Purification and partial sequences of Aralia cordata cinnamyl alcohol dehydrogenase

Phytochemistry, Vol. 32, No. 3, PP.565-567,1993 Printedin GreatBritain.

0031-9422/93$6.00+ 0.00 0 1993Perpmon PressLtd

PURIFICATION AND PARTIAL SEQUENCES OF ARALIA CORDATA CINNAMYL ALCOHOL DEHYDROGENASE TAKASHI HIBINO, DAISUKE SHIBATA,* TOSHIAKI UMEZAWA~ and TAKAYOSHIHIGUCHI$ Mitsui Plant Biotechnology Research Institute, TCI-AlO, Sengen 2-l-6, Tsukuba, Ibaraki 305, Japan; tWood Research Institute, Kyoto University, Gokasho, Uji, Kyoto 611, Japan; SDepartment of Forestry, College of Agriculture and Veterinary Medicine, Nihon University, 3-34-l Shimouma, Setagaya-ku, Tokyo 154, Japan (Received 18 November 1991) IN HONOUR

OF PROFESSOR

MEINHART

ZENK’S SIXTIETH

BIRTHDAY

Key Word Index-Aruliu cordata; Araliaceae; lignin biosynthesis; cinnamyl alcohol dehydrogenase; coniferyl aldehyde; sinapyl aldehyde.

Abatraet-Cinnamyl alcohol dehydrogenase (CAD) (EC 1.1.1.195) from a dicot, Arabia cordata, was purified to homogeneity and its properties were characterized. The enzyme shows a preference for cinnamyl alcohols and cinnamyl aldehydes as substrates. The M, is estimated at 72000. The enzyme is composed of two heterogeneous subunits of slightly different sizes, and it differs from the bean enzyme in the size of subunits. Partial amino acid sequencing of the purified enzyme was carried out both from the N-terminus and using selected peptides obtained by cyanogen bromide cleavage.

INTRODUCIlON

Cinnamyl alcohol dehydrogenase (CAD) catalyses the reduction of cinnamyl aldehydes to corresponding alcohols in the presence of NADPH. The reaction is involved in the last step of the formation of lignin precursors (monolignols) [l]. Purification and characterization of the enzyme from Forsythia suspensa [2], soybean [3,4], spruce [4], Japanese black pine [S] and poplar [6] have been reported. Recently cDNA cloning of a putative bean CAD was carried out by Walter at al. [TJ. However, the same research group later expressed doubts about the identity of that cDNA clone (CAD4) [8]. They found extensive sequence similarity in the protein sequence deduced from the sequence of the CAD4 clone to the NADP+ malic enzyme from maize [9] and based upon this evidence suggested that the CAD4 clone encodes the bean mahc enzyme rather than CAD. Western blotting experiments with bean extracts and an antibody against a poplar CAD showed that the size of the subunit was 64000 [7, lo], which is consistent with the results from the molecular cloning of the enzyme. However, the subunit size differs from those of well characterized CADS from spruce (M, of subunit = 41700) [4], soybean (40 000) [4] and poplar (38 000) [6]. Since the same antibody was uwd to isolate the CAD4 clone from an expression library, it is unclear whether the Western blotting experiments detected the bean malic enzyme or a bean CAD enzyme of novel size. This uncertainty is exacerbated by *Author to whom correspondence should be addressed.

the lack of enzymological characterization and amino acid sequence data for the bean CAD enzyme. The uncertain identity of the CAD4 clone illustrates the need for more information useful for the cloning and confirmation of bonajde CAD genes. Here we report the purification of CAD from young shoots of the perennial, Aralia cordata, and several partial amino acid sequences from the protein. This is the first report of amino acid sequencing of a purified CAD.

RESUL’LS AND

DISCUSSION

Purijication of CAD

We used young shoots of Aralia cordata as a source for purification of CAD because of the high specific activity of the enzyme in the tissue (1160 nkat kg- ‘) and tissue is commercially available in large amounts. Two peaks of CAD activity were isolated using butyl-Toyopearl column chromatography. The major CAD (95% of the total CAD activity applied to the column) was further purified by Mono Q chromatography and affinity chromatography using NADP+-agarose (Table 1). The final preparation showed one band on native PAGE. Substrate specificity

GC-mass spectrometric investigation of the acetylated enzymatic reaction products showed formation of coniferyl alcohol as an enzymatic reduction product of coniferyl aldehyde. Acetate of coniferyl alcohol due to the

565

T. HIBIND et al.

566 No. 1

HL?A??KT?G

No. 2

KETEE

No. 3

ALGAE

No. 4

VGDQKFVVKIPD

No. 5

DYINTAFQ

No. 6

GV ? NQIQ

although gradient gel separation of the subunits of these enzymes has not been reported. The subunits of the bean CAD are much larger than those reported for other CADS (M,=64000) [7].

