Plant a

Planta (1984)160:455-463

9 Springer-Verlag 1984

lnternode length in Pisum The Le gene controls the 3p-hydroxylation of gibbereilin A2o to gibberellin A t

Timothy J. Ingtam 1.2, James B. Reid 1, Ian C. Murfet ~, Paul Gaskin 2 Christine L. Willis 2 and Jake MacMillan 2 a Department of B~tany, University of Tasmania, Hobart, Tasmania, Australia 7001, and 2 School of Chemigtry, University of Bristol, Bristol BS8 1TS, UK

Abstract. The !nfluence of the Na and Le genes in peas on gib.berellin (GA) levels and metabolism were examined iby gas chromatographic-mass spectrometric analylsis of extracts from a range of stemlength genotypes fed with [13C, 3H]GA20. The substrate was metabolised to [t3C, 3H]GA~, [13C, all]GAs and [!3C, 3H]GA29 in the immature, expanding apical tissue of all genotypes carrying Le. In contrast, [tac, 3 H ] G A 2 9 and, in one line, [t3C, 3H]GAz9-catabolite, were the only products detected in plants homozygous for the le gene. These results confirm that the Le gene in peas controls the 3fl-hydroxylation of GA20 to GAp Qualitatively the same results were obtained irrespective of the genotype at the Na locus. In all Na lines the [13C, 3H]GA20 metabolites were considerably diluted by endogenous [12C]GAs, implying that the metabolism of [t3C, 3H]GAz0 mirrored that of endogenous [12C]GA2o. In contrast, the [t3C, 3H]GA2o metabolites in na lines showed no dilution with [12C]GAs, confirming that the na mutation prevents the production of C19-GAs. Estimates of the levels of endogenous GAs in the apical tissues of Na lines, made from the 12C: 13C isotope ratios and the iradioactivity recovered in respective metabolites, varied between 7 and 40 ng of each GA per plant: in the tissue expanded during the 5 d between treatment with [13C, 3H]GA2o and extraction. No [12C]GA1 and only traces of [12C]GAs (in one line) were detected in the two Na le lines examined. These results are discussed in relation to recent observations on dwarfism in rice and maize. Key words: Gibberellin metabolism (isotope ratios) - Internode growth - Pisum (gibberellin metabolism). Abbreviations: GA n= gibberellin An; GC-MS = gas chromatography-mass spectrometry; HPLC=high-pressure liquid chromatography

Introduction

Although the gibberellins (GAs) have long been implicated in the regulation of stem growth in plants their precise role is still unclear. Recent progress in defining the physiological relevance of these hormones has centred on the use of stemlength mutants, following the pioneering studies of Phinney (1956, 1961) and Brian (1957). In maize, the dwarf-5 (d5) mutation reduces the Bactivity of ent-kaurene synthetase (Hedden and Phinney 1979), an enzyme which acts early in the GA biosynthetic pathway. There is also good bioassay evidence for the metabolic steps blocked by the dl, d2 and d3 mutations in maize (Phinney and Spray 1982) and the dx and dy mutations in rice (Murakami 1972). Our interest has been directed towards the garden pea (Pisum sativum L.) in which at least five loci are known to directly influence internode length (Reid et al. 1983). Until recently the action of the most extensively studied of these, the Le locus (which confers either a tall (Le-) or dwarf (le le) phenotype on a wild-type background), remained obscure (see McComb 1977). However, it has now been shown that the gene Le is associated with the presence of GAt in the immature shoot tissues of tall peas (Potts et al. 1982; Potts and Reid 1983; Ingrain etal. 1983). Gibberellin A 1 appears to be absent or limiting in dwarf peas. However, vegetative tissues of the dwarf line G2 do contain GA20 (Davies et al. 1982), a major GA of developing seeds of peas (Eeuwens et al. 1973; Frydman and MacMillan 1973; Frydman et al. 1974) and a known precursor of GA~ in at least two other species (Yamane et al. 1975, 1977; Lance et al. 1976). The simplest interpretation of these observations is that the Le gene in pea enables the 3fl-hydroxylation of GA2o to GA t. Evidence supporting this hypothesis is described by Potts and Reid (1983) and Ingram et al.

456

T.J. Ingram et al. : Gibberellins and internode growth in peas

Fig. 1. Phenotypic expression of the genes controlling internode length in peas. (Plants photographed at 25 d old)

(1983). We now report on the identification by gas chromatography-mass spectrometry (GC-MS) of metabolites obtained from feeds of [lac, 3H]GA2o to Le and le genotypes of peas. Similar results are reported for the D 1 locus in maize in the accompanying paper (Spray et al. 1984). Materials and methods Plant material. With the exception of John Innes line 15 the pure Pisum sativum L. lines used in the present study came from the collection at Hobart (Department of Botany, University of Tasmania). Details of their phenotypes (see Fig. 1) and genotypes are as follows: line 23 (tall, Le La and-or Cry Na Lm) ; line 15 (tall, Le La and-or Cry Na Lm); line 53 (dwarf, le La cry s Na Lm); line 57 (microcryptodwarf, le la cry c Na lm); line 1766 (nana, L e L a Cry na Lm). The na le plants used in Section A were nana F 4 descendants of dwarf F 2 segregates from a cross between lines 1766 and 58 (dwarf, le La cry c or cry s Na Lm). Further details of the genetic basis of internode length in peas are given by Reid et al. (1983). The plants used in Section A were grown in heated glasshouses as described by Potts et al. (t982). All other plants were grown in controlledenvironment cabinets under the following conditions : day temperature, 20 ~ C; night temperature, 15~ C; daylength, 18 h; light intensity, 450 gE m - ~ s-1 (photosynthetically active radiation). Compost was John Innes No. 3. Gibberellin treatments. [17-13C, 3H2]Gibberellin Azo (88% 13C,

