Planta (1983)159:226-230

P l a n t a 9 Springer-Verlag 1983

Chalcone isomerase in flowers of mutants of Petunia hybrida S. van Weely, A. Bleumer, R. Spruyt and A.W. Schram Section Biosynthesis of Flavonoids, Departments of Genetics and Plant Physiology, Kruislaan 318, NL-1098 SM Amsterdam, The Netherlands

Abstract. The effect of the gene Po on the activity of chalcone isomerase was investigated in Petunia hybrida. Furthermore, isomerase activities isolated from petals were compared with those extracted from anthers. No effect of Po on the pH-dependence of the isomerase and its kinetic properties was observed. With respect to these criteria, the enzyme extracted from anthers behaved in an identical manner to that extracted from petals. Upon chromatofocussing of a petal extract two peaks of activity were present with slightly different isoelectric points (pI 4.8 and pI 5.1). The occurrence of these activities was dependent on the method of enzyme extraction. An isolation procedure using polyvinylpyrrolidone besides Dowex to remove phenolic compounds, followed by ( N H 4 ) 2 S O 4 precipitation of the protein, resulted in only one peak of isomerase at a pI of 5.3. This observation was independent of Po and did not occur in anthers. In anthers one peak of enzyme activity with a pI of 4.5 was present. The moleuclar weight of the isomerase from flowers (62,500 dalton in Po-dominant and Po-recessive plants) differed from the molecular weight of the anther enzyme (44,000 dalton). In Po-recessive mutants the isomerase activity in mature flowers was low compared with Po-dominant mutants, indicating that the mutation in Po either reflects a temporal mutation in the expression of chalcone isomerase or an increased degradation of the enzyme.

Key words: A n t h o c y a n i n - Chalcone isomerase Flavonoic - Petunia.

Introduction

The enzyme chalcone isomerase (CI; EC 5.5.1.6) catalyses the isomerization of chalcones to flavanAbbreviation: CI = Chalcoue isomerase

ones, one of the early steps in flavanoid biosynthesis (Hahlbrock and Grisebach 1975). This enzyme has been the object of several studies (Boland and Wong 1975; Dixon etal. 1982; Forkmann and Dangelmayr 1980; Forkmann and Kuhn 1979; Hahlbrock et al. 1970; Kuhn et al. 1978; Moustafa and Wong 1967; Wiermann 1972). In Petunia hybrida the accumulation of chalcone in pollen is controlled by the gene Po (Wiering 1974; Wiering et al. 1979a, b). In pollen of Po-recessive lines, 2',4',4',6'-tetrahydroxy-chalcone is accumulated, whereas in pollen of Po-dominant lines no chalcone is found (de Vlaming and Kho 1976). Enzymatic investigations in anthers of Petunia hybrida revealed a correlation between accumulation of chalcone and almost total deficiency of CI activity (Forkmann and Kuhn 1979). In petal, however, CI activity in homozygous mutants, recessive for Po, was only 50% of that in the pollen of Podominant lines. This might indicate a different genetic control of CI in anthers and flowers or the existence of tissue-specific isoenzymes of CI. We therefore characterised the chalcone-isomerase activities extracted from petals and anthers of Podominant and Po-recessive mutants. Furthermore, the effect of Po on the expression of CI during flower development was investigated.

Materials and methods Plant material. Genetically defined lines of Petunia hybrida, differing in pollen colour, were cultivated in the greenhouse. Anthocyanin synthesis in pollen occurs in the presence of dominant alleles of the anthocyanin genes Anl, An3 and An4 together with the alleles Hfl or hfl-1 of the hydroxylation gene Hfl (Wiering 1974; Wiering et al. ]979a, b). When anthocyanins are present in the anthers their pollen colour is blue in Podominant mutants or green in homozygous Po recessive mutants, whereas if no anthocyanins are present the pollen is either white or yellow, respectively. The yellow pigment was identified as 2',41,4',6'-tetrahydroxychalcone (de Vlaming and Kho 1976).

