3β,13-dihydroxylated C20 gibberellins from inflorescences of Rumex acetosa L.

Article abstract of DOI:10.1016/S0031-9422(02)00442-9

Using full scan GC-MS a wide range of gibberellins (GAs) was identified in the young inflorescences of the dioecious species Rumex acetosa L., consistent with the ubiquitous early 13-hydroxylation pathway in both male and female plants. In addition, R. acetosa is the first species in which all three 3β, 13-dihydroxylated C₂₀-GAs - GA₁₈, GA₃₈ and GA23 - have been identified in the same organism, suggesting an early 3β,13-dihydroxylation biosynthesis pathway in this species. Authentic GA₁₈, GA₃₈ and GA₂₃ were synthesized and their effects and that of GA₁, a GA common to both pathways, on the time to inflorescence emergence was investigated. GA₁ accelerated the emergence of inflorescences in both male and female plants. In addition some evidence for biological activity per se of the C₂₀-GA₃₈ was obtained.

Full text of DOI:10.1016/S0031-9422(02)00442-9

Phytochemistry 62 (2003) 165–174
www.elsevier.com/locate/phytochem
3b,13-Dihydroxylated C20 gibberellins from inflorescences of
Rumex acetosa L.
Tania S. Stokesa,*, Lewis N. Manderb, Stephen J. Crokerc, Bruce Twitchinb,
David E. Hanked
aPlant Sciences Division, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, LE12 5RD, UK
bResearch School of Chemistry, Australian National University, Canberra, ACT 2601, Australia
cIACR-Long Ashton Research Station, Department of Agricultural Sciences, University of Bristol, Long Ashton, Bristol BS419AF, UK
dDepartment of Plant Sciences, Cambridge University, Downing Street, Cambridge CB2 3EA, UK
Received 26 February 2002; received in revised form 11 August 2002
Abstract
Using full scan GC–MS a wide range of gibberellins (GAs) was identified in the young inflorescences of the dioecious species
Rumex acetosa L., consistent with the ubiquitous early 13-hydroxylation pathway in both male and female plants. In addition, R.
acetosa is the first species in which all three 3b,13-dihydroxylated C20-GAs—GA18, GA38 and GA23—have been identified in the
same organism, suggesting an early 3b,13-dihydroxylation biosynthesis pathway in this species. Authentic GA18, GA38 and GA23
were synthesized and their effects and that of GA1, a GA common to both pathways, on the time to inflorescence emergence was
investigated. GA1 accelerated the emergence of inflorescences in both male and female plants. In addition some evidence for bio-
logical activity per se of the C20-GA38 was obtained.
# 2003 Elsevier Science Ltd. All rights reserved.
Keywords: Rumex acetosa; Polygonaceae; Sorrel; Synthesis; GC–MS; Gibberellin; Early 13-hydroxylation pathway; Early 3b,13-hydroxylation
pathway; GA18; GA38; GA23
1. Introduction
the sex of the plants was not specified, the samples from
male and female plants were analysed separately to
reveal any qualitative differences in their content of
GAs.
Rumex acetosa L. (common sorrel) is a dioecious
perennial (Korpelainen, 1992). Inflorescences are initi-
ated in spring and early summer and it is during the
reproductive phase of development that the phenologi-
cal differences between the sexes become most evident
(Ainsworth et al., 1999). Gibberellins (GAs) in R. acet-
osa shoots have been analysed previously by full-scan
GC–MS (Rijnders et al., 1997). GA53, GA29, GA19,
GA4 and GA1 were identified, which suggested the
operation of two biosynthesis pathways in these tissues:
the early 13-hydroxylation pathway and the non 13-
hydroxylation pathway. In the current paper we report
the results from a study using full-scan GC–MS to
identify the suite of endogenous GAs in young inflor-
escences of R. acetosa. Unlike the previous study, where
Since the seminal workof Lang, many studies have
demonstrated the involvement of GAs in the floral
induction of photoperiod-sensitive species under non-
inductive conditions and/or in accelerating flowering
under inductive conditions (Michniewicz and Lang,
1962; Wilson et al., 1992; Hisamatsu et al., 1998; Lee et
al., 1998). Rosette plants do not show any obvious stem
development in the vegetative state, but produce a
flowering stem during the reproductive phase of devel-
opment. The use of GA biosynthesis inhibitors on some
rosette plants, such as Beta vulgaris (Van Roggen et al.,
1998) and Silene armeria (Talon and Zeevaart, 1990),
has revealed that, although GAs are not involved in
floral induction in these species, they are required for
stem elongation. In Rumex acetosa, the involvement of
GAs during the reproductive phase of development has
not been previously investigated. However, in a related
* Corresponding author. Tel.: +44-115-951-6379; fax: +44-115-
951-6334.