1.27 GBq retool- 1), prepared as described later, was fed to approx. 20-d-old plants in microdrops of methanol (1 gg of GA gl- 1) applied to each of the uppermost pair of stipules enclosing the unexpanded apical tissue. In the experiments described in Section A, 80 plants ofgenotype na Le (line 1766) and 50 plants of genotype nale were fed with 2 gg of [13C, 3H]GAzo per

plant. After 6 d, the tissue above the treated node was harvested, frozen in liquid nitrogen and extracted. In Section B, 30 plants of each Na line (lines 23, 15, 53 and 57) were fed with 2 t.tg of [a3C, 3H]GA20 per plant and the shoot above the treated node and the treated node itself separately extracted after 5 d. Gibberellin extraction and purification. Extraction and initial purification of GAs were as described before (Ingram et al. 1983). Acidic ethyl-acetate extracts from genotypes na Le and nale (see Results, Section A) were chromatographed on silicic acid columns (Powell and Tautvydas ~967; Durley et al. 1972) using Woelm 'Silica Gel for Partition Chromatography' (Woelm Pharma, Eschwege, FRG). Columns were eluted with a gradient of ethyl acetate in n-hexane generated either by a four-chambered mixer (ha Le extract) or stepwise by hand (ha le extract). Appropriate fractions were purified further by highpressure liquid chromatography (HPLC) (see below). Acidic ethyl-acetate extracts from Na lines (Section B) were purified by HPLC, first as the free acids and then as the methyl esters. This technique enabled separation of the methyl esters of GA 1 and the c~, fl-unsaturated ketone catabotite of GA29 , which as the free acids co-eluted on semi-preparative scale reverse-phase HPLC under our conditions. The HPLC was performed on a stainless-steel column (250 mm long, 8 mm diameter i.d.) packed with ODS Hypersil (5 p.m) (Shandon Southern Products, Runcorn, Cheshire, UK). The column was eluted with methanol in distilled water (containing 1% acetic acid in the case of the GA-free acids), delivered by two pumps controlled by a solvent programmer (see Jones et al. 1980). Metabolites obtained from genotypes na Le and nale (Fig. 2; Results, Section A) were purified by isocratic elution with 35% methanol. Extracts from Na lines (see Results, Section B) were eluted with an exponential gradient of methanol, 40 to 70% in 30 rain (see Fig. 4). The solvent flow rate was 2.5 ml rain-1. Radioactivity was determined by liquid scintillation counting of aliquots from chromatographic fractions.

T.J. Ingrain et al.: Gibberellins and internode growth in peas

457

Gas chromatograph~:-mass spectrometry. Appropriate fractions from HPLC were combined, reduced to dryness and the na Le and nale extracts @hich remained as the free acids after HPLC) methylated with ethereal diazomethane. The trimethylsilyl ethers (TMS) of tl~e methyl esters were prepared immediately prior to GC-MS b~ addition of bis-trimethylsilyltrifluoroacetamide to the extra~zts. GC-MS was performed on a VG 70-50 computerised mas~ spectrometer (V.G. Analytical, Wythenshawe, Manchester I UK) coupled to a DANI-3800 gas chromatograph (Kontron j[nstruments, St. Albans, Herts., UK) fitted with a vitreous silich capillary column (25 m long, 0.2 mm inner diameter) coated *ida OV-1 (GC/(Chromatography) Northwich, Cheshire, U~). The interface temperature was 250~ C. Conditions for GC! were: injector temperature, 260~ C; Helium carrier-gas pressurt, 2 bar. Samples were injected at an oven temperature of 35~ C in the Grob splitless injection mode and the GC programmed from 35~ to 150~ at 13~ ram- . Data were acquired from 150~ C to 300~ C at 3~ C rain- 1. Conditions for MS we}e: electron energy, 24 eV; source temperature, 220~ C ; scan rate, 0.7 s decade- ~. A mixture of n-alkanes (C~6-C34; BDH dhemicals, Dagenham, Essex, UK and Parafilm, Gallenkamp,i London, UK) was co-injected with each sample as internal standards for GC retention times (Gaskin et al. 1971). Gibberellin mass spectra were obtained after background subtraction of averaged neighbouring scans. Percentages of [lzC]- and [~3C]GJXswere calculated from the relative intensities of ions in th~ molecular ion (M +) clusters (see Sponsel and MacMillan ~ 19.,78).The 'fit factor' (Tables 2, 3) is defined as 1 - Z 1/xZ where!x equals the + ve or - v e difference in intensity between each ibn in the normalised observed and calculated M + clusters. I

"

and reverse-phase HPLC and 97% pure by HPLC radio-counting.

~~ A c COOCH3

~OAc R - i COOCH3

:

