be compared with that of the wild type. The enzymatic activities are determined by dividing the amounts of product produced in vivo by the amount of EtAS expressed in the mutants and the wild type. Figure 12 shows the relative activities estimated by in vivo experiments. The amounts of protein expressed were estimated by western blot analysis (approximately 5.5–6.7 mg for both the wild type and mutants per liter of culture; Fig. S9), and the quantities of β-amyrin were determined by GC analysis. No other triterpene was found to be produced by these variants. The vertical line indicates the activity of the mutants relative to that of the wild type. No activity was observed for the D485N and D485E variants, as was also observed in the in vitro experiments with the purified enzymes (Fig. 10; Table 3). The C486A variant showed ~ 58% activity relative to the wild type, and the C564A variant only 2.5% activity. These results of the in vivo experiments (Fig. 11) are in agreement with the results of the in vitro experiments (Fig. 10; Table 3). Thus, it can be concluded that the in vivo assay could be an alternative to the in vitro assay in cases where the enzymatic activities of the mutated OSCs are too weak for measurement of the Km and kcat and/or the expression levels of the mutated OSCs are significantly low.
There are presently several reports identifying the functions of the active site residues of some OSCs. This is especially true for S. cerevisiae lanosterol and plant cycloartenol synthases, the functions of which which were inferred mainly from the product ratios of each of the triterpenes produced in vivo by the mutants [39-42]. However, no report describing the OSC expression level of the mutants has since been published. We achieved the first complete purification of a β-amyrin synthase. Furthermore, we report that the enzymatic activities of the mutated β-amyrin synthases can be estimated by in vitro and/or in vivo experiments. CD measurements are necessary for probing of the protein architecture. However, the CD spectrum of any OSC has never been reported to date. We successfully obtained the first CD spectrum of a β-amyrin synthase. On the basis of these studies, we propose a method with which to determine the exact function of active site residues. We are now identifying the functions of the active site residues through site-directed mutagenesis experiments, followed by enzyme activity analysis. These results will be reported in the near future.
Instruments and chemicals
NMR spectra were recorded in CDCl3 with a Bruker DMX 600 spectrometer. GC analyses were performed on a Shimadzu GC-2014 chromatograph fitted with a flame ionization detector (J&W DB-1 capillary column, 0.32 mm × 30 m). GC/MS spectra were obtained with a JMS-Q1000 GC K9 (JEOL) instrument under electronic impact at 70 eV, with a Zebron ZB-5 ms capillary column (0.25 mm × 30 m), and the oven temperature being elevated from 220 to 270 °C (3 °C·min−1). CD spectra were measured on a Jasco J-725 with a 0.1-cm quartz cell (Jasco; Tachikawa, Tokyo, Japan). EtAS was dissolved in 50 mm potassium phosphate buffer containing 0.01% Brij35 (0.1 mg·mL−1). Compounds 1 and 5 were synthesized as previously described [29, 32-34]. Compounds 6 and 7 were from Sigma (St Louis, MO, USA) and Merck Calbiochem (Darmstadt, Germany), respectively. β-OG and Tween-80 were from Kanto Kagaku (Tokyo, Japan). Triton X-100 and Brij35 were from Wako Pure Chemical Industries (Osaka, Japan). Restriction enzymes were from Toyobo Co. (Osaka, Japan).