?I

Amino acid sequencing

Fig. 1. Partial amino acid sequences of Arafia cordata cinnamyl alcohol dehydrogenase. The N-terminal sequence (No. 1) was determined using 25 pg of the purified enzyme. Peptides (No. ~-NO. 6) obtained by cyanogen bromide cleavage of the enzyme were sequenced. Unidentified residues in the sequencing

are indicated by questron marks (?)

enzymatic reduction gave a mass spectrum [m/z (%) 264 (ll.l), 222 (lOO), 180 (27.3), 179 (45.0), 163 (15.0), 162 (23.4), 1.51 (13.6), 147 (16.2). 131 (30.9), 124 (11.3), 119 (19.2), 91 (9.2)] at a retention time [7.1 min], which was identical to that of a chemically synthesized authentic sample [mass spectrum, m/z (%) 264 ([Ml+, 16.6), 222 (lOO), 180 (25.9), 179 (42.7), 163 (13.0), 162 (11.7), 151 (11.3), 147 (11.3), 131 (25.9), 124 (9.6), 119 (13.0), 91 (7.4): retention time, 7.1 min]. Formation of coniferyl alcohol did not occur when CAD was absent. These results confirmed unequivocal formation of coniferyl alcohol from coniferyl aldehyde by the purified CAD. The K, values for reduction of coniferyl aldehyde and sinapyl aldehyde were 1.8 PM and 7.2 PM, respectively. The K, values for oxidation of coniferyl alcohol and sinapyl alcohol were 6.9 PM and 16.6 PM, respectively. The K,s for NADPH and NADP” were 8.1 and 4.8, respectively. The enzyme did not use propylaldehyde, acetaldehyde, NADH or NAD+. Determination

of M,

The M, was estimated to be 72000 by gel filtration. This value is comparable within experimental error to the M, found for the CAD proteins in spruce (72000) [4], soybean (69 000) [4], Japanese black pine (67 000) [S] and poplar (70 000) [6]. SDS-PAGE of the purified enzyme on 7% and 20% gels revealed a band of approximately 39 000 M,, which resolved into two closely spaced bands after 8-25% gradient SDS-PAGE. These results indicate that the Aralia cordata enzyme is composed of two heterogeneous subunits of very close M,. This enzyme differs from the CADS isolated from spruce, poplar and soybean which are reported to be homodimers [4, 61, Table

1. Purification

The purified enzyme was cleaved by cyanogen bromide at the methionyl bonds and subjected to HPLC analysis and selected peptides were used for amino acid sequencing. The identity and position of 43 amino acid residues in the CAD protein were determined as shown in Fig. 1. Amino acid sequence analysis of the entire protein beginning at the N-terminus of the enzyme yielded amino acid identity up to the 10th residue. The identity of four amino acids among the first 10 residues at the N-terminus could not be determined because of low yields of the derivatives of these amino acids in the sequencing cycles. It seems possible that the N-terminal regions of the two subunits are composed of similar but slightly different polypeptide sequences or perhaps one of the subunits is blocked at the amino terminus. We compared the amino acid sequences obtained with the protein sequences predicted by the bean CAD [7] and maize malic enzyme [9] clones. These comparisons showed no significant sequence homology, but the limited numbers of amino acid sequences which we have obtained are not enough to determine with certainty whether the enzyme we have purified shares homology with the bean or maize enzymes. We have used these amino acid sequences, to prepare oligonucleotides as molecular probes or primers which will be used to clone the CAD gene by polymerase chain reaction (PCR). The resulting amino acid sequences will be compared with that of the bean CAD. Molecular cloning of the CAD gene is curkntly underway in our laboratory. EXPERIMENTAL

Plant material. Etiolated young shoots of Aralia cordata were purchased from local markets. Before protein extraction the tissues were placed in the light for 1 day, which increased CAD activity in the tissue about 2fold. Enzyme assay. Coniferyl aldehyde was synthesized as reported in ref. [ll]. Coniferyl alcohol was purchased from Aldrich. Sinapyl alcohol was a gift from Dr S. Kawai. CAD activity was measured at pH 6.5 using

of CAD from 4 kg of young

shoots

of Aralia cordata

Specific (mg)

activity (nkat mg-‘)

(%)

Punfication (fold)