1

Recoveries of GAsJ High quenching of radioactivity in the initial methanolic extrac{s of Na lines prohibited direct measurements of GA losses incurred prior to obtaining the acidic ethyl-acetate fractions. Since between 75% and 85% of the applied radioactivity was recovered after partitioning in the summed extracts of treated and expanded tissues, and no account was taken of possible basipetal transport of label from the treated node, or of [13C, 3H]QA2o turnover during the 5 d between [13C, 3H]GA20 application and extraction, these losses were probably relatively low. However, the results in Table 3 are adjusted for losses which variod between 20% and 30% after obtaining the acidic ethyt-aceta{e fractions. Preparation of[lTJ3C, 3H2]GA2o. Conjugate reduction of the enone (structure i) with sodium borohydride in the presence of copper (I) chloride (Duri et al. 1981) gave a separable mixture of 3-epiGA~ methyl ester-13-acetate (2) and a small amount of GA 1 methyl ester-13-acetate (3). As described by Beale et al. 0980), treatment of 3-epiGA~ methyl ester 13-acetate (2) with phosphoryl chloride in pyridine gave the 3fl-chloro compound (4) wliich on treatment with tri-n-butylstannane and initiator gave GA2o methyl ester-13-acetate (5). Oxidation of (5) with osmium ~etroxide-sodinm periodate gave GA2o methyl ester t7-norketone 13-acetate (6), m.p. 228-229 ~ C. fi(CDCI3) 1.12(s, 18-H3), 2.59(d, J=10.5 Hz, 5-H), 2.78(d, J=10.5 Hz, 6-H), 3.77(s, -OCH3). m/z 390 (M +, 10%), 359(8), 348(11), 320(14), 302(60),'288(61), 274(64), 260(26), 215(31). The ylide, triphenyl P= 13C, 3H2, was prepared as follows. [~3C]Methyltriph~enylphosphonium bromide was converted to the [i 3C, ~H3]salt by exchange with 3H20 (Bearder et al. 1976) and the doubly-labelled salt was treated with sodium hydride. Addition of the resultant [t~C, 3H2]ylide to the ketone (6) gave [17-~3C, 3H2]GA20 methyl ester 13-acetate (7). Alkaline hydrolysis of (7) (Beale et al. 1980) afforded the required [17-1~C, 3H2]GA2o with 0.88 atoms of 13C per molecule and specific activity 1.27 GBq mmol -~. The product was 99% pure by GC

(1)

(2) (31 (4) (5)

R =o~OH R =/50H R = #[:L R: H

(6) ~= o (7) R = 13C,3H2

Results A. Metabolism of [13C, 3H]GA2o by na Le and na le plants. P r e v i o u s studies h a v e p r o v i d e d s t r o n g c i r c u m s t a n t i a l evidence t h a t the L e gene in p e a regulates the 3/~-hydroxylation o f GA20 to G A 1 (Potts a n d R e i d 1983; I n g r a i n et al. 1983). In o r d e r to c o n f i r m this hypothesis, large scale feeds o f [13C, 3H]GAzo to na L e a n d n a l e g e n o t y p e s were c o n d u c t e d . T h e acidic ethyl-acetate fractions obtained f r o m extracts o f the tissue e x p a n d e d in the 6 d following t r e a t m e n t with [13C, 3H]GA2o were c h r o m a t o g r a p h e d on silicic acid c o l u m n s (Fig. 2). As before (Ingrain et al. 1983) the na L e plants p r o d u c e d three m a j o r m e t a b o l i t e s whilst the n a l e plants p r o d u c e d only one. A p p r o p r i a t e fractions were c o m b i n e d , purified by H P L C a n d the m e t h y l T M S derivatives analysed by capillary G C - M S . T h e three m e t a b o l i t e s present in the na Le plants were identified as [i3C, 3H]GA1, [13C, 3H]GA29 a n d [13C, a H ] G A 8 (Fig. 3). Only [laC, 3H]GA29 was present at detectable levels in the na le plants a l t h o u g h traces o f o t h e r m e t a b o l i t e s were indicated b y the radioactivity profile f r o m the silicic-acid colu m n . N e i t h e r g e n o t y p e c o n t a i n e d any [~3C, 3 H ] G A z g - c a t a b o l i t e which is the m a j o r end p r o d uct o f GA20 m e t a b o l i s m in d e v e l o p i n g seeds (Sponsel a n d M a c M i l l a n 1978, 1980; D u r l e y et al. 1979; G a s k i n et al. 1981). M o r e o v e r , neither genotype c o n t a i n e d a p p r e c i a b l e levels o f highly p o l a r m e t a b o l i t e s (such as G A - c o n j u g a t e s ) ; in b o t h cases only small a m o u n t s o f radioactivity p a r t i t i o n e d into n - b u t a n o l at p H 3 (Table 1). C a l c u l a t i o n o f the ~3C-isotope i n c o r p o r a t i o n in each m e t a b o l i t e s h o w e d no dilution with e n d o g enous [12C]GAs (Table 2). This is in a g r e e m e n t with the results o f Potts a n d R e i d (1983) which indicate t h a t na blocks G A synthesis p r i o r to the p r o d u c t i o n o f biologically active G A s , H o w e v e r , it does n o t preclude the p r o d u c t i o n o f G A s in na

458

T.J. Ingram et al. : Gibberellins and internode growth in peas

[13C.3H] GA20

25-

/

20-

a

[13C.3H] GA1 / [13C.3HlGA29

15~D

10-

.....r'

,'o 2'0

Fraction

No.

5'o

t13E.3HlGA2O

60-

/I

/

[13C3H1GA29

b

50406m 3020100-

~.~ ~

x\xxx\x\\\~x\xx\xx~

2d40gOr 5~456 518gO 62 6k 66 18 7b 42 7~4716"]8 dO ~5 90 100 % ethyl acetate in n-hexane Fig. 2 a, b. Silicic-acid partition chromatography of acidic ethylacetate extracts from genotypes na Le (a) and n a l e (b) fed with [13C, 3H]GAzo. Aliquots (1%) were removed from each fraction for determination of radioactivity9 The bars indicate fractions combined and analysed by HPLC and GC-MS (see Fig. 3)

plants at levels below those detectable by current methods. B. Metabolism o f / 1 3 C , 3H]GA2o by Na Le and Na le plants. The results in the previous section were confirmed and extended by conducting feeds of [13C, 3H]GA/o to a number of lines carrying Na and either Le or le. In all lines less than 10% of the applied radioactivity was recovered in the tissue expanded after treatment (Table 1), and, of this, 65% or more partitioned into the acidic ethylacetate fraction. These figures contrast with the na lines in Section A, in which approx. 50% of the applied radioactivity was transported to the expanding tissue (Table 1). The difference between Na and na lines was mainly due to the levels of unmetabolised [13C, 3H]GA20 recovered in the expanded tissue (Table i). In Na plants high levels of label remained in the treated tissue (between

