- Email: [email protected]
Cloned and Expressed Fungal phyA Gene in Alfalfa Produces a Stable Phytase
Biochemical and Biophysical Research Communications 290, 1343–1348 (2002) doi:10.1006/bbrc.2002.6361, available online at http://www.idealibrary.com on
Cloned and Expressed Fungal phyA Gene in Alfalfa Produces a Stable Phytase Abul H. J. Ullah,* ,1 Kandan Sethumadhavan,* Edward J. Mullaney,* Thomas Ziegelhoffer,† and Sandra Austin-Phillips† *Southern Regional Research Center, ARS, USDA, New Orleans, Louisiana 70124; and †University of Wisconsin, Madison, Wisconsin 53706
Received December 14, 2001
The phyA gene from Aspergillus ficuum that codes for a 441-amino-acid full-length phosphomonoesterase (phytase) was cloned and expressed in Medicago sativa (alfalfa) leaves. The expressed enzyme from alfalfa leaves was purified to homogeneity and biochemically characterized, and its catalytic properties were elucidated. The expressed phytase in alfalfa leaves retained all the biochemical properties of the benchmark A. ficuum phytase. Although the characteristic bi-hump pH optima were retained in the cloned phytase, the optimal pH shifted downward from 5.5 to 5.0. Also, the recombinant phytase was inhibited by the pseudo-substrate myo-inositol hexasulfate and also by antibody raised against a 20-mer peptide belonging to fungal phytase. The expressed phytase in alfalfa could also be modified by phenylglyoxal. Taken together, the results indicate that fungal phytase when cloned and expressed in alfalfa leaves produces stable and catalytically active phytase while retaining all the properties of the benchmark phytase. This affirms our view that “molecular biofarming” could be an alternative means of producing stable hydrolytic enzymes such as phytase.
Phytic acid, myo-inositol hexakis phosphate, is an abundant plant metabolite that serves as a storage form of phosphorus in a wide variety of legumes and cereals (1). Monogastric animals such as swine and poultry have difficulty digesting phytic acid when they ingest large amounts of soybean meals because they lack phytase (2). As a result, the manure rich in phytic acid may become a potential source of phosphate pollution through the action of microbial phytase. The excess amount of phytic acid in animal feed also poses a health risk to the animal. That is, phytate binds nutritionally important minerals and peptides and 1
To whom correspondence and reprint requests should be addressed.
thereby render the nutrients unavailable to monogastric animals (3, 4). To solve both the nutritional and environmental problems, researchers all over the world had been searching for an active phytase for the last four decades (5, 6). An extensive survey by Shieh and Ware in 1968 identified a strain of Aspergillus ficuum that secreted active phytase (7). This enzyme, myo-inositol hexakisphosphate phospho-hydrolase, is popularly known as 3-phytase, which is an important member of ‘Histidine Acid Phosphatase’ or HAP (8). This phytase is now widely used by the animal feed industry to degrade phytate in the stomach of monogastric animals such as poultry and hogs. The USDA laboratory had purified, characterized, and sequenced this phytase in the 1980s, which paved the way for cloning and expression of the phyA gene in A. niger (9 –12). The cloned A. ficuum phytase was later commercialized and it thus became the benchmark phytase. Its three dimensional structure was also elucidated and a “consensus phytase” was developed by a commercial firm that could not match the activity of the native A. ficuum phytase (13, 14). The fungal phytase, especially the one produced by A. ficuum, is one of the most intensely studied hydrolytic enzymes (7). Because of its high activity and stability several enzyme producers tried to improve other fungal phytases with low activity based on the three dimensional structure of A. ficuum phytase for commercial exploitation (15). The phyA gene coding for the enzyme was not only overexpressed in commercial fungal strains, but also in Nicotiana tabacum (tobacco) plants (12, 16). This paved the way for biofarming of fungal phytase in traditional crops (17). In this communication, we present evidence showing that fungal phytase can be produced by cultivation of the fodder crop Medicago sativa (alfalfa) in which the expressed phyA gene produces stable and fully active phytase in the leaves. This opens up an alternative cost-effective way to produce bulk enzymes in traditional crop plants.
1343
0006-291X/02
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Plasmid Development and Plant Transformation The plasmid and Agrobacterium tumefaciens strain used in this study are identical to those reported in the transformation procedure described earlier (17). Similarly, the transformation protocol used in this study is identical to the one described previously by one of us (18).
Phytase Assay
linear salt gradient (0 to 0.3 M NaCl) was developed in the imidazole buffer at a flow rate of 2.5 ml 10 min. Fractions containing 2.5 ml each column eluate were collected and essayed for both activity and protein. The fraction containing the highest specific activity was used to characterize the cloned and expressed phytase in alfalfa.
Protein Assay The protein concentration of leaf extracts and column fractions was determined by the Coomassie blue G-250 dye binding method using ovalbumin as standard (19).
Phytase assays were performed in 1.0-ml volume at 58°C in 50 mM sodium acetate buffer at pH 5.0 similar to the fungal phytase assay reported earlier (9). The liberated inorganic ortho phosphates were determined spectrophotometrically using a freshly prepared solution of acetone, molybdate, and sulfuric acid whose composition was mentioned earlier (9). A standard curve for inorganic ortho phosphate ranging from 10 to 500 nmol was also made prior to each experiment.
Gel Electrophoresis
Extraction of Phytase
Phytase’s Peptide-Specific Antibody Generation
Mature leaves were harvested from primary transformants and stored at ⫺60°C until fractionation. About 15 g of leaf tissue was homogenized in 120 ml imidazole buffer containing 25 mM imidazole, pH 7.0, supplemented with 0.1 mM CaCl 2, 1.5 mM phenylmethylsulfonyl fluoride (PMSF), and 10% glycerol using a chilled Waring commercial blender at low speed for 1 min. The resulting extract was spun at 12,000 rpm in a Sorval RC-5B centrifuge using a chilled SS-34 rotor. After centrifugation, the supernatant was assayed for both phytase activity and protein concentration.
A synthetic 20-mer peptide mimicking residues 151 through 170 of the fungal phytase (phyA) was conjugated to antigenic protein keyhole limpet hemocyanine (KLH). The primary structure of the peptide is 151IEGFFQSTKLKDPRAQPGQSS 170. A 6.0-mg portion of the 20-mer peptide was conjugated to KLH using carbodiimides as the coupling reagent. The conjugated protein was dialyzed extensively to remove the free peptide. Antibodies were raised in female New Zealand rabbits. The IGGs were finally purified using AVID AL (BioProbe International, Tustin, CA) affinity column chromatography. The purified IGGs eluting from the column were neutralized using Tris solid crystals for long-term stability.