E-mail address: tania.stokes@nottingham.ac.uk (T.S. Stokes).
0031-9422/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(02)00442-9

166
T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
species, R. acetosella, exogenous treatment of plants with
GA3 under long-day (LD) conditions advanced the pro-
duction of ‘floral buds’ by 70 days from the time of treat-
ment, compared to control plants (Bavrina et al., 1991).
From early biochemical studies to the more recent
molecular studies there is persuasive evidence for the
role of GAs in floral induction of the facultative LD
rosette plant Arabidopsis thaliana. Non-vernalizable
‘races’ of this species flower under both LD and SD
conditions, but flowering is considerably accelerated
under LD conditions. An early study by Langridge
(1957) demonstrated that inflorescence induction was
promoted by GA3 under SD conditions. There is now
evidence of at least three flowering pathways in A.
thaliana: a GA-dependent, a LD and an autonomous
pathway (Reeves and Coupland, 2001).
The ga1-3 mutants in A. thaliana are impaired in their
synthesis of CPP synthase which catalyses the conver-
sion of GGPP to CPP in early GA biosynthesis. Conse-
quently the GA levels in these mutants is greatly
reduced (Sun and Kamiya, 1994). Wilson et al. (1992)
found that under continuous light the ga1–3 mutants
initiated flowers, but flowering was delayed compared
to wild-type plants, possibly indicating a partial
requirement for GAs for flowering under such condi-
tions. However, under non-inductive SD conditions,
ga1-3 mutants did not flower unless treated with exo-
genous GA3, therefore showing an absolute requirement
for GAs under such conditions.
In A. thaliana the transition to flowering is partly
dependent on the inflorescence meristem identity gene
LEAFY (LFY). As the plant develops, levels of LFY
expression in the primordia on the shoot apex increase
until a threshold is reached and subsequently floral
rather than leaf primordia are developed (Blazquez et
al., 1998). Blazquez et al. (1998) found that the failure
of the severely GA-deficient mutant ga1-3 to flower in
SD corresponded to the lackof significant upregulation
in expression of LFY during the three months of the
experiment. Constitutive expression of LFY in such
mutants restored their ability to flower under SD
conditions.
Biochemical studies using exogenous applications of a
wide range of authentic GAs have revealed that the
extent of biological activity of GAs in advancing flow-
ering and/or promoting bolting may depend on the
structural features of the GA molecules, as well as the
species under investigation (Evans et al., 1990; Mandel
et al., 1992). For example, GAs without a hydroxyl
group on C-13 but carrying one at C-3, such as GA4,
promote both inflorescence initiation and stem elonga-
tion in Matthiola incana (Hisamatsu et al., 2000), but in
Lolium temulentum GAs with a hydroxyl group at C-3
have reduced levels of activity in promoting inflores-
cence initiation, while the presence of a hydroxyl group
on C-13 increased activity (Evans et al., 1990).