70% and 80% of the applied label), of which approx. 90% was [laC, 3H]GA2o. Although traces of GA29 (in line 53) and GA29and GA 1 (in line 23) were detected by GC-MS in the treated tissue (results not shown), their levels were very low in relation to the amounts present in the expanded tissue of the same lines. These results indicate that the major compound exported from the treated leaf was [13C, 3H]GA20 and, therefore, that the metabolites measured in the expanded tissue were primarily produced 'in situ' and not in the treated node itself. The very high levels of [13C, 3H]GA20 transported to the expanding tissue of na (compared with Na) lines were probably the result of a combination of factors, including a) more efficient penetration of the applied [13C, 3H]GA2o into the treated leaf, b) the much closer proximity of the treated leaf to the apex, and c) the absence of dilution with endogenous GAs. Analysis of the acidic ethyl-acetate fractions from extracts of the apical tissues of Na lines by HPLC and GC-MS showed that all genotypes contained appreciable amounts of [13C, 3H]GA29.As predicted, however, only the tall (Le) lines 15 and 23 contained [13C, 3H]GA1 and [t3C, 3H]GAs (Fig. 4). The presence of 13C label (88%) in the applied GA2o and knowledge of the radioactivity in appropriate HPLC fractions permitted estimation of the levels of endogenous GAs (Table 3), in the manner described by Sponsel and MacMillan (1978). Several important points emerge from these results. Firstly, the [13C]GA s showed considerable dilution with endogenous [~2C]GAs, lending support to the conclusion that the metabolism of [13C, 3H]GA2o mirrored that of endogenous [12C]GA2o. Secondly, the endogenous levels of GA29 in line 53 and of GA29 plus GA29-catabolite in line 57 were greater than those of GA29 in the tall lines 23 and 15. This might be expected in view of the inability of the former two lines to metabolise GA2o to GA 1 and GA 8. No evidence was obtained for increased levels of [12C]GAzo in the two Na le lines. Clearly, however, differences in the absolute levels of GAs exist between all lines. Thirdly, traces of GA 8 were identified by GC-MS in the dwarf line 53 (but not in line 57), indicating that the le gene may show a small degree of leakage. However, comparison of the molecular ion currents of methyl TMSGA s and methyl TMSGAz9 in line 53 gave a ratio of approx. 0.02 (compared with approx. 1.6 in the tall lines 15 and 23), showing that the 3fl-hydroxylation pathway is extremely minor in the le genotype. Finally, on the basis of their HPLC radioactivity profiles, lines 23, 15 and 53 produced only traces of putative [~ac, 3H]GA29-catabolite (Fig. 4). Where the appro-

T.J. Ingram et al. : Gibberellins and internode growth in peas

9

323

73

a

-:t l l. . ,.....1 201 ~0 8~ 6o4

40

208

1~o

314

~o

2~o

26o

L

2~o

~49

377

. .

3~o

,.

.~

~o

~b0

~o

~"8 0

I

b

l

73 1

0 ~

507

.,.1....... & ..~L_.IL. . . . . . . . . ~.L. . . . . . . . . .

~'0

'~

459

~

:

80

so7

120

160

200

280

240

320

360

~00

4~0

480

520

C

80 60 595 ~0

208

20 0

,

4o

"'-

.

k

.

~o

._ . L . . [ . ~ . ~ , J

Iko

....

i;o

~

239 ,,t . . . .

2bo

L..

2~o

380 ~.

~

z'8o

..~

. . . . . .

..ihJ

3'2o

3~o

m/e

4~9 I;

~oo

'

"

i' [~

.,o

"

r

gso

slo

I

6bo

6ko

Fig. 3a-c. Mass spectra of [13C, aH]GA 1 (a), [laC, 3H]GA29 (b) and [t3C, 3H]GA8 (e) obtained from the [13C, 3H]GAzo feed to genotype na L~ (see Fig. 2a). Kovats G C retention indices (Gaskin et al. 1971) for G A 1, GA29 and G A 8 were 2662, 2680 and 2816, in the order given T a b l e 1. [t3C, 3H]Oibberellin A2o feeds to Le and le genotypes.

Distribution of ra~tioactivity after partitioning

Table 2. [laC, 3H]Gibberellin Gzo feeds to na Le and n a l e genotypes, it 3C]_isotope incorporation

1

Line

Ge~otype

Percent of radioactivity fed a Acidic ethyl acetate

Acidic Residual n-butanol aqueous

57.5 (48.8) b 45.1 (38.1) 5.2 (3.0) 6.6 (3.2) 2.5 (1.1) 2.3 (1 A)

t.5 3.0 0.5 1.4 0.4 0.4

Line

Genotype

Metabolite Percent isotopic composition ~ 13C

z2C

Fit factor b

88 n.d. c n.d. 88 92 90

12 n.d. n.d. 12 8 10

0.9863 0.9194 0.8580 0.8250

t

-

1766 53 57 23 15

nal F na Le Na !e Na ie Na Le Na Le

-

-

na Ie

0.6 0.8

1766

na Le

a Fed with approx. 6 kBq p l a n t - 1 b Percent remaining as [13C, 3H]GA20 (in parentheses)

priate HPLC fractions were examined by GC-MS (line 15), correspondingly low levels of endogenous [t2C]GA29-ca~tabolite were observed (Table3). These results I are consistent with the absence of [13C, 3H]GA~9-catabolite in the apical tissues of na plants described in Section A. In contrast line 57, which differs from the other three lines in carrying recessive alleles at the la, cry and lm loci (see Reid et al. 1983), contained appreciable levels of both [13C] and endogenous [12C]GAzgcatabolite (Fig. 4, Table 3).