Purification of Phytase Step 1: Macroprep Q anion-exchange chromatography. The supernatants from the leaf extract were dialyzed against 25 mM imidazole buffer, pH 7.0, supplemented with 5% (v/v) glycerol, by changing buffers three times. A 10.0-ml column was made using Kontes Flex-Column (2.5 ⫻ 10 cm) with Macro-Prep High Q (Bio-Rad) as an-ion exchanger. A low pressure Econo System column chromatography workstation (Bio-Rad) was used to perform the chromatography. After loading the supernatant and washing with imidazole buffer, the column was developed running a linear salt gradient of 0 to 0.3 M in imidazole buffer. A flow rate of 3.0 ml per minute was maintained for 30 min and 5.0-ml fractions were collected for both phytase activity and protein measurements. The column was then washed with 20 ml of 0.5 and 1.0 M NaCl in the buffer, respectively. Step 2: Macroprep S cation-exchange chromatography. The active fractions from the previous step were pooled and dialyzed against 50 mM sodium acetate buffer, pH 3.75, supplemented with 5% (v/v) glycerol. A 5.0-ml column was made using Kontes Flex-Column (1.5 ⫻ 5 cm) with Macro-Prep High S (Bio-Rad) as cation exchanger. After loading the dialyzed active fractions from the previous step, the column was washed with acetate buffer and a linear salt gradient (0 to 0.3 M NaCl) was developed in the buffer at a flow rate of 2.0 ml for 20 min. At the end, the column was washed with 0.5 M NaCl in the acetate buffer. Two-milliliter fractions were collected and assayed for both phytase activity and protein. Step 3: Second Macro-Prep Q anion-exchange chromatography. The active fractions from step 2 were pooled and the buffer exchange from sodium acetate to imidazole was accomplished by using an Amicon Stirred Ultrafiltration Cell and PM 10 membrane (Amicon, Beverly, MA). A 5.0-ml Kontes Flex-Column was packed with 2.5 ml anion-exchange resin and was equilibrated in 25 mM imidazole buffer. The protein sample that had gone through ultrafiltration was loaded onto the column and washed with imidazole buffer. Finally, a
Electrophoresis of the purified phytase from alfalfa leaf extract was performed using Xcell II, Mini-Cell and 4 –12% NuPage Bis–Tris gels (Novex, San Diego, CA). Successful separation was achieved by running at a constant 200 V for 70 min. For gel calibration, prestained and multicolored molecular weight markers (4 –250 kDa) were used as standards.
Inhibition of Phytase Activity by myo-Inositol Hexasulfate and Phenylglyoxal We have shown earlier that recombinant fungal phytase expressed in tobacco leaves was inhibited by the pseudo-substrate myo-inositol hexasulfate (MIHS) and the known arginine modifier phenylglyoxal (17). The details of the methods of MIHS and phenylglyoxal inhibition of phytase activity are mentioned in that publication.
RESULTS Overexpression of Fungal Phytase in Alfalfa Leaves Phytase gene (phyA) from A. ficuum was expressed in alfalfa leaves. Table 1 shows the comparison of phytase activities in the leaves of control and transgenic alfalfa leaves. The specific activity data show that phytase activity had increased about 13-fold in the leaves produced by transgenic plant (Table 1). The same leaf extract showed about a 17-fold increase in specific activity for phytase. Purification Purification of overexpressed phyA protein in alfalfa was achieved using both anion and cation exchange columns. Table 2 summarizes the purification of overexpressed phytase. To obtain a maximum purification, it was needed to purify 170-fold starting from the crude
1344
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Comparison of Phytase Activity in the Leaves of Transgenic and Control Alfalfa Parameters
Control plants
Transgenic plants
Amount of leaves used to grind Volume of leaf extract Phytase activity per milliliter Phytase activity per gram of leaves Phytase activity per milligram of protein Total protein in milligrams Total activity Phytase (in milligrams)
3.0 g 68 mL 1.29 nKat 29.24 nKat 1.29 12.85 87.72 0.03
14.75 g 87 mL 66.7 nKat 389.3 nKat 22 261 5742 1.91
Note. Specific activity of 3000 nKat/mg is used for estimation.
leaf extract. The purified phytase had a specific activity of 3761 nKat/mg of protein with an overall yield of 18%. Purity, Molecular Weight, and Glycosylation Figure 1 shows the SDS–PAGE profile of purified phytase from alfalfa leaf. The cloned and overexpressed phytase in alfalfa gave two broad bands in the regions 73 and 100 kDa. The high specific activity (3761 nKat/mg) associated with the purified phytase indicates that a homogenous preparation was obtained through the chromatographic steps employed for the purification. The estimated molecular mass of phytase overexpressed in alfalfa leaf is about 16% lower than the one produced by A. ficuum (9). Similar to the recombinant A. ficuum phytase, the expressed fungal phyA gene in alfalfa leaves also gave a diffused protein band in SDS–PAGE indicating extensive glycosylation. When duplicate gel was visualized for glycosylation through periodic acid Schiff (PAS) staining, it gave positive results (data not presented). Therefore, the cloned phytase in alfalfa leaves also underwent glycosylation similar to fungal phytase (9). Kinetic Characterization pH optima. Although pH optima of the cloned phytase show the characteristic bi-hump peaks similar to native phytase (9), there is a noticeable shift in the optima (Fig. 2). Unlike the native fungal phytase
FIG. 1. SDS–PAGE profile of native A. ficuum phytase and recombinant phytase in alfalfa leaves. Lane A shows molecular weight standards. Lanes B and C show the banding pattern of A. ficuum and expressed phytase in alfalfa leaves, respectively.