Relatively few GAs are considered to be biologically
active (Lange, 1998). Treatment of mutants defective in
GA biosynthesis with authentic labelled or unlabelled
GAs has demonstrated that certain C19-GAs such as
GA1, GA3, GA4 and GA5 have biological activity per se
(Zeevaart and Talon, 1992; Spray et al., 1996). How-
ever, as suggested by Lange (1998), GAs with relatively
low or negligible activity in common bioassays may be
physiologically active in some plant species, at specific
development stages and/or in certain tissues. Early
bioassays have indicated that some 3b,13-hydroxylated
C20-GAs may have intrinsic biological activity, that is
their activity is not entirely dependent on metabolic
conversion to C19-GAs (Reeve and Crozier, 1974). Fol-
lowing the identification of GA18, GA38 and GA23 in R.
acetosa tissues by full-scan GC–MS, authentic samples
of these GAs were prepared and their biological activity
was examined in flowering time studies.
2. Results and discussion
Nine GAs—GA8, GA29, GA17, GA97, GA17, 16, 17-
dihydro-17-hydroxy GA53, GA18, GA38 and GA23—
were identified for the first time in this species and were
detected in inflorescence tissues (Table 1). GA17 and
GA97, the inactive metabolites of GA19 and GA53
respectively, were only identified in the sample of
inflorescences from female plants. The presence of
GA18, GA38 and GA23 is consistent with the operation
of the putative early 3b, 13-dihydroxylation pathway
(MacMillan, 1997) in floral tissues, although local
accumulation of these GAs from other sites of synthesis
cannot be discounted. As reported previously for R.
acetosa shoot tissues (Rijnders et al., 1997), GA19 and
GA1 were found in the inflorescences, although GA1
was only found in male inflorescences. In addition
GA20, previously identified in shoot tissues (Rijnders
et al., 1997), was identified in GC–SIM quantitative
studies of both male and female inflorescences (data not
included), although it was not detected in the qualitative
studies. The spectrum of GA53 was too contaminated to
allow unambiguous identification, but it was confirmed
in leaves of male plants. Although GA4 and GA9 of the
non-early-13 hydroxylation pathway have been reported
as present in shoot tissues of R. acetosa by Rijnders et
al. (1997), they were not detected in the inflorescences of
either male or female plants in this study.
GA18, GA38 and GA23 are 20-carbon GAs that are
both 3b and 13 hydroxylated. They were initially iden-
tified in the seeds of species such as Lupinus luteus,
Wisteria floribunda and Phaseolus vulgaris (Koshimizu
et al., 1968b; Fukui et al., 1972; Hiraga et al., 1974). It
has been suggested that such GAs may be synthesized
via a pathway in which hydroxylation of GA12 at C-13
to give GA53 and subsequent hydroxylation of C-3 of

T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
167
Table 1
Endogenous gibberellins identified in Rumex acetosa tissues by full scan GC–MS. Identification was based on mass spectra and Kovats retention
a
indices (KRI) of methyl ester trimethylsilyl ether derivatives of GAS
GA
Source
HPLC
KRI
Major ions (m/z) with relative abundance (%)
fractions
GA8
Standard
2791
2790
2791
594 (M+ 100), 579 (3), 565 (2), 535 (7), 504 (4), 448 (24),
379 (6),238 (5), 207 (5), 194 (8)
594 (M+ 100), 579 (4), 565 (4), 535 (11), 504 (11), 448 (29),
Inflorescences (F)
Inflorescences (M)
11–13
11–13
379 (7), 238 (10), 207 (12), 194 (12)
594 (M+ 100), 579 (4), 565 (2), 535 (11), 504 (4), 448 (20),
379 (6), 238 (4), 207 (8), 194 (8)
GA38
Standard
2991
2989