S. van Weely et al. : Chalcone isomerase in Petunia hybrida

227

Enzyme preparation. Essentially three methods of enzyme extraction were used. In the first method (further designated as method 1), five petals (wet weight about 250 mg) of flowers of bud length about 40 rnm were homogenized in an UltraTurrax (Janke and Kunkel, Staufer, ERG) in 2 ml 100 mM K phosphate buffer (pH 7.5) in the presence of 1.4 m M fl-mercaptoethanol and 0.25 g Dowex l-X2 (Bio-Rad, Richmond, USA). The homogenate was centrifuged at 38,000 g for 20 rain and the supernatant was used as the enzyme preparation. In the second method (further designated as method 2), several hundred petals of flowers and flower buds were homogenized in a Waring Blendor in 50-/00 ml 10 mM 2-amino-2(hydroxymethyl)-l,3-propanediol (Tris)-HCL buffer (pH 7.5) in the presence of 4 mM fl-mercaptoethanol and 10-20 g Dowex 1-X2. The homogenate was centrifuged at 38,000 g for 30 min. The pellet was resuspended and homogenized again. The combined supernatants were used for further purification by treatment with solid (NH~)2SO 4 to a final concentration of 80% (w/v). After incubation for 20 rain the incubation-mixture was centrifuged at 38,000 g for 20 min. The pellet was resuspended in distilled water and dialysed overnight against 10 mM TrisHCL buffer (pH 7.5) containing 4 mM fl-mercaptoethanol. The dialysed enzyme preparation was loaded onto a diethylaminoethyl (DEAE)-cellulose column (Sigma, St. Louis, Mo., USA; dimensions 10 cm long, 2.8 cm diameter). The column was washed with l0 m M Tris-HCL (pH 7.5) and the activity was eluted with 10 mM Tris-HCL to which 250 mM NaCI was added. The active fractions were pooled, concentrated by Diaflo ultrafiltration (Amicon, PM 10, Danvers, USA) and dialysed overnight against 25 mM imidazole-HCL (pH 7.4). After dialysis, the enzyme preparation was loaded onto a PBE Chromatofocussing column (Pharmacia, Uppsala, Sweden; dimensions 22 cm long, 1.0 cm diameter) and eluted with Polybuffer 74 (Pharmacia; pH 4.0). The individual fractions were measured for pH and CI-activity. The third method (further designated as method 3) was based on the extraction of protein in the presence of both polyvinylpyrrolidine (PVP) and Dowex ]-X2. Several hundred flower buds were homogenized in a Waring Blendor in the presence of 50-100 ml Tris-HCL (50 raM, pH 7.5), 20 m M fl-mercaptoethanol and 5% PVP. After filtration, the homogenate was centrifuged at 38,000 g for 20 rain The supernatant was saturated to 30% (NH4)aSO4 (w/v) and centrifuged for 20 rain at 38,000 g to remove PVP. Solid (NH4)zSO 4 was added to the supernatant to a final saturation of 90% (w/v) and centrifuged. The pellet, containing CI-activity, was suspended in distilled water and dialysed against 25 m M imidazole-HCL (pH 7.4). After dialysis the preparation was loaded onto a PBE column (22 cm long, 1.0 can diameter) and eluted with Polybuffer 74 (pH 4.0). The CI-activity and pH of all fractions were measured.

Enzyme and protein assays. The reaction mixture (1 ml) contained 0.1 M K phosphate (pH 7.5), 10 m M KCN, 10 mM ethylene glycol monomethyl ether, 10-100 lal enzyme, 18.4 mM 2',4',4',6'-tetra hydroxy-chalcone (dissolved in 10 ~tl of 96% ethanol). The reaction was started by addition of chalcone. The blank contained i ml distilled water. Activity was detected by the decrease in absorption at 385 nm. All enzyme-activity data were corrected for non-enzymatic isomerization. Naringenin was identified as the reaction product by high-performance liquid chromatography and spectral analysis (Mabry et al. /970; Schram et al. /983). Potassium cyanide was added to the assay to prevent interference by peroxidase activity (Rathmell and Bendall 1972). The reaction velocity was linear with enzyme concentration up to 100 gg protein. The chalcone was synthesized according to Moustafa and Wong (1967). Pro-

.-,/~l I

I

I

100

-

\

>

50 x

o w

3

5

9 pH

Fig. 1. Relation between pH and activity of CI of the Petunia mutants, W39, Po dominant ( e - - e , limbs; x - - - x , anthers) and R27, Po recessive ( o - - o , petals). Chalcone concentration was 18.4 [aM; pH 3.0-5.5 (sodium acetate); pH 6.0-9.0 (potassium phosphate)

tein was determined by the Bio-Rad Protein assay with bovine gamma globulin as a standard.