whose activity is highest at pH 5–5.5, the cloned and expressed phytase in alfalfa gave the highest activity at pH 5.0; a second peak occurred at pH 3.0 rather than at pH 2.5 for the native fungal phytase. Temperature optima. The temperature optima profile of the overexpressed phytase paralleled the native fungal phytase (Fig. 3). While its activity peaked at 58°C, just as it is for the native phytase, the cloned phytase from alfalfa leaf became more sensitive to thermal denaturation compared to the native phytase. For example, at 68°C the cloned phytase was inactivated completely. Kinetic Parameters The comparison between the native fungal phytase and overexpressed recombinant phytase in alfalfa is shown in Table 3. While the computed turnover number for both the native fungal and overexpressed phytase was found to be very similar, the native phytase gave a lower K m value over the overexpressed phytase produced in alfalfa leaf. The kinetic perfection
TABLE 2
Purification of Expressed Phytase from Alfalfa Leaves Details
Volume (mL)
Total activity (nKats)
Activity (nKat/mL)
Protein (mg/mL)
Specific activity
Leaves extract MacroPrep Q MacroPrep S MacroPrep S
87 100 10 1.5
5742 1607 1073 1038
66.7 16.1 107.3 692.0
3.065 0.193 0.048 0.184
21.8 83.3 2235.4 3760.9
1345
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. The pH versus activity of overexpressed phytase in alfalfa leaves. The buffer used for pH 1.0 –2.5 was 50 mM glycine–HCl; 3.0 –5.0, 50 mM sodium acetate; and 6.0 –9.0, 50 mM imidazole.
of the native phytase was double the value of recombinant phytase because of the higher K m value for phytate in overexpressed phytase (Table 3). When the inhibition constant, K i, was determined for myoinositol hexasulfate (MIHS), a potent inhibitor of phytase, the K i value was about 50-fold lower for the recombinant phytase compared to the value obtained for native phytase. Therefore, the recombinant phytase was more sensitive to MIHS inhibition. Inactivation by Phenylglyoxal Phenylglyoxal, a well-known modifier of arginine inhibits fungal phytases (17, 20). We performed inactivation of the expressed recombinant phytase in alfalfa leaf by the same arginine modifier. The results are shown in Fig. 4. Phenylglyoxal modified arginine quantitatively up to a concentration of 400 nM concentration. At that low concentration, the activity of the overexpressed phytase in alfalfa was inhibited 96%. Also, at about 100 nM inhibitor concentration, 50% of the
FIG. 4. Inhibition of recombinant phytase by phenylglyoxal. Enzyme activity measured as a function of increasing concentration of phenylglyoxal. The X-axis representing the inhibitor concentration in log scale.
activity in cloned phytase was inhibited. These values are similar to those observed previously in the case of native and cloned phytases in tobacco leaves (17). The inactivation of recombinant phytase overexpressed in alfalfa leaf by phenylglyoxal is indicative of a similar active site geometry in both the recombinant and the native phytase. Effect of Phytase-Specific Antibody on Recombinant Phytase The results of antibody binding studies are shown in Fig. 5. The phosphomonoesterase activity in cloned alfalfa phytase was quantitatively inhibited by the purified antibody developed against the internal peptide sequence from phyA protein. The native fungal phytase was also inhibited similarly by the same concentration of antibodies made against the same 20-mer internal peptide of native phytase (data not shown).
TABLE 3
Kinetic Parameters of Native Fungal Phytase and Recombinant Phytase in Alfalfa
Kinetic parameters
FIG. 3. Temperature versus activity of overexpressed phytase in alfalfa leaves. The enzyme assay was performed at pH 5.0.
K m (M) K i for myo-inositol hexaphosphate (M) Turnover number per second K cat/K m (M ⫺1 s ⫺1)
1346
Native fungal phytase
Recombinant and overexpressed phytase
27
50
4.6 348 1.29 ⫾ 10 7
0.08 300 0.6 ⫻ 10 7
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 5. Inhibition of recombinant phytase by native phytasespecific antibody. Phytase activity was measured as a function of increasing concentration of native phytase-specific antibody.
DISCUSSION The phyA gene coding for phytase, an important member of HAP, was successfully introduced and expressed in alfalfa leaves with no apparent loss of activity. We extracted the proteins from alfalfa leaves and then purified the expressed recombinant phytase with a minimal loss of activity. The recombinant phytase from alfalfa leaves was purified using a series of ion-exchange chromatographies. The purified phytase was characterized and its kinetic properties compared with that of fungal phytase. Based on phytase activity, we determined that about a 17-fold increase in enzyme activity resulted from the expression of phytase gene in the leaves of the transgenic alfalfa plant (Table 1). An earlier report from our laboratory had indicated a 5-fold increase in phytase activity in tobacco leaves transformed with fungal phyA gene from A. ficuum (17). Another laboratory had reported a much higher level of expression of this gene in tobacco, but the value was solely based on the amounts of phytase determined by an immunological technique (16). In this communication, we report the level of phytase activity based solely on hydrolysis of phytate. The same phyA gene from A. ficuum when expressed in Escherichia coli was shown to be stored in inclusion bodies and lacked activity. Attempts were made to refold the protein with concomitant regeneration of the activity but without success (21). This could be due to the lack of glycosylation of fungal phytase after expression in E. coli. It remains to be seen what role glycosylation plays in the folding of fungal phytase. Similar to cloned and expressed phyA gene in tobacco leaves, the same phytase expressed in alfalfa also gave a lower mass than the native phytase. The fungal phytase expressed in alfalfa was 16% less glycosylated (Fig. 1) than in A. ficuum; whereas, the phytase expressed in tobacco was 17% less glycosylated (17). Despite this lower level of glycosylation, the recombinant
phytase was fully functional and had the same kinetic parameters as the native fungal phytase (Fig. 2, Fig. 3, and Table 3). It was shown in yeast, Pichia pastoris, that glycosylation affects both secretion and stability of functional phytase (22). Even though the kinetic parameters of the cloned and overexpressed fungal phyA gene product in alfalfa gave nearly identical values to those of the native phytase, the pH optima had shifted one-half pH unit for one of the bi-humps; moving from pH 2.5 to pH 3.0 (Fig. 2). Since both the native and the recombinant phytase in alfalfa had identical primary structure, the lowered glycosylation level in alfalfa may have induced the structural change which may have raised the optimum pH from 2.5 to 3.0. The view that glycosylation plays an important role in the folding of phytase is corroborated by the findings of this study. The importance of glycosylation in folding of phytase to generate the active site was studied when the A. ficuum phytase gene was expressed in E. coli. The expressed fungal phytase in the bacteria was not only devoid of any activities, but it accumulated in the inclusion bodies (21). Although a very small portion of phytase activity was reconstituted, the kinetic properties of the refolded phytase remain unchanged compared to the native phytase. The accumulation of inactive phytase inside the inclusion bodies of the bacteria had clearly showed the role glycosylation plays in folding of the fungal phytase. Alteration of the pH optima of one of the bi-humps may strengthen the view that post-translational modification of phytase such as glycosylation may play a leading role “structure–function” relationship of this industrially important biocatalyst. The temperature optima profile of the overexpressed phytase in alfalfa matched with that of the native and overexpressed phytase in tobacco leaves even though a diminished glycosylation in overexpressed phytases in both alfalfa and tobacco were observed (17). Fungal phytase’s inherent resistance to higher temperature and its ability to hydrolyze phytic acid at 58°C could be attributed to the presence of five disulfide bridges (13, 23). Both the integrity and fidelity of the disulfide bridges in the recombinant phytase should have to be maintained to allow the correct folding of the molecule. The K m and K cat for phytate and K i for myo-inositol hexasulfate in native phytase and recombinant phytase were very similar. These indicate that active site geometry in overexpressed recombinant phytase may be near identical to the native fungal phytase. The kinetic perfection as measured by the ratio of K cat over K m in native fungal phytase was only two-fold higher than it was in recombinant phytase expressed in alfalfa and tobacco (17). This was due to a lower K m value for phytate in the native phytase. All these kinetic data indicate that the active site geometry of the recombinant phytase matched very closely with the native phytase.