2988
520 (M+ 100), 505 (10), 461 (19), 430 (21), 417 (6), 371 (12),
305 (5), 238 (10), 207 (76), 129 (6)
520 (M+ 46), 505 (1), 461 (9), 430 (13), 417 (11), 371 (16),
Inflorescences (F)
Inflorescences (M)
11–13
11–13
305 (nd), 238 (nd), 207 (55), 129 (18)
520 (M+ 75), 505 (1), 461 (15), 430 (18), 417 (3), 371 (11),
305 (nd), 238 (18), 207 (100), 129 (12)
GA29
Standard
2692
2690
2693
506 (M+ 100), 491 (9), 477 (11), 447 (6), 389 (11), 375 (15),
303 (20), 235 (10), 207 (38), 193 (7)
weakspectrum 506 (M + 100), 491 (nd), 477 (8), 447 (nd),
389 (nd), 375 (4), 303 (19), 235 (22), 207 (23), 193 (11)
506 (M+ 100), 491 (11), 477 (8), 447 (17), 389 (16), 375(7),
303 (26), 235 (4), 207 (21), 193 (7)
Inflorescences (F)
Inflorescences (M)
14–15
14–15
GA1
Standard
2679
506 (M+ 100), 491 (9), 448 (18), 377 (12), 376 (14), 313 (9),
305 (5), 208 (8)
Inflorescences (F)
Inflorescences (M)
16–17
16–17
Not detected
506 (M+ 100), 491 (7), 448 (31), 377 (27), 376 (32), 313 (31),
2678
2654
305 (15), 208 (6)
GA18
GA97
GA19
Standard
536 (M+ 71), 521 (23), 504 (49), 477 (42), 375 (29), 347 (9),
319 (21), 269 (17), 239 (13), 238 (13), 208 (100)
Not analysed
Inflorescences (F)
Inflorescences (M)
18–22
18–22
2656
2692
536 (M+ 65), 521 (17), 504 (30), 477 (51), 375 (22), 347 (11),
319 (17), 269 (25), 239 (27), 238 (14), 208 (100)
Standard
536 (M+15), 504 (6), 477 (10), 446 (4), 387 (8), 371 (9),
327 (12), 239 (46), 208 (100), 207 (84), 179 (25)
536 (M+42), 504 (21), 477 (33), 446 (11), 387 (20), 371 (20),
327 (35), 239 (10), 208 (100), 207 (66), 179 (18)
Not detected
Inflorescences (F)
18–22
18–22
Inflorescences (M)
Standard
2695
2667
462 (M+3), 434 (100), 402 (53), 375 (50), 374 (95), 345 (46),
285 (15), 239 (35), 208 (3), 193 (13)
Inflorescences (F)
Inflorescences (M)
28–30
28–30
2669
2670
462 (M+ 5), 434 (100), 402 (60), 375 (42), 374 (95), 345 (27),
285 (40), 239 (15), 208 (29), 193 (19)
462 (M+ 7), 434 (100), 402 (60), 375 (46), 374 (91), 345 (32),
285 (27), 239 (22), 208 (22), 193 (nd)
16, 17–Dihydro-17-hydroxy GA53
Library spectrum
Inflorescences (F)
Inflorescences (M)
538 (M+ 7), 448 (10), 407 (54), 389 (8), 375 (100), 347 (24),
297 (19), 241 (13), 207 (17), 181 (19)
538 (M+ 9), 448 (12), 407 (35), 389 (7), 375 (100), 347 (22),
28–30
28–30
2774
2776
297 (8), 241 (7), 207 (5), 181 (8)
538 (M+ 8), 448 (11), 407 (33), 389 (4), 375 (100), 347 (27),
297 (10), 241 (8), 207 (2), 181 (13)
GA17
Library spectrum
Inflorescences (F)
492 (M+ 43), 460 (23), 433 (26), 432 (15), 401 (10), 373 (23),
372 (14), 208 (100), 207 (96)
492 (M+ 30), 460 (64), 433 (18), 432 (46), 401 (15), 373 (20),
31–33
31–33
637
372 (26), 208 (100), 207 (38)
Inflorescences (M)
Standard
Not detected
GA53
2551
448 (M+ 74), 433 (18), 416 (49), 389 (56), 357 (12), 251 (13),
241 (34), 208 (100), 193 (15), 181 (31)
Contaminated spectrum
Inflorescences (F)
Inflorescences (M)
Leaves (M)
34–37
34–37
34–37
Contaminated spectrum
448 (M+ 31), 433 (7), 416 (15), 389 (21), 357 (9), 251 (5),
2554
241 (14), 208 (38), 193 (5), 181 (19)
GA23
*
Standard
2677
2679
522 (78), 493 (46), 490 (81), 462 (48), 447 (16), 400 (38),
373 (36), 372 (57), 346 (38), 318 (50)
Inflorescences pooled (F+M)
12–13
522 (72), 493 (47), 490 (27), 462 (29), 447 (8), 400 (12), 373 (nd),
372 (8), 346 (8), 318 (58)
a
GA23 was identified by GC–MS/MS. Its retention time on HPLC is not comparable with those of other GAs as it was separated on the HPLC as a free acid, not a
methylated GA, using a different HPLC system and elution conditions. Abbreviations: not detected (nd), female (F), male (M).