5107 769 2.50 0.24

0.91 6.05 551 2100

100 99 29 11

1 6 605 2307

Protem Purification

method

4&70% (NH&SO, Butyl-Toyopearl Mono Q NADP+-agarose

Yield

Cinnamyl alcohol dehydrogenase of ha&a cordata

65 hM coniferyl aldehyde and 100 PM NADPH as substrates, as described by Wyrambik and Grisebach [3]. Protein was determined by the Bradford method [l2] using a kit (Biolad). Purijcation of CAD. All purification steps were carried out at 4”. Etiolated young shoots of Aralia cordata (4.5 kg) were homogenized with 4 1 of 0.2 M Tris-Cl, 10 mM EDTA (pH 7.5). Protein was pptd from the supernatant of the homogenate with 40%-70% ammonium sulphate. This was desalted and equilibrated with buffer A (10 mM Tris-Cl, 10 mM EDTA, pH 7.5). The pptd protein was dissolved in buffer A contai~ng 0.5 M ammonium sulphate and applied to a butyl-Toyopearl column (5 x 15 cm, Tosoh Co.), which had been equilibrated with 0.5 M ammonium sulphate in buffer A. CAD activity was eluted with a linear gradient of ~monium sulphate (0.5-0.0 M). The protein fraction containing the major peak of CAD activity was pptd with 70% satd ammonium sulphate. After desalting on a Sephadex G-25 column (2.5 x 20 cm), the enzyme fraction was applied to a Mono Q column (Pharmacia). CAD activity was eluted with a linear gradient of NaCl (0 to 0.3 M) in buffer A. After desalting, the enzyme fraction was subjected to affinity CC (1 x 6 cm) using an NADP” agarose matrix (Ph~a~a). The CAD activity was eluted with buffer A containing 1 mM NADP+. The enzyme preparation was desalted using a Centricon filter (Amicon). Identijication of con$eryl alcohol. Coniferyl aldehyde (65 nmol) was incubated with the purified CAD (2 nkat) for 1 hr under the standard assay conditions. The reaction mixt. was extracted with EtOAc (5 ml). The EtOAc soln was dried over anhydrous Na,SO,. Following evapn to dryness, the EtOAc extract was acetylated with A@-pyridine (1: 1, 0.4 ml) at room temp. for 12 hr. Following evapn to dryness, the acetylation product was dissolved in Me,CO (50 4). An aliquot of the soln was then analysed by GC-MS. For a control experiment, coniferyl aldehyde was incubated and anaiysed exactly as above, but without the enzyme preparation. CC-MS operating conditions were as follows: electron impact MS (ionizing voltage, 70 eV), column: Shimazu Hi-cap CBPIWl2-l~(O.53 mm x 12 m) with a guard column (Sup&co Fused silica capillary tubing, 0.53 mm x 1 m), column temperature: 150”, carrier gas: He, 28.5 ml mm’. Estimation of M,. The M, of purified CAD was estimated by gel filtration chromatography with Superose 12

567

(Pharmacia) using M, standards (Pharmacia). PAGE (gradient gels of 8% to 25% with or without SDS) was carried out using the Phast system (Pharmacia). The proteins were visualized using a silver staining kit (Pharmacia). Amino acid sequencing. The purified enzyme (50 pg) was digested using 43 mg of cyanogen bromide in 70% HCO,H at room temp. for 20 hr. After lyophilization, the resulting peptides were sepd by HPLC on a Cl8 reverse phase column (250 x 4.6 mm) with an acetonitrile gradient of &80% in 0.1% TFA under 40 kg cm-‘. Elution of the peptides was monitored at 220 nm. Peptide fractions were lyophilized and dissolved in 20 ~1 of 0.1% TFA. Amino acid sequencing of the peptides was carried out using a gas phase sequencer (ABI). Acknowledgement-We gratefully acknowledge Dr Singo Kawai (Gifu University) for his gift of sinapyl alcohol.

REFERENCES

1. Higuchi, T. (1990) Wood Sci. Technol. 24, 23. 2. Mansell, R. L., Gross, G. G., Stijckigt, J., Franke, H. and Zenk, M. H. (1974) ~hyt~hern~~y 13,2427. 3. Wyrambik, D. and Grisebach, H. (1975) Eur. J. Biochem. 59,9. 4. Liideritz, T. and Grisebach, H. (1981) Eur. J. Biochem. 119, 115. 5. Kutsuki, H., Shimada, M. and Higuchii T. (1982) Phytochemistry 21, 19. 6. Sami, F., Grand, C. and Boudet, A. M. (1984) Eur. J. Biochem. 139, 259. 7. Walter, M. H., Grima-Pettenati, J., Grand, C., Boudet, A. M. and Lamb, C. J. (1988) Proc. Natn. Acad. Sci. 85, 5546. 8. Walter, M. H., Grima-Pettenati,

J., Grand, C., Boudet, A. M. and Lamb, C. J. (1990) Plant Mol. Biol. 15, 525. 9. Rothermel, B. A. and Nelson, T. (1989) J. Biol. Chem. 264, 19587.

10. Grand, C., Sami, F. and Lamb, C. J. (1987) Eur. f. Biochem. 11. Kutsuki, Mokuzai 12. Bradford,

169, 73.

H., Nakatsubo,

F. and Higuchi, T. (1981)

Gakkaishi 27,520. M. M. (1976) Analyt. ~ioche~

72,248.