Cloning, expression in yeast, and protein purification
Construction of pYES2–EtAS, expression of EtAS in S. cerevisiae GIL77, and analysis of triterpene products
The pPICZ A plasmid harboring the EtAS gene was kindly provided by K. Ohyama (Ishikawa Prefectural University, Japan). The EtAS gene was amplified by PCR with the following primers: as sense primer, 5′-CACGTTGAAAAGCTTATGTGGAAGCTGAAG-3′ (the HindIII site is underlined); and as antisense primer, 5′-GGCTGGGCGGCCGCTTAAAGAGTAG-3′ (the NotI site is underlined). PCRs were conducted with a 30-cycle program: 98 °C for 1 min, 60 °C for 1 min, 68 °C for 2 min, 68 °C for 2 min, and a final extension at 68 °C for 7 min. KOD-plus DNA polymerase (Toyobo) was used with dNTPs (0.2 mm) in a final volume of 50 μL, according to the manufacturer's protocol. The PCR product was digested with HindIII and NotI (Toyobo). The resulting 2.3-kbp fragment was ligated into the yeast expression vector pYES2 (Life Technologies, Carlsbad, CA, USA), which was previously digested with the same restriction enzyme. The constructed plasmid was named pYES2–EtAS. The integrity of the gene was verified by sequencing. The expression plasmid pYES2–EtAS was transformed into S. cerevisiae GIL77 lacking lanosterol synthase (erg7, ura3-167, hem3-6, gal2), by use of a Frozen-EZ Transformation II Kit (Zymo Research, Irvine, CA, USA). The transformants were plated onto synthetic complete medium without uracil (SC-U), supplemented with ergosterol (20 μg·mL−1), hemin chloride (13 μg·mL−1), and Tween-80 (5 mg·mL−1), and then incubated at 30 °C for 3 days. Expression of EtAS in the yeast was conducted as reported by Kushiro et al. . A 100-mL culture of S. cerevisiae GIL77 harboring pYES2–EtAS was grown in SC-U supplemented with ergosterol (20 μg·mL−1), hemin chloride (13 μg·mL−1), and Tween-80 (5 mg·mL−1) at 30 °C with shaking (150 r.p.m.). After being grown for 2 days, cells were collected and resuspended in SC-U without glucose, supplemented with ergosterol, hemin chloride, and Tween-80, and 2% galactose was then added to induce EtAS expression at 30 °C for 24 h. The yeast cells were collected and resuspended in 0.1 m potassium phosphate (pH 7.0) supplemented with 3% glucose and hemin chloride, and further incubated at 30 °C for 24 h. KOH/MeOH (15%) was added to the grown cells collected, and the suspension was refluxed by heating to disrupt the cells. The lipophilic materials including the cyclization products were extracted with hexane, and the extract was subjected to GC/MS and GC analysis.
Construction and expression of polyhistidine-tagged EtAS
The pYES2–EtAS construct was used as template. The following primers were employed for PCR amplification: 5′-GAATATTAAGCTTGGTACCATGTGGAAGC-3′ (the KpnI site is underlined) as sense primer; and 5′-CTCGAGCCTCGAGTAAAAGAGTAGTGGAAG-3′ (the XbaI site is underlined) as antisense primer, which was designed to delete the stop codon. PCR conditions were the same as described above. The amplified PCR product was digested with KpnI and XbaI (Toyobo). The resulting fragment was ligated into the yeast expression vecor pYES2/CT (Life Technologies), which was previously digested with the same restriction enzymes. The constructed plasmid was named pYES2–EtAS/CT, and was introduced into S. cerevisiae GIL77. His-tagged EtAS (His6) was expressed by culture of the transformed yeast and by induction with galactose, added according to the same protocols described above. The cultured cells collected by centrifugation (10 000 g) were disrupted with glass beads, and purified on an Ni2+–nitrilotriacetic acid column according to the manufacture's protocol (Qiagen, Valencia, CA, USA); the purity was checked by SDS/PAGE on 10% gels. The detailed purification procedure is described in the text; the imidazole concentration in the elution buffer is crucial for obtaining high purity. The protein concentration was measured by the Bradford method with a Bio-Rad protein assay solution with BSA as a standard.
EtAS was dissolved in potassium phosphate buffer (pH 7.0, 50 mm) containing 0.01% Brij35, and the final protein concentration was adjusted to 0.1 mg·mL−1 for CD measurements. The temperatures were varied in the range 5–40 °C. The data were expressed as molar ellipticity.
In vitro enzyme assays
A reaction mixture (2.5 mL, pH 7.0, 100 mm potassium buffer) consisting of Triton X-100 (0.05%, w/v), 1 (120 μg), BSA (1 mg·mL−1), dithiothreitol (1 mm) and purified His-tagged EtAS (2 μg, 8.8 nm) was incubated at 30 °C for 20 min. The reaction was quenched by heating at 100 °C for 3 min. To the reaction mixture, 3 mL of 15% KOH/MeOH, including squalene (25 μg) as an internal standard, was added. The lipophilic materials were extracted with hexane, and the extract was subjected to GC analysis to measure the quantities of β-amyrin produced.