1347
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
The fungal phytase belongs to a class of phosphomonoesterase call HAP because in the active site is located a histidine residue (H342) which is implicated in the hydrolysis of the phosphomonoester bond (8, 13). A few arginine residues in the active site were also implicated to play a role in the substrate docking into the active site (13). We probed this substrate-binding site chemically using phenylglyoxal, which is a known arginine modifier. Phenylglyoxal reacted with the sensitive arginine residues inside the active site at a very low concentration of 100 nM (Fig. 4). This value matched the published observations with both the native and overexpressed phytase from tobacco leaves (17). This is indicative of no major alteration of the substrate binding sites in the recombinant and overexpressed phytase in alfalfa leaves. Antibody probing of the activity of phytase, which was done by antibodies developed against a 20-mer internal peptide of native phytase, revealed that the cloned phytase was inhibited quantitatively by the probe (Fig. 5). A similar result was obtained in the native phytase. Therefore, the over all structure of the recombinant phytase matched with that of the native phytase. Based on catalytic characterization of the cloned and overexpressed fungal phyA gene in two plants namely, tobacco and alfalfa, we conclude that molecular biofarming of stable hydrolytic enzymes is a reasonable proposition. Biochemically speaking, there is not much difference between the phytase produced by the fungus and the recombinant phytase produced in either tobacco or alfalfa leaves. Therefore, production of stable hydrolytic enzymes from microbial sources in crop plants could open a new door for commercial exploits. REFERENCES 1. Reddy, N. R., Sathe, S. K., and Salunkhe, D. K. (1982) Adv. Food Res. 28, 1–92. 2. Nelson, T. S., Shieh, T. R., Wodzinski, R. J., and Ware, J. H. (1968) Poult. Sci. 47, 1985–1989. 3. Erdman, J. W., Jr. (1979) J. Am. Oil Chem. Soc. 56, 736 –741.
4. Maga, J. A. (1982) J. Agric. Food Chem. 30, 1–9. 5. Dvorakova, J. (1998) Folia Microbiol. 43, 323–338. 6. Wodzinski, R. J., and Ullah, A. H. J. (1996) Adv. Appl. Microbiol. 42, 263–302. 7. Shieh, T. R., and Ware, J. H. (1968) Appl. Microbiol. 16, 1348 – 1351. 8. Mullaney, E. J., Daly, C. B., and Ullah, A. H. J. (2000) Adv. Appl. Microbiol. 47, 157–199. 9. Ullah, A. H. J., and Gibson, D. M. (1987) Prep. Biochem. 17, 63–91. 10. Ullah, A. H. J., and Dischinger, H. C., Jr. (1993) Biochem. Biophys. Res. Commun. 192, 747–753. 11. Mullaney, E. J., Gibson, D. M., and Ullah, A. H. J. (1991) Appl. Microbiol. Biotechnol. 35, 611– 614. 12. Van Hartingsveldt, W., Vaan Zeijl, C. M. J., Harteveld, G. M., Gouka, R. J., Suykerbuyk, M. E. J., Luiten, R. G. M., Van Paridon, P. A., Selten, G. C. M., Veenstra, A. E., Van Gorcom, R. F. P., and Van den Hondel, C. A. M. J. J. (1993) Gene 127, 87–94. 13. Kostrewa, D. F., Gruninger-Leitch, F., D’Arcy, A., Broger, C., Mitchell, D., and van Loon, A. P. G. M. (1997) Nat. Struct. Biol. 4, 185–190. 14. Lehmann, M., Pasamontes, L., Lassen, S. F., and Wyss, M. (2000) Biochim. Biophys. Acta 1543, 408 – 415. 15. Lehmann, M., Lopez-Ulibarri, R., Loch, C., Viarouge, C., Wyss, M., and van Loon, A. P. (2000) Protein Sci. 9, 1866 –1872. 16. Verwoerd, T. C., van Paridon, P. A., van Ooyen, A. J. J., van Lent, J. W. M., Hoekma, A., and Pen, J. (1995) Plant Physiol. 109, 1199 –1205. 17. Ullah, A. H. J., Sethumadhavan, K., Mullaney, E. J., Ziegelhoffer, T., and Austin-Phillips, S. (1999) Biochem. Biophys. Res. Commun. 264, 201–206. 18. Austin, S., Bingham, E. T., Mathews, D. E., Shahan, M. N., Will, J., and Burgess, R. (1995) Euphytica 4, 1–13. 19. Sedmark, J. J., and Grossberg, S. E. (1977) Anal. Biochem. 79, 544 –552. 20. Ullah, A. H. J., Sethumadhavan, K., Lei, X. G., and Mullaney, E. J. (2000) Biochem. Biophys. Res. Commun. 275, 279 –285. 21. Phillippy, B. Q., and Mullaney, E. J. (1997) J. Agric. Food Chem. 243, 458 – 462. 22. Han, Y., and Lei, X. G. (1999) Arch. Biochem. Biophys. 364, 83–90. 23. Ullah, A. H. J., and Mullaney, E. J. (1996) Biochem. Biophys. Res. Commun. 227, 311–317.