GAz9 GA z GA 8 GA29 GA I GA 8

Substrate [13C] c o m p o s i t i o n = 8 8 % b Fit of calculated to observed molecular ion clusters. Values of 1.0 mean an exact fit. Factors below 0.9 imply relatively inaccurate fit ~ Not detected a

Discussion

In the expanding apical tissues of all Le lines of peas examined, [I3C, 3H]GA20 was metabolised to [13C, 3H]GAt, [13C, 3H]GA8 and [13C, 3H]GA29. In contrast, [13C, 3H]GA29 and, in one line, [13C, 3H]GA29-catabolite, were the only detected products in plants homozygous for the le gene. Qualitatively the same metabolism occurred irrespective of the genotype at the N a locus, which

460

T.J. Ingram et al. : Gibberellins and internode growth in peas GA 8 GA29 GA 1 H i.~ h-~

OA29~.cafabo[ife 1-4

a 2O

l

10

S

/

//t

r .........

100

)

70

//

-60 -50

0

........ I

I

l+0 I I

III

l l l l l l l l l l l l l l l 1 1 1 1 1 1 1 1 1 1

GA8 H

25-

GA29 GA1 H H

@A29-cafabo[ife H

b

20-

r- . . . . . . . . .

100

,

i15c13

10-

//

I

70 T

L

/

//

si

- 60 %

////

._~

///

L~\\\\\\\\\\\\\\~\\\~ l III

"50

I

IIII111111111

GAs GA29 GAI H I-~ I-4

t~

-4o ~ 1 I I I I I I I I I I

GA29-cafabo[ife t-.4

53 Bq

C r........

100

I I

$ I

/)

70

/ //

k \ \ \ \ \ \ \ \ N \ \ ~I III

I

I I

I

I11~II

II

II

-60

k\\\\\\\\~l I11

III

20

~'0 I i

ii

iii

I

i

i i i

i i

30

Fracfion

Fig. 4a--e. High-pressure liquid chromatography of extracts from the tall line 23 (a), dwarf line 53 (b) and microcryptodwarf line 57 (e) fed with [13C, 3H]GA2o. Aliquots (2%) were removed from each fraction for determination of radioactivity. The bars indicate fractions combined and analysed by GC-MS (see Table 3). Methylated GA standards eluted in the positions indicated. The tall line 15 (data not shown) gave similar results to line 23

is believed to control a step early in the GA biosynthetic pathway (Potts and Reid 1983). Gibberellin A 8 and GA29, respectively, are the 2fl-hydroxylated deactivation products of GA 1 and GA2o (Fig. 5) and exhibit low to negligible biological activity in a wide range of assays (Reeve and Crozier 1974). The 2fl-hydroxylation of [3H]GA20 to [3H]GA29 has been described before in both developing seeds and seedlings of dwarf peas (Railton et al. 1974; Frydman and MacMillan 1975; Sponsel and MacMillan 1977; Durley et al. 1979). From the present work, the occurrence of [13C, 3H]GA1 and [13C, 3H]GAs only in L e lines confirms our earlier proposals (Potts and Reid 1983; Ingrain et al. 1983) that the L e gene controls the 3fl-hydroxylation of GAge to give GA1 (Fig. 5). The same step is controlled by the d 1 gene in maize (Spray et al. 1984). The physiological importance of this conversion was suggested by Phinney and Spray (1982) from comparison of the biological activities of GA2o and GA~ on different dwarf mutants of maize and the same conclusion may be drawn from similar studies on rice (Murakami 1972) and peas (Ingram et al. 1983). Therefore, in at least three species 3fl-hydroxylation is crucial for the stimulation of stem growth by GAs. GA~ has been identified by GC-MS or GC-MS multiple ion monitoring in all three plants (Kuroguchi et al. 1979; Suzuki et al. 1981; Hedden et al. 1982; Potts and Reid 1983; Ingram et al. 1983) but appears to be present at much lower levels in vegetative tissues of rice and maize than in peas. Estimates of the levels of endogenous GAs in N a lines of peas (Table 3) are similar to those observed before in vegetative tissues of peas (Davies et al. 1982) and other species (Kuroguchi et al. 1979; Jones and Zeevaart 1980; Suzuki etal. 1981). Therefore, it is likely that these values reflect the normal endogenous levels, despite perturbation of the GA economy by exogenous [13C, aH]GA2o. This conclusion is supported by the strong dilution of [13C, 3H]GA2o metabolites by endogenous [~2C]GAs (Table 3). Taken together these observations indicate that the metabolism of [~3C, 3H]GA2o accurately represents GA metabolism within the endogenous system. In contrast to N a lines, the absence of any dilution of [13C, 3H]GA2o metabolites in na lines (Table 2) corresponds with previous reports that the na mutation prevents the production of biologically active C19-GAs in vegetative tissues (Potts and Reid 1983). However, normal levels of GAs are present in developing seeds of na lines (Potts and Reid 1983), a finding that has important physiological implications.

T.J. Ingrain et al. : GibberelIins and internode growth in peas

461

Table 3. [~3C, aH]Gibberellin A~0 feeds to Na le and Na Le genotypes. [~3C]-isotope incorporations and endogenous G A levels Line

53

Gen&ype

Metabolite

Na le

Percent isotopic composition"

Approximate level b (ng p l a n t - 1)

13C

12C

Fit factor ~

Exogenousa

Endogenous

GAzo GA29 GAz9-catabolite ~ GA 1 GA 8

44.8 23.3

55.2 76.7 _ n.d.

0.9905 0.9942 _ -

24.6 12.6 _ 0

23.7 34.9

n.d. ~ Trace f

0 Trace

57

Na le (lm la cry ~)

GAzo GA29 GAzg-catabolite GA 1 GA 8

78.8 65.1 46.1 n.d. n.d.