Molecular weight determination. Molecular weights were estimated using gel chromatography on Sephacryl S-200 (Pharmacia) and cytochrome c, chymotrypsinogen, bovine serum albumin and blue dextran as markers.

Results

In order to investigate a possible effect of Po on the kinetic properties of CI we extracted (method 1) the enzyme from flower buds of mutants dominant or homozygous recessive for Po. Furthermore, we extracted CI from anthers and compared the effect of pH on the activity of the different enzyme preparations. The results shown in Fig. 1 indicate that the activities of CI in the three extracts behave identically upon variation of the pH. It should be noted that the relatively high spontaneous isomerization exhibits the same dependence on pH as the enzymic isomerization. Each individual measurement was therefore corrected for spontaneous isomerization at that particular pH. The relationship between substrate concentration and reaction velocity for the three activities is shown in Fig. 2. The three curves all indicate substrate inhibition of enzyme activity at substrate concentrations above ]0-20 ~tM and reach maximal activity at 10 ~tM. It is interesting that between a chalcone concentration of 0 and 25 gM the spontaneous isomerization of the chalcone was linear with chalcone concentration (results not shown). No deviation from apparent linearity, which would represent a clear difference in the kinetics of spontaneous and enzymic isomerization of chalcone, was observed at high concentrations of chalcone.

228

S. van Weely et al. : Chalcone isomerase in Petunia hybrida i

9~

0.10

0.010

150 Q "~

O.05 ~

0.005 ';> A~

>

G .< - -

I

0

[

0

0

0.5 S - l ( p M ) -~

Fig. 2. Lineweaver-Burk plot for the activity of CI extracted from the Petunia mutants, W39, Pc dominant ( 0 - - 9 petals; x - - x , anthers) and R27, Pc recessive (o-----o, limbs) at pH 7.5; mU: arbitrary units 4/ "~00

i

i

,

,',

i

io

0

L.5

',~

I

i@

,

,

~o

5.0

55 pH

Fig. 3, Chromatofocussing of protein extracts from limbs and

anthers of the Pc-dominant Petunia mutant W39. 9 9 W39 petals, extraction method 2; 0 - - 0 , W39 petals, extraction method 3 ; x - - x , W39 anthers, extraction method 2

10

i x

x•

30 50 Budlength (mm)

Fig. 4. Activity of CI during bud development in extracts of the Petunia mutants W39, Pc-dominant ( o - - o , petals; x - - x , anthers) and R27, Pc-recessive ( 0 - - 9 petals), mU: arbitrary units

is9 values and the assumed dissociation of enzyme are independent of genotype. The enzyme from anthers (mutant W39) was extracted using method 2. Upon chromatofocussing, only one peak of activity of low is9 point (pH 4.5) was identified. This result might be indicative of a difference in activity between anthers and flower buds. This difference is further confirmed by molecular-weight determinations. The molecular weight of the lower enzyme in both Pc-dominant and Pc-recessive mutants is identical (about 62,500dalton), but differs significantly from the molecular weight of the enzyme extracted from anthers (about 44,000 dalton).