1348
Cloned and Expressed Fungal phyA Gene in Alfalfa Produces a Stable Phytase Abul H. J. Ullah,* ,1 Kandan Sethumadhavan,* Edward J. Mullaney,* Thomas Ziegelhoffer,† and Sandra Austin-Phillips† *Southern Regional Research Center, ARS, USDA, New Orleans, Louisiana 70124; and †University of Wisconsin, Madison, Wisconsin 53706
Received December 14, 2001
The phyA gene from Aspergillus ficuum that codes for a 441-amino-acid full-length phosphomonoesterase (phytase) was cloned and expressed in Medicago sativa (alfalfa) leaves. The expressed enzyme from alfalfa leaves was purified to homogeneity and biochemically characterized, and its catalytic properties were elucidated. The expressed phytase in alfalfa leaves retained all the biochemical properties of the benchmark A. ficuum phytase. Although the characteristic bi-hump pH optima were retained in the cloned phytase, the optimal pH shifted downward from 5.5 to 5.0. Also, the recombinant phytase was inhibited by the pseudo-substrate myo-inositol hexasulfate and also by antibody raised against a 20-mer peptide belonging to fungal phytase. The expressed phytase in alfalfa could also be modified by phenylglyoxal. Taken together, the results indicate that fungal phytase when cloned and expressed in alfalfa leaves produces stable and catalytically active phytase while retaining all the properties of the benchmark phytase. This affirms our view that “molecular biofarming” could be an alternative means of producing stable hydrolytic enzymes such as phytase.
Phytic acid, myo-inositol hexakis phosphate, is an abundant plant metabolite that serves as a storage form of phosphorus in a wide variety of legumes and cereals (1). Monogastric animals such as swine and poultry have difficulty digesting phytic acid when they ingest large amounts of soybean meals because they lack phytase (2). As a result, the manure rich in phytic acid may become a potential source of phosphate pollution through the action of microbial phytase. The excess amount of phytic acid in animal feed also poses a health risk to the animal. That is, phytate binds nutritionally important minerals and peptides and 1
To whom correspondence and reprint requests should be addressed.
thereby render the nutrients unavailable to monogastric animals (3, 4). To solve both the nutritional and environmental problems, researchers all over the world had been searching for an active phytase for the last four decades (5, 6). An extensive survey by Shieh and Ware in 1968 identified a strain of Aspergillus ficuum that secreted active phytase (7). This enzyme, myo-inositol hexakisphosphate phospho-hydrolase, is popularly known as 3-phytase, which is an important member of ‘Histidine Acid Phosphatase’ or HAP (8). This phytase is now widely used by the animal feed industry to degrade phytate in the stomach of monogastric animals such as poultry and hogs. The USDA laboratory had purified, characterized, and sequenced this phytase in the 1980s, which paved the way for cloning and expression of the phyA gene in A. niger (9 –12). The cloned A. ficuum phytase was later commercialized and it thus became the benchmark phytase. Its three dimensional structure was also elucidated and a “consensus phytase” was developed by a commercial firm that could not match the activity of the native A. ficuum phytase (13, 14). The fungal phytase, especially the one produced by A. ficuum, is one of the most intensely studied hydrolytic enzymes (7). Because of its high activity and stability several enzyme producers tried to improve other fungal phytases with low activity based on the three dimensional structure of A. ficuum phytase for commercial exploitation (15). The phyA gene coding for the enzyme was not only overexpressed in commercial fungal strains, but also in Nicotiana tabacum (tobacco) plants (12, 16). This paved the way for biofarming of fungal phytase in traditional crops (17). In this communication, we present evidence showing that fungal phytase can be produced by cultivation of the fodder crop Medicago sativa (alfalfa) in which the expressed phyA gene produces stable and fully active phytase in the leaves. This opens up an alternative cost-effective way to produce bulk enzymes in traditional crop plants.
1343
0006-291X/02
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Plasmid Development and Plant Transformation The plasmid and Agrobacterium tumefaciens strain used in this study are identical to those reported in the transformation procedure described earlier (17). Similarly, the transformation protocol used in this study is identical to the one described previously by one of us (18).
Phytase Assay
linear salt gradient (0 to 0.3 M NaCl) was developed in the imidazole buffer at a flow rate of 2.5 ml 10 min. Fractions containing 2.5 ml each column eluate were collected and essayed for both activity and protein. The fraction containing the highest specific activity was used to characterize the cloned and expressed phytase in alfalfa.
Protein Assay The protein concentration of leaf extracts and column fractions was determined by the Coomassie blue G-250 dye binding method using ovalbumin as standard (19).
Phytase assays were performed in 1.0-ml volume at 58°C in 50 mM sodium acetate buffer at pH 5.0 similar to the fungal phytase assay reported earlier (9). The liberated inorganic ortho phosphates were determined spectrophotometrically using a freshly prepared solution of acetone, molybdate, and sulfuric acid whose composition was mentioned earlier (9). A standard curve for inorganic ortho phosphate ranging from 10 to 500 nmol was also made prior to each experiment.
Gel Electrophoresis
Extraction of Phytase
Phytase’s Peptide-Specific Antibody Generation
Mature leaves were harvested from primary transformants and stored at ⫺60°C until fractionation. About 15 g of leaf tissue was homogenized in 120 ml imidazole buffer containing 25 mM imidazole, pH 7.0, supplemented with 0.1 mM CaCl 2, 1.5 mM phenylmethylsulfonyl fluoride (PMSF), and 10% glycerol using a chilled Waring commercial blender at low speed for 1 min. The resulting extract was spun at 12,000 rpm in a Sorval RC-5B centrifuge using a chilled SS-34 rotor. After centrifugation, the supernatant was assayed for both phytase activity and protein concentration.
A synthetic 20-mer peptide mimicking residues 151 through 170 of the fungal phytase (phyA) was conjugated to antigenic protein keyhole limpet hemocyanine (KLH). The primary structure of the peptide is 151IEGFFQSTKLKDPRAQPGQSS 170. A 6.0-mg portion of the 20-mer peptide was conjugated to KLH using carbodiimides as the coupling reagent. The conjugated protein was dialyzed extensively to remove the free peptide. Antibodies were raised in female New Zealand rabbits. The IGGs were finally purified using AVID AL (BioProbe International, Tustin, CA) affinity column chromatography. The purified IGGs eluting from the column were neutralized using Tris solid crystals for long-term stability.