21.2 34.9 53.9 n.d, n.d,

0.9872 0.9921 0.9742 -

49.8 36,5 7.4 0 0

5.8 12.8 6.7 0 0

23

Na Le

GAzo GA29 GAz9-catabolite g GA 1 GA 8

29.4 14.2 . 10.5 15.6

70.6 85.8

0.9949 0.9828 . 0.9901 0.9852

18.3 3.2

36.5 16.8

4.2 6.0

31.0 27.8

GAzo GA29 GA29-catabolite GA 1 GA 8

53.8 28.5

0.9844 0.9571 0.9111 0.9663

37.5 4.2

11.1 8.7

15

Na Le

.

. 89.5 84.4 46.2 71.5

Trace ~ 23.5 23.1

76.5 76.9

Trace 3.4 5.2

9.4 14.5

Substrate [1~C] ~ o m p o s i t i o n = 8 8 % Approximate levels of G A s in tissue expanded after treatment with [13C, 3H]GA2o Fit of calculated to observed molecular ion clusters. Factors below 0.9 imply relatively inaccurate fit Estimated from HPLC radioactivity profile Not detected by GC-MS f Very weak spect}um (insufficient intensity to calculate [13C] incorporation) g Not analysed by GC-MS but only traces indicated by HPLC a b c d

GA29

:~

"

GA29- catabo[ite

cooH

/O

%

coo~

%

OH

HO

/,0

OH

COOR

GA1

fiA 8

Fig. 5. The effect of the le mutation on G A metabolism in peas

A feature 0f the metabolism of [~C, 3H]GA2o was the absence of appreciable levels of [13C, 3H]GA29-catabolite in all lines except the microcryptodwarf l~ne 57. This observation is interesting in view of the ::clear accumulation of GA29-catabolite in developing seeds (Sponsel and MacMillan 1978, 1980; Durley et al. 1979) but has been found consistently in all feeds of [SH]GA2o to vegetative tissues of lines carrying a dominant allele at one or both of the loci La and Cry. Davies et al. (1982) also report that GA29-catabotite was present at

appreciably lower levels than GAao and GAa9 in vegetative tissues of the dwarf line G2. However, lines carrying recessive alleles at both of these loci, expressed in the most extreme version by the slender phenotype (genotype la cryS), do appear to contain relatively high levels of GA29-catabolite (line 57, Table 3 and Fig. 4). The importance of this difference is not clear but it may help towards our understanding of how the La and Cry genes operate to regulate stem growth in peas (see Reid et al. 1983). From their work on dwarf maize mutants, Phinney and Spray (1982) have made the important prediction that GA t is the only GA active per se in regulating stem growth. The presence of trace amounts of GA 8 in the dwarf pea line 53 (Table 3) is consistent with this conclusion, since it implies limited leakage of the le gene. Suzuki et al. (1981) also reported low levels of GA 1 in the dwarf rice cultivar Tan-ginbozu (genotype dx Dy), which is believed to be blocked early in the GA biosynthetic pathway (Murakami 1972). However, GAzo may also show intrinsic activity,

462

a possibility presently under investigation in peas. Endogenous GAs are also implicated in the control of pod development in peas (Eeuwens and Schwabe 1975; Rosenstand 1978; Garcia-Martinez and Carbonell 1980). In this system, GA2o may be active without conversion to GA 1 as there is no obvious phenotypic difference between the pods of L e and le genotypes. Furthermore, only GA20 and G A 2 9 have been identified in the pods of either genotype (Komoda etal. 1968; Ingrain and Browning 1979). However, although applied GA20 can fully restore pod elongation and inflation in 'deseeded' fruits of the dwarf cultivar Progress No. 9, it has only approx. 10% the activity of GA 3 (Sponsel 1982). Comparison of the activities of GAzo and GA1 on pod growth of tall and dwarf peas may indicate whether or not the gene L e is expressed within the pod tissue. This is of particular relevance in view of the absence of GA~ in developing seeds of tall peas (Eeuwens et al. 1973; Ingrain 1980) and the conclusion from this and more recent work (Potts and Reid 1983) that L e is not expressed within the seed tissues. In contrast, GA2o and its 2fl-hydroxylated derivative, G A 2 9 , are major GAs of developing seeds of both tall and dwarf peas (Eeuwens et al. 1973; Frydman and MacMillan 1973; Frydman etal. 1974; Ingram 1980). Although we have now established the importance of 3fl-hydroxylation in the regulation of stem growth in peas, GA20 (both [13C, 3H] and endogenous [~2C] species) also underwent appreciable 2flhydroxylation to form GA29 in the apical tissues of L e genotypes (lines 1766, 15 and 23). 2fl-Hydroxylation is widely recognised as a deactivation step (Rappaport et al. 1974; Hoad et al. 1982). The relative properties and compartmentation of the GA20 2fl- and 3fl-hydroxylases must therefore be critical for the control of stem growth. Although the turnover of GA20 via these competing pathways cannot be determined from the present work, the [13C]:[12C] isotope ratios of GA1, GA s and G A 2 9 in lines 23 and 15 (Table 3) indicate that they occur to a similar extent in the expanding tissue of L e lines. The same conclusion does not necessarily hold for the mature tissues of L e plants. Both grafting studies (Lockhart 1957; McComb and McComb 1970; Lockhard and Grunwald 1970; Reid et al. 1983) and extraction of endogenous GAs (Potts et al. 1982) indicate that the influence of the L e gene is largely limited to the immature, expanding shoot tissue. However, graft-transmission of a hormonal factor that influences stem growth has been demonstrated between mature N a stocks and na scions (Reid et al. 1983). These find-

T.J. Ingram et al. : Gibberellins and internode growth in peas

ings have important consequences for the action of GAs in regulating stem growth in peas and are presently under wider investigation. We are grateful to Drs. W.C. Potts and V.M. Sponsel for constructive advice during the work. In addition we thank Miss J. Smith, Miss A. Combes and Miss S. Hickey for technical assistance, and the Australian Research Grants Committee and the Royal Society for financial support. T.J.I. gratefully acknowledges the receipt of a Royal Society Pickering Research Fellowship.