Quantitative western blotting for estimation of in vivo protein expression levels
Purified EtAS or the yeast cell lysates were dissolved or suspended in SDS/PAGE sample buffer (62.5 mm Tris/HCl pH 6.8, 2% SDS, 5% β-mercaptoethanol, 10% glycerol, 0.2 mg·mL−1 bromophenol blue) and boiled for 5 min, and 2 μL per lane was applied to 7.5% polyacrylamide gels. The amounts of purified EtAS (10–80 ng) were used to make a standard curve. The gels were electrophoresed at 40 mA for 60 min, and the proteins then were transferred to a poly(vinylidene difluoride) membrane by semidry electroblotting at 144 mA for 60 min. Blots were blocked with blocking buffer [20 mm Tris/HCl pH 7.4, 150 mm NaCl, 0.05% (v/v) Tween-20, 5% (w/v) skimmed milk] overnight at room temperature, and then washed three times for 10 min each with NaCl/Tris-Tween [20 mm Tris/HCl pH 7.4, 150 mm NaCl, 0.05% (v/v) Tween-20]. Blots were incubated with antibody against V5 (1 : 5000 dilution; Life Technologies) for 1 h at room temperature, and then washed three times in NaCl/Tris-Tween for 10 min each. Blots were then incubated with goat anti-(mouse IgG2a) human ads-fluorescein isothiocyanate (1 : 1250 dilution; Beckman Coulter, Brea, CA, USA) in NaCl/Tris-Tween for 1 h at room temperature, and then washed three times in NaCl/Tris-Tween for 10 min each. Signal detection and imaging were performed with A Luminescent Image Analyzer LAS3000 (Fujifilm, Akasaka, Tokyo, Japan), and protein band densities were analyzed WITH multigauge software (Fujifilm). The details are given in Fig. S9.
Mutagenesis of Asp485, Cys486 and Cys564 in wild-type pYES2–EtAS/CT was performed with the QuickChange site-directed mutagenesis method. The following oligonucleotide primers were used; substitutions are underlined.
For the D485N variant: sense primer, 5′-CATGGTTGGCAAGTTTCTAATTGCACTGCTGAAGG-3′; and antisense primer, 5′-CCTTCAGCAGTGCAATTAGAAACTTGCCAACCATG-3′.
For the D485E variant: sense primer, 5′-GGTTGGCAAGTTTCTGAGTGCACTGCTGAAGG-3′; and antisense primer, 5′-CCTTCAGCAGTGCACTCAGAAACTTGCCAACC-3′.
For the C486A variant: sense primer, 5′-GATCATGGTTGGCAAGTTTCTGATGCCACTGCTGAAGG-3′; and antisense primer, 5′-CCTTCAGCAGTGGCATCAGAAACTTGCCAACCATGATC-3′.
For the C564A variant: sense primer, 5′-GAACATGAATATGTAGAGGCCACAGCATCTGCAATCCATG-3′: and antisense primer, 5′-CATGGATTGCAGATGCTGTGGCCTCTACATATTCATGTTC-3′.
PCRs were conducted with a 16-cycle program: 98 °C for 0.5 min, 60 °C for 1 min, 68 °C for 9 min, and a final extension at 68 °C for 9 min. KOD-plus DNA polymerase (Toyobo) was used with dNTPs (0.2 mm), dimethylsulfoxide (5%) and MgSO4 (0.1 mm) in a final volume of 50 μL. Mutations were confirmed by DNA sequencing with the dideoxy chain-termination method and a Beckman Coulter CEQ8000 (Beckman Coulter).
We are indebted to K. Ohyama (Ishikawa Prefectural University) for the kind gift of the plasmid harboring the EtAS gene, and also to Y. Ebizuka (Tokyo University) and M. Shibuya (Niigata University of Pharmacy and Applied Life Sciences) for the gift of S. cerevisiae GIL77. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (No. 18380001).