Purification of Phytase Step 1: Macroprep Q anion-exchange chromatography. The supernatants from the leaf extract were dialyzed against 25 mM imidazole buffer, pH 7.0, supplemented with 5% (v/v) glycerol, by changing buffers three times. A 10.0-ml column was made using Kontes Flex-Column (2.5 ⫻ 10 cm) with Macro-Prep High Q (Bio-Rad) as an-ion exchanger. A low pressure Econo System column chromatography workstation (Bio-Rad) was used to perform the chromatography. After loading the supernatant and washing with imidazole buffer, the column was developed running a linear salt gradient of 0 to 0.3 M in imidazole buffer. A flow rate of 3.0 ml per minute was maintained for 30 min and 5.0-ml fractions were collected for both phytase activity and protein measurements. The column was then washed with 20 ml of 0.5 and 1.0 M NaCl in the buffer, respectively. Step 2: Macroprep S cation-exchange chromatography. The active fractions from the previous step were pooled and dialyzed against 50 mM sodium acetate buffer, pH 3.75, supplemented with 5% (v/v) glycerol. A 5.0-ml column was made using Kontes Flex-Column (1.5 ⫻ 5 cm) with Macro-Prep High S (Bio-Rad) as cation exchanger. After loading the dialyzed active fractions from the previous step, the column was washed with acetate buffer and a linear salt gradient (0 to 0.3 M NaCl) was developed in the buffer at a flow rate of 2.0 ml for 20 min. At the end, the column was washed with 0.5 M NaCl in the acetate buffer. Two-milliliter fractions were collected and assayed for both phytase activity and protein. Step 3: Second Macro-Prep Q anion-exchange chromatography. The active fractions from step 2 were pooled and the buffer exchange from sodium acetate to imidazole was accomplished by using an Amicon Stirred Ultrafiltration Cell and PM 10 membrane (Amicon, Beverly, MA). A 5.0-ml Kontes Flex-Column was packed with 2.5 ml anion-exchange resin and was equilibrated in 25 mM imidazole buffer. The protein sample that had gone through ultrafiltration was loaded onto the column and washed with imidazole buffer. Finally, a
Electrophoresis of the purified phytase from alfalfa leaf extract was performed using Xcell II, Mini-Cell and 4 –12% NuPage Bis–Tris gels (Novex, San Diego, CA). Successful separation was achieved by running at a constant 200 V for 70 min. For gel calibration, prestained and multicolored molecular weight markers (4 –250 kDa) were used as standards.
Inhibition of Phytase Activity by myo-Inositol Hexasulfate and Phenylglyoxal We have shown earlier that recombinant fungal phytase expressed in tobacco leaves was inhibited by the pseudo-substrate myo-inositol hexasulfate (MIHS) and the known arginine modifier phenylglyoxal (17). The details of the methods of MIHS and phenylglyoxal inhibition of phytase activity are mentioned in that publication.
RESULTS Overexpression of Fungal Phytase in Alfalfa Leaves Phytase gene (phyA) from A. ficuum was expressed in alfalfa leaves. Table 1 shows the comparison of phytase activities in the leaves of control and transgenic alfalfa leaves. The specific activity data show that phytase activity had increased about 13-fold in the leaves produced by transgenic plant (Table 1). The same leaf extract showed about a 17-fold increase in specific activity for phytase. Purification Purification of overexpressed phyA protein in alfalfa was achieved using both anion and cation exchange columns. Table 2 summarizes the purification of overexpressed phytase. To obtain a maximum purification, it was needed to purify 170-fold starting from the crude
1344
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1
Comparison of Phytase Activity in the Leaves of Transgenic and Control Alfalfa Parameters
Control plants
Transgenic plants
Amount of leaves used to grind Volume of leaf extract Phytase activity per milliliter Phytase activity per gram of leaves Phytase activity per milligram of protein Total protein in milligrams Total activity Phytase (in milligrams)
3.0 g 68 mL 1.29 nKat 29.24 nKat 1.29 12.85 87.72 0.03
14.75 g 87 mL 66.7 nKat 389.3 nKat 22 261 5742 1.91
Note. Specific activity of 3000 nKat/mg is used for estimation.
leaf extract. The purified phytase had a specific activity of 3761 nKat/mg of protein with an overall yield of 18%. Purity, Molecular Weight, and Glycosylation Figure 1 shows the SDS–PAGE profile of purified phytase from alfalfa leaf. The cloned and overexpressed phytase in alfalfa gave two broad bands in the regions 73 and 100 kDa. The high specific activity (3761 nKat/mg) associated with the purified phytase indicates that a homogenous preparation was obtained through the chromatographic steps employed for the purification. The estimated molecular mass of phytase overexpressed in alfalfa leaf is about 16% lower than the one produced by A. ficuum (9). Similar to the recombinant A. ficuum phytase, the expressed fungal phyA gene in alfalfa leaves also gave a diffused protein band in SDS–PAGE indicating extensive glycosylation. When duplicate gel was visualized for glycosylation through periodic acid Schiff (PAS) staining, it gave positive results (data not presented). Therefore, the cloned phytase in alfalfa leaves also underwent glycosylation similar to fungal phytase (9). Kinetic Characterization pH optima. Although pH optima of the cloned phytase show the characteristic bi-hump peaks similar to native phytase (9), there is a noticeable shift in the optima (Fig. 2). Unlike the native fungal phytase
FIG. 1. SDS–PAGE profile of native A. ficuum phytase and recombinant phytase in alfalfa leaves. Lane A shows molecular weight standards. Lanes B and C show the banding pattern of A. ficuum and expressed phytase in alfalfa leaves, respectively.