References Beale, M.H., Gaskin, P., Kirkwood, P.S., MacMillan, J. (1980) Partial synthesis of gibberellin A 9 and [3~- and 3fl-2H1] gibberellin A9; gibberellin A s and [lfl,~3-2H2 and -3H2]gibberellin As; and gibberellin A2o and [1/?,3~-2H2 and -3H2]gibberellin A2o. J. Chem. Soc. Perkin I, 885-891 Bearder, J.R., Frydman, V.M., Gaskin, P., MacMillan, J., Wels, C.M.~ Phinney, B.O. (1976) Fungal products. Part XVI. Conversion of isosteviol and steviol acetate into gibberellin analogues by mutant B1-41 a of Gibberellafujikuroi and the preparation of [3H] gibberellin A2o. J. Chem. Soc. Perkin I, 173-178 Brian, P.W. (1957) The effects of some microbial metabolic products on plant growth. Symp. Soc. Exp. Bot. II, 166-182 Davies, P.J., Emshwiller, E., Gianfagna, T.J., Proebsting, W.M., Noma, M., Pharis, R.P. (1982) The endogenous gibberellins of vegetative and reproductive tissues of G2 peas. Planta 154, 266-272 Duri Z.J, Fraga, B.M., Hanson, J.R. (1981) Preparation of gibberellins A 9 and A2o from gibberellic acid. J. Chem. Soc. Perkin I, 161-163 Durley, R.C., Crozier, A., Pharis, R.P., McLaughlin, G.E. (1972) Chromatography of 33 gibberellins on a gradient eluted silica gel partition column. Phytochemistry 11, 3029-3033 Durley, R.C., Sassa, T., Pharis, R.P. (1979) Metabolism of tritiated gibberellin A2o in immature seeds of dwarf pea, cv. Meteor. Plant Physiol. 64, 214-219 Eeuwens, C.J., Gaskin, P., MacMillan, J. (1973) Gibberellin A2o in seed of Pisum sativum L., cv. Alaska. Planta 115, 73-76 Eeuwens, C.J., Schwabe, W.W. (1975) Seed and pod wall development in Pisum sativum L. in relation to extracted and applied hormones. J. Exp. Bot. 26, 1-14 Frydman, V.M., Gaskin, P., MacMillan, J. (1974) Qualitative and quantitative analyses of gibberellins throughout seed maturation in Pisum sativum cv. Progress No. 9. Planta 118, 123-132 Frydman, V.M., MacMillan, J. (1973) Identification of gibberellins A2o and A29 in seed of Pisum sativum cv. Progress No. 9 by combined gas chromatography-mass spectrometry. Planta 115, 11-15 Frydman, V.M., MacMillan, J. (1975) The metabolism of gibberellins A9, A2o and Az9 in immature seeds of Pisum sat# rum cv. Progress No. 9. Planta 125, 181-195 Garcia-Martinez, J.L., Carbonell, J. (1980) Fruit set of unpollinated ovaries of Pisum sativum L. Influence of plant growth regulators. Planta 147, 451-456 Gaskin, P., Kirkwood, P.S., MacMillan, J. (1981) Partial synthesis of ent-13-hydroxy-2-oxo-20-norgibberella-l(lO),16diene-7,19-dioic acid, a catabolite of gibberellin Az9 , and of related compounds J. Chem. Soc. Perkin I, 1083-1091

463

T.J. Ingram et al. : ~ibberellins and internode growth in peas Gaskin, P., MacMillan, J., Firn, R.D., Pryce, RJ. (1971) 'Parafilm': a conver~ient source of n-alkane standards for the determination & gas chromatographic retention indices. Phytochemistry 10, 1155-1157 Hedden, P., Phinn~y, B.O. (1979) Comparison of ent-kaurene and ent-isokaurene synthesis in cell-free systems from etiolated shoots of ~aormal and dwarf-5 maize seedlings. Phytochemistry 18, 1475-1479 Hedden, P., Phinney, B.O., Heupel, R., Fujii, D., Cohen, H., Gaskin, P., MaCMillan, J., Graebe, J.E. (1982) Hormones of young tasselsf of Zea mays. Phytochemistry 21,391-393 Hoad, G.V., MacNllillan, J., Smith, V.A., Sponsel, V.M., Taylor, D.A. (1982~ Gibberellin 2fl-hydroxylases and biological activity of 2tq-a!~yl gibberellins. In:Plant growth substances 1982, pp. 91-100, Wareing, P.F., ed. Academic Press, London New York Ingrain, T.J. (1980) Gibberellins and reproductive development in peas. Ph.D. t~aesis, University of East Anglia, Norwich Ingrain, T.J., Bro~vning, G. (1979) Influence of photoperiod on seed development in the genetic line of peas G2 and its relation to @anges in endogenous gibberellins measured by combined igas chromatography-mass spectrometry. Planta 146, 423~432 Ingrain, T.J., Reid~ J.B., Potts, W.C., Muffet, I.C. (1983) Internode length inl Pisum. IV. The effect of the gene Le on gibberellin metabolism. Physiol. Plant. 59, 607-616 Jones, M.G., Zeevaart, J.A.D. (1980) The effect of photoperiod on the levels of seven endogenous gibberellins in the longday plant Agro)temma githago L. Planta 149, 274-279 Jones, M.G., Metzger, J.D., Zeevaart, J.A.D. (1980) Fractionation of gibberetlins in plant extracts by reverse phase high performance l~quid chromatography. Plant Physiol. 65, 218-221 Komoda, Y., Isog}ai, Y., Okamoto, T. (1968) Isolation of gibberellin A2o frbm pea pods. Sci. Pap. College Gen. Educ. Univ. Tokyo 1~, 221-230 Kuroguchi, S., Mlrtrofushi, N., Ota, N., Takahashi, N. (1979) Identification ~f gibberellins in the rice plant and quantitative changes 0f gibberellin A19 throughout its life cycle. Planta 146, 185-191 Lance, B., Durley, R.C., Reid, D.M., Thorpe, T.A., Pharis, R.P. (1976) Metabolism of [~H]gibberellin A2o in light- and dark-grown tobacco callus cultures. Plant Physiol. 58, 387-392 Lockhard, R.G., Grunwald, C. (1970) Grafting and gibberellin effects on the growth of tall and dwarf peas. Plant Physiol. 45, 160-162 Lockhart, J.A. (1057) Studies on the organ of production of the natural gibberellin factor in higher plants. Plant Physiol. 32, 204-207 McComb, A.J. (1977) Control of root and shoot development. In: The physiology of the garden pea, pp. 235-263, Sutcliffe, J.F., Pate, J.S!, eds. Academic Press, London New York McComb, AJ., NlcComb, J.A. (1970) Growth substances and the relation between phenotype and genotype in Pisum satirum. Planta 9t, 235-245 Murakami, Y. (1972) Dwarfing genes in rice and their relation to gibberellin biosynthesis. In: Plant growth substances 1970, pp. 166+174, Carr, D.J., ed. Academic Press, London New York Phinney, B.O. (11956) Growth response of single gene dwarf mutants in maize to gibberellic acid. Proc. Natl. Acad. Sci. USA 42, 185-.189 Phinney, B.O. (1961) Dwarfing genes in Zea mays and their