Enzyme activity during development of flower buds. We conclude that the three activities behave identically upon variation of both pH and substrate concentration. Upon chromatofocussing of enzyme extracted from Pc-dominant mutants using method 2 (includes DEAE-cellulose), two peaks of activity were detected at the is9 values of pH 4.8 and pH 5.1 (Fig. 3) indicating the presence of two is 9 of CI. Upon extraction using method 3 (includes DEAE-Cellulose and polyvinylpyrrolidone), only one peak of activity at pH 5.3 was observed (Fig. 3). These results can be explained by assuming a total dissociation of CI in to two active units of lower is 9 point, or by the modification (proteolysis) of enzyme during enzyme extraction. However, no relationship was observed between length of time of isolation and occurrence of the different peaks of activity. Similar results were obtained with CI activity extracted from Pc-recessive mutants, indicating that both the

As shown by Forkmann and Kuhn (1979) and pointed out in the introduction, the effect of Pc being homozygous recessive is most striking in the pollen. In mature pollen of Pc-recessive mutants, CI activity decreased to values not significantly different from those obtained when spontaneous isomerization occurs. In flowers there is residual activity present. Pollen of Pc-recessive mutants accumulate chalcone, but flowers with decreased CI activity synthesize normal amount of anthocyanins. The question arose whether CI activity in recessive genotypes is decreased at every stage of bud development. Anthers and petals from buds of different length were collected from mutants homozygous recessive for Pc (R27) and dominant for Pc (W39).Protein was extracted as described in Materials and methods (method 1). In each individual preparation CI activity was determined (Fig. 4). In young buds, the activity in petals is low and increases upon increasing bud length. In petals of

S. v a n W e e l y et al. : C h a l c o n e i s o m e r a s e i n

Petunia hybrida

229

T a b l e 1. A c t i v i t y o f c h a l c o n e i s o m e r a s e i n a n t h e r s a n d l i m b s o f f l o w e r s o f Specific a c t i v i t y i n n k a t a l m g - x p r o t e i n Mutants

Pollen" colour

Petunia hybrida

at different stages of development.

Specific activity of CI Anthers

Petals

3.8-4.2 b

12.8--20 b

F1 o w e r ~

12.8-20 b

F1 o w e r ~

0.53 0.42 -

3.7 4.9 3.2 3.0 1.1 3.4

0.63 -

12.1 26.9 10 3.8 26.5 15.2

8.9 2.0 3.4 4.9 3.9 1.7

0.20 0.83 -

0.40 0.17 0.05 1 1.1 0.17

0.01 -

7.0 14.2 18.l 5.8 21.4 21.2

0.93 0.35 0.62 1.27 0.58 -

Dominant for Po M43 R8 W39 A9 S6 V10

White White White Blue Blue Blue

Homozygous recessive for Po M73 R27 S1 W37 A26 V28

Yellow Yellow Yellow Yellow Green Green

See M a t e r i a l s a n d m e t h o d s B u d l e n g t h in m m c Separate experiment

mature flowers the activity in the Po-recessive mutant is lowered compared with the activity in flowers of the Po-dominant mutant. This is in agreement with the observations of Forkmann and Kuhn (1979). In anthers from buds of the same length as those used for petal assays, the activity decreases during bud development, which is in agreement with the earlier development of anthers compared with flowers. To obtain more data on the CI activity in the early and late stages of bud development CI activities were investigated in several mutants of Petunia hybrida at different bud stages (Table 1). In petals of young buds, comparable activities in both dominant and recessive mutants were obtained. In mature flowers, however, the activity in Po-recessive mutants was always low compared with the activities extracted from dominant mutants. Obviously, at the developmental stage at which maximal anthocyanin synthesis takes place (bud length 15-35 ram; Gerats et al. 1982), there is no quantitative difference in CI activity. This explains the comparable values for anthocyanin content in flowers of Po-dominant and Po-recessive mutants. In both Po dominant and Po-recessive mutants, CI activity decreases at the mature state of development. A similar pattern of activity is found in anthers fiom dominant and recessive mutants. At later stages of development, the activity in recessive mutants is decreased compared with

dominant mutants. Obviously, at stages of maximal flavonoid synthesis in anthers of Po-recessive mutants, the activity of CI is too low to give maximal synthesis of flavanones, as is observed by the accumulation of the yellow chalcone in Po-recessive mutants. Discussion