whose activity is highest at pH 5–5.5, the cloned and expressed phytase in alfalfa gave the highest activity at pH 5.0; a second peak occurred at pH 3.0 rather than at pH 2.5 for the native fungal phytase. Temperature optima. The temperature optima profile of the overexpressed phytase paralleled the native fungal phytase (Fig. 3). While its activity peaked at 58°C, just as it is for the native phytase, the cloned phytase from alfalfa leaf became more sensitive to thermal denaturation compared to the native phytase. For example, at 68°C the cloned phytase was inactivated completely. Kinetic Parameters The comparison between the native fungal phytase and overexpressed recombinant phytase in alfalfa is shown in Table 3. While the computed turnover number for both the native fungal and overexpressed phytase was found to be very similar, the native phytase gave a lower K m value over the overexpressed phytase produced in alfalfa leaf. The kinetic perfection
TABLE 2
Purification of Expressed Phytase from Alfalfa Leaves Details
Volume (mL)
Total activity (nKats)
Activity (nKat/mL)
Protein (mg/mL)
Specific activity
Leaves extract MacroPrep Q MacroPrep S MacroPrep S
87 100 10 1.5
5742 1607 1073 1038
66.7 16.1 107.3 692.0
3.065 0.193 0.048 0.184
21.8 83.3 2235.4 3760.9
1345
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. The pH versus activity of overexpressed phytase in alfalfa leaves. The buffer used for pH 1.0 –2.5 was 50 mM glycine–HCl; 3.0 –5.0, 50 mM sodium acetate; and 6.0 –9.0, 50 mM imidazole.
of the native phytase was double the value of recombinant phytase because of the higher K m value for phytate in overexpressed phytase (Table 3). When the inhibition constant, K i, was determined for myoinositol hexasulfate (MIHS), a potent inhibitor of phytase, the K i value was about 50-fold lower for the recombinant phytase compared to the value obtained for native phytase. Therefore, the recombinant phytase was more sensitive to MIHS inhibition. Inactivation by Phenylglyoxal Phenylglyoxal, a well-known modifier of arginine inhibits fungal phytases (17, 20). We performed inactivation of the expressed recombinant phytase in alfalfa leaf by the same arginine modifier. The results are shown in Fig. 4. Phenylglyoxal modified arginine quantitatively up to a concentration of 400 nM concentration. At that low concentration, the activity of the overexpressed phytase in alfalfa was inhibited 96%. Also, at about 100 nM inhibitor concentration, 50% of the
FIG. 4. Inhibition of recombinant phytase by phenylglyoxal. Enzyme activity measured as a function of increasing concentration of phenylglyoxal. The X-axis representing the inhibitor concentration in log scale.
activity in cloned phytase was inhibited. These values are similar to those observed previously in the case of native and cloned phytases in tobacco leaves (17). The inactivation of recombinant phytase overexpressed in alfalfa leaf by phenylglyoxal is indicative of a similar active site geometry in both the recombinant and the native phytase. Effect of Phytase-Specific Antibody on Recombinant Phytase The results of antibody binding studies are shown in Fig. 5. The phosphomonoesterase activity in cloned alfalfa phytase was quantitatively inhibited by the purified antibody developed against the internal peptide sequence from phyA protein. The native fungal phytase was also inhibited similarly by the same concentration of antibodies made against the same 20-mer internal peptide of native phytase (data not shown).
TABLE 3
Kinetic Parameters of Native Fungal Phytase and Recombinant Phytase in Alfalfa
Kinetic parameters
FIG. 3. Temperature versus activity of overexpressed phytase in alfalfa leaves. The enzyme assay was performed at pH 5.0.
K m (M) K i for myo-inositol hexaphosphate (M) Turnover number per second K cat/K m (M ⫺1 s ⫺1)
1346
Native fungal phytase
Recombinant and overexpressed phytase
27
50
4.6 348 1.29 ⫾ 10 7
0.08 300 0.6 ⫻ 10 7
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 5. Inhibition of recombinant phytase by native phytasespecific antibody. Phytase activity was measured as a function of increasing concentration of native phytase-specific antibody.
DISCUSSION The phyA gene coding for phytase, an important member of HAP, was successfully introduced and expressed in alfalfa leaves with no apparent loss of activity. We extracted the proteins from alfalfa leaves and then purified the expressed recombinant phytase with a minimal loss of activity. The recombinant phytase from alfalfa leaves was purified using a series of ion-exchange chromatographies. The purified phytase was characterized and its kinetic properties compared with that of fungal phytase. Based on phytase activity, we determined that about a 17-fold increase in enzyme activity resulted from the expression of phytase gene in the leaves of the transgenic alfalfa plant (Table 1). An earlier report from our laboratory had indicated a 5-fold increase in phytase activity in tobacco leaves transformed with fungal phyA gene from A. ficuum (17). Another laboratory had reported a much higher level of expression of this gene in tobacco, but the value was solely based on the amounts of phytase determined by an immunological technique (16). In this communication, we report the level of phytase activity based solely on hydrolysis of phytate. The same phyA gene from A. ficuum when expressed in Escherichia coli was shown to be stored in inclusion bodies and lacked activity. Attempts were made to refold the protein with concomitant regeneration of the activity but without success (21). This could be due to the lack of glycosylation of fungal phytase after expression in E. coli. It remains to be seen what role glycosylation plays in the folding of fungal phytase. Similar to cloned and expressed phyA gene in tobacco leaves, the same phytase expressed in alfalfa also gave a lower mass than the native phytase. The fungal phytase expressed in alfalfa was 16% less glycosylated (Fig. 1) than in A. ficuum; whereas, the phytase expressed in tobacco was 17% less glycosylated (17). Despite this lower level of glycosylation, the recombinant
phytase was fully functional and had the same kinetic parameters as the native fungal phytase (Fig. 2, Fig. 3, and Table 3). It was shown in yeast, Pichia pastoris, that glycosylation affects both secretion and stability of functional phytase (22). Even though the kinetic parameters of the cloned and overexpressed fungal phyA gene product in alfalfa gave nearly identical values to those of the native phytase, the pH optima had shifted one-half pH unit for one of the bi-humps; moving from pH 2.5 to pH 3.0 (Fig. 2). Since both the native and the recombinant phytase in alfalfa had identical primary structure, the lowered glycosylation level in alfalfa may have induced the structural change which may have raised the optimum pH from 2.5 to 3.0. The view that glycosylation plays an important role in the folding of phytase is corroborated by the findings of this study. The importance of glycosylation in folding of phytase to generate the active site was studied when the A. ficuum phytase gene was expressed in E. coli. The expressed fungal phytase in the bacteria was not only devoid of any activities, but it accumulated in the inclusion bodies (21). Although a very small portion of phytase activity was reconstituted, the kinetic properties of the refolded phytase remain unchanged compared to the native phytase. The accumulation of inactive phytase inside the inclusion bodies of the bacteria had clearly showed the role glycosylation plays in folding of the fungal phytase. Alteration of the pH optima of one of the bi-humps may strengthen the view that post-translational modification of phytase such as glycosylation may play a leading role “structure–function” relationship of this industrially important biocatalyst. The temperature optima profile of the overexpressed phytase in alfalfa matched with that of the native and overexpressed phytase in tobacco leaves even though a diminished glycosylation in overexpressed phytases in both alfalfa and tobacco were observed (17). Fungal phytase’s inherent resistance to higher temperature and its ability to hydrolyze phytic acid at 58°C could be attributed to the presence of five disulfide bridges (13, 23). Both the integrity and fidelity of the disulfide bridges in the recombinant phytase should have to be maintained to allow the correct folding of the molecule. The K m and K cat for phytate and K i for myo-inositol hexasulfate in native phytase and recombinant phytase were very similar. These indicate that active site geometry in overexpressed recombinant phytase may be near identical to the native fungal phytase. The kinetic perfection as measured by the ratio of K cat over K m in native fungal phytase was only two-fold higher than it was in recombinant phytase expressed in alfalfa and tobacco (17). This was due to a lower K m value for phytate in the native phytase. All these kinetic data indicate that the active site geometry of the recombinant phytase matched very closely with the native phytase.