relation to the gibberellins. In: Plant growth regulation, pp. 489-511, Klien, R.M., ed. Iowa State University Press, Ames IOWA Phinney, B.O., Spray, C. (1982) Chemical genetics and the gibberellin pathway in Zea mays L. In: Plant growth substances 1982, pp. 101-110, Wareing, P.F., ed. Academic Press, London New York Potts, W.C., Reid, J.B. (1983) Internode length in Pisum. III. The effect and interaction of the Na/na and Le/le gene differences on endogenous gibberellin-like substances. Physiol. Plant. 57, 448-454 Potts, W.C., Reid, J.B., Murfet, I.C. (1982) Internode length in Pisum. I. The effect of the Le/le gene difference on endogenous gibberellin-like substances. Physiol. Plant. 55, 323-328 Powell, L.E., Tautvydas, K.J. (1967) Chromatography of gibbereUins on silica gel partition columns. Nature (London) 213, 292-293 Railton, I.D., Murofushi, N., Durley, R.C., Pharis, R.P. (1974) Interconversion of gibberellin A2o to gibberellin A29 by etiolated seedlings and germinating seeds of dwarf Pisum sativum. Phytochemistry 13, 793-796 Rappaport, L., Davies, L., Lavee, S., Nadeau, R., Patterson, R., Stolp, C.F. (1974) Significance of metabolism of [3H]GA1 for plant regulation. In: Plant growth substances 1973, pp. 314-324. Hirokawa, Tokyo Reeve, D.R., Crozier, A. (1974) An assessment of gibberellin structure-activity relationships. J. Exp. Bot. 25, 431-445 Reid, J.B., Murfet, I.C., Potts, W.C. (1983) Internode length in Pisum. II. Additional information on the relationships and action of loci Le, La, Cry, Na and Lm. J. Exp. Bot. 34, 349-364 Rosenstand, A. (1978) GA treatment of pea pods: implications on intergeneric hybridisation. Pisum Newslett. 10, 75 Sponsel, V.M. (1982) Effects of applied gibberellins and naphthylacetic acid on pod development in fruits of Pisum satirum L. cv. Progress No. 9. J. Plant Growth Regul. 1, 147-152 Sponsel, V.M., MacMillan, J. (1977) Further studies on the metabolism of gibberellins (GAs) A 9, A2o and A29 in immature seeds of Pisum sativum cv. Progress No. 9. Planta 135, 129-136 Sponsel, V.M., MacMillan, J. (1978) Metabolism of gibberellin A29 in seeds of Pisum sativum cv. Progress No. 9; use of [ZH] and [3H]GAs, and the identification of a new GA catabolite. Planta 144, 69-78 Sponsel, V.M., MacMillan, J. (1980) Metabolism of [13Cl]gibberellin A29 to [13C1]gibberellin-catabolite in maturing seeds ofPisum sativum cv. Progress No. 9. Planta 150, 46-52 Spray, C., Phinney, B.O., Gaskin, P., Gilmour, S.J., MacMillan, J. (1984) Internode length in Zea mays L. The dwarf-1 mutation controls the 3fl-hydroxylation of gibberellin A2o to gibberellin A~. Planta 160, 464-468 Suzuki, Y., Kuroguchi, S., Murofushi, N-, Ota, Y., Takahasi, N. (1981) Seasonal changes of GAt, GA~9 and abscisic acid in three rice cultivars. Plant Cell Physiol. 22, 1085-1093 Yamane, H., Murofushi, N., Takahashi, N. (t975) Metabolism of gibberellins in maturing and germinating bean seeds. Phytochemistry 14, 1195-1200 Yamane, H., Murofushi, N., Osada, H., Takahashi, N. (1977) Metabolism of gibberellins in early immature bean seeds. Phytochemistry 16, 831-835 Received 18 October; accepted 15 November 1983