The mutation in Po has no qualitative effect on CI in flower limbs. The pH dependence and kinetic properties of CI from Po-dominant and Po-recessive mutants are identical. The behaviour of CI during the different isolation procedures is independent of genotype. Furthermore, the mutation in Po has an identical effect on the expression of CI in limbs and anthers during bud development and does not effect other enzymes of anthocyanin biosynthesis (personal communication by Dr. Gerats; Department of Genetics, University of Amsterdam, The Netherlands). The enzyme present in anthers is probably different from the enzyme in limbs. Therefore, in our opinion, there are two possibile explanations for the regulatory effect of Po on CI activity: either Po recessive is indicative of a temporal mutation leading to a shorter period of effective transcription or translation of the CI loci (and therefore to lower actual values of CI activity in mature flowers and anthers), or Po recessive re-

230 flects a n e n h a n c e d d e g r a d a t i o n o f C I d u r i n g t h e last stages of flower development. The authors are grateful to Mr. R. Vermij and Mr. J.A.J. van der Meijden for making the illustrations and to Mr. J. Bakker and Mr. T.L. Thio for growing the plants.

References Boland, M.J., Wong, E. (1975) Purification and kinetic properties of chalcone-flavanone isomerase from soya bean. Eur. J. Biochem. 50, 383-389 de Vlaming, P., Kho, K.F.F. (1976) 4',2',4',6'-Tetrahydroxychalcone in pollen of Petunia hybrida. Phytochemistry 15, 348-349 Dixon, R.A., Dey, P.M., Whitehead, I.M. (1982) Purification and properties of chalcone isomerase from cell suspension cultures of Phaseolus vulgaris. Biochim. Biophys. Acta 715, 25-33 Forkmann, G., Dangelmayr, B. (1980) Genetic control of chalcone isomerase activity in flowers of Dianthus caryophyllus. Biochem. Genet. 18, 519-527 Forkmann, G., Kuhn, B. (1979) Genetic control of chalcone isomerase activity in anthers of Petunia hybrida. Planta 144, 189-192 Gerats, A.G.M., Cornelissen, P.T.J., Groot, S., Hogervorst, J.M.V., Schram, A.W., Bianchi, F. (1982) A gene controlling rate of anthocyanin synthesis and mutation frequency of the gene Anl in Petunia hybrida. Theor. Appl. Genet. 62, 199-203

S. van Weely et al. : Chalcone isomerase in Petunia hybrida Hahlbrock, K., Grisebach, H. (1975) Biosynthesis of flavonoids. In: The flavonoids, pp. 866-9/5, Haborne, J.B., Mabry, T.J., Mabry, H., eds. Chapman and Hall, London Hahlbrock, K., Wong, E., Schill, L., Grisebach, H. (1970) Comparison of chalcone-flavanone isomerase heteroenzymes and isoenzymes. Phytochemistry 9, 949-958 Kuhn, B., Forkmann, G., Seyffert, W. (1978) Genetic control of chalcone-flavanone isomerase activity in Callistephus chinensis. Planta 138, 199-203 Moustafa, E., Wong, E. (1967) Purification and properties of chalcone-flavanone isomerase from soya bean seed. Phytochemistry 6, 625-632 Rathmell, W.G., Bendall, D.S. (1972) The peroxidase-catalysed oxidation of a chalcone and its possible physiological significance. Biochem. J. 127, 125-132 Schram, A.W., Jonsson, L.M.V., de Vlaming, P. (1983) Identification of anthocyanins and intermediates of anthocyanin biosynthesis from Petunia hybrida using HPLC. Z. Naturforsch. Teil C 38, 342-345 Wiering, H. (1974) Genetics of flower colour in Petunia hybrida Hort. Genen Phaenen 17, 117-134 Wiering, H., de Vlaming, P., Cornu, A., Maizonnier, D. (1979 a) Petunia genetics. I. List of genes. Ann. Amelior. Plant. 29, 611-622 Wiering, H., de Vlaming, P., Cornu, A., Maizonnier, D. (1979b) Petunia genetics. II. A comparison of two gefie banks. Ann. Amelior. Plantes 29, 699-708 Wiermann, R. (1972) Aktivitfit der Chalkon-Flavanon Isomerase und Akkumulation von phenylpropanoiden Verbindungen in Antheren. Planta 102, 55-60 Received 25 February; accepted 19 May 1983