1347
Vol. 290, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
The fungal phytase belongs to a class of phosphomonoesterase call HAP because in the active site is located a histidine residue (H342) which is implicated in the hydrolysis of the phosphomonoester bond (8, 13). A few arginine residues in the active site were also implicated to play a role in the substrate docking into the active site (13). We probed this substrate-binding site chemically using phenylglyoxal, which is a known arginine modifier. Phenylglyoxal reacted with the sensitive arginine residues inside the active site at a very low concentration of 100 nM (Fig. 4). This value matched the published observations with both the native and overexpressed phytase from tobacco leaves (17). This is indicative of no major alteration of the substrate binding sites in the recombinant and overexpressed phytase in alfalfa leaves. Antibody probing of the activity of phytase, which was done by antibodies developed against a 20-mer internal peptide of native phytase, revealed that the cloned phytase was inhibited quantitatively by the probe (Fig. 5). A similar result was obtained in the native phytase. Therefore, the over all structure of the recombinant phytase matched with that of the native phytase. Based on catalytic characterization of the cloned and overexpressed fungal phyA gene in two plants namely, tobacco and alfalfa, we conclude that molecular biofarming of stable hydrolytic enzymes is a reasonable proposition. Biochemically speaking, there is not much difference between the phytase produced by the fungus and the recombinant phytase produced in either tobacco or alfalfa leaves. Therefore, production of stable hydrolytic enzymes from microbial sources in crop plants could open a new door for commercial exploits. REFERENCES 1. Reddy, N. R., Sathe, S. K., and Salunkhe, D. K. (1982) Adv. Food Res. 28, 1–92. 2. Nelson, T. S., Shieh, T. R., Wodzinski, R. J., and Ware, J. H. (1968) Poult. Sci. 47, 1985–1989. 3. Erdman, J. W., Jr. (1979) J. Am. Oil Chem. Soc. 56, 736 –741.
4. Maga, J. A. (1982) J. Agric. Food Chem. 30, 1–9. 5. Dvorakova, J. (1998) Folia Microbiol. 43, 323–338. 6. Wodzinski, R. J., and Ullah, A. H. J. (1996) Adv. Appl. Microbiol. 42, 263–302. 7. Shieh, T. R., and Ware, J. H. (1968) Appl. Microbiol. 16, 1348 – 1351. 8. Mullaney, E. J., Daly, C. B., and Ullah, A. H. J. (2000) Adv. Appl. Microbiol. 47, 157–199. 9. Ullah, A. H. J., and Gibson, D. M. (1987) Prep. Biochem. 17, 63–91. 10. Ullah, A. H. J., and Dischinger, H. C., Jr. (1993) Biochem. Biophys. Res. Commun. 192, 747–753. 11. Mullaney, E. J., Gibson, D. M., and Ullah, A. H. J. (1991) Appl. Microbiol. Biotechnol. 35, 611– 614. 12. Van Hartingsveldt, W., Vaan Zeijl, C. M. J., Harteveld, G. M., Gouka, R. J., Suykerbuyk, M. E. J., Luiten, R. G. M., Van Paridon, P. A., Selten, G. C. M., Veenstra, A. E., Van Gorcom, R. F. P., and Van den Hondel, C. A. M. J. J. (1993) Gene 127, 87–94. 13. Kostrewa, D. F., Gruninger-Leitch, F., D’Arcy, A., Broger, C., Mitchell, D., and van Loon, A. P. G. M. (1997) Nat. Struct. Biol. 4, 185–190. 14. Lehmann, M., Pasamontes, L., Lassen, S. F., and Wyss, M. (2000) Biochim. Biophys. Acta 1543, 408 – 415. 15. Lehmann, M., Lopez-Ulibarri, R., Loch, C., Viarouge, C., Wyss, M., and van Loon, A. P. (2000) Protein Sci. 9, 1866 –1872. 16. Verwoerd, T. C., van Paridon, P. A., van Ooyen, A. J. J., van Lent, J. W. M., Hoekma, A., and Pen, J. (1995) Plant Physiol. 109, 1199 –1205. 17. Ullah, A. H. J., Sethumadhavan, K., Mullaney, E. J., Ziegelhoffer, T., and Austin-Phillips, S. (1999) Biochem. Biophys. Res. Commun. 264, 201–206. 18. Austin, S., Bingham, E. T., Mathews, D. E., Shahan, M. N., Will, J., and Burgess, R. (1995) Euphytica 4, 1–13. 19. Sedmark, J. J., and Grossberg, S. E. (1977) Anal. Biochem. 79, 544 –552. 20. Ullah, A. H. J., Sethumadhavan, K., Lei, X. G., and Mullaney, E. J. (2000) Biochem. Biophys. Res. Commun. 275, 279 –285. 21. Phillippy, B. Q., and Mullaney, E. J. (1997) J. Agric. Food Chem. 243, 458 – 462. 22. Han, Y., and Lei, X. G. (1999) Arch. Biochem. Biophys. 364, 83–90. 23. Ullah, A. H. J., and Mullaney, E. J. (1996) Biochem. Biophys. Res. Commun. 227, 311–317.
1348