168
T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
GA53 occurred early, the so-called early 3b, 13-hydro-
xylation pathway (Sponsel, 1995; MacMillan, 1997).
Although the possibility of the pathway has been
demonstrated in Pisum sativum seedlings using in vivo
radiolabelled feeding studies (Durley et al., 1974), none
of the 20-carbon GAs of the pathway have yet been
reported in this species. Based on the results of radi-
olabelled feeding studies and bioassays (Fukui et al.,
1972), Durley et al. (1974) proposed that the following
conversions occur in the putative early 3b,13-dihydroxy-
lation pathway. Successive oxidation of C-20 in GA18
gives initially the d-lactone GA38 and subsequently the
aldehyde GA23. Further oxidation of C-20 in GA23
leads to the formation of the g-lactone GA1 or the tri-
carboxylic acid GA28 (see Fig. 1). To the present
authors’ knowledge R. acetosa is the first species in
which three successive 20-carbon GAs—GA18, GA38
and GA23—of this potential pathway have been identi-
fied together, suggesting that the pathway may be
operational in R. acetosa.
Although the metabolism of GAs has been researched
for decades, it is evident that research on certain aspects
of GA biosynthesis is still incomplete. In his review
paper MacMillan (1997) showed that some of the stages
of the putative early 3b,13-hydroxylation pathway await
confirmation by metabolic studies. As the only species
in which three consecutive GAs of the pathway have
been endogenously identified by GC–MS, R. acetosa
may be a useful system in which to investigate the
metabolic conversions involved in this pathway. In
Fig. 1. The putative early 3b,13-hydroxylation pathway. The pathway illustrates the conversions from GA18 to GA1 and its inactive metabolite
GA8. The GA conversions are putatively catalysed by the enzymes 20-oxidase (20-ox) and 2b-hydroxylase (2b-OH).

T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
169
other species such as Cucurbita maxima, although GA38
and GA23 have been identified endogenously (Lange et
al., 1993), the lackof identification of GA may be
18
because this GA is present at levels too low for current
detection techniques. However, it is equally likely that
the putative early 3b,13-hydroxylation pathway is not
operational in this species. GA38 and GA23 may be
synthesized by 3b-hydroxylation of the 13-hydroxylated
GA44 and GA19 respectively (MacMillan, 1997).
The biological activities of the C20-GAs of the puta-
tive early 3b,13-hydroxylation pathway have not been
previously investigated in flowering time studies. Fig. 2A
shows the typical effects of a range of 3b,13-dihy-
droxylated GAs and a GA-biosynthesis inhibitor on the
time to visible emergence of the inflorescence shoot for
male clones <A. GA18 and GA23 had no detectable
effects on the time to emergence of the inflorescence
shoot in treated compared to the control plants. GA38
had a slight promotive effect, giving a mean time to
inflorescence shoot emergence of 16 days compared to
20 days in the control plants. GA1 strongly promoted
early emergence by approximately one weekcompared
to control plants. In contrast, paclobutrazol treatment
retarded the appearance of the inflorescence shoot by an
average of 2 weeks compared to control plants.
A similar trend was observed using female clones ,B
(Fig. 2B). Although the effect was not statistically sig-
nificant, inflorescence shoots were observed approxi-
mately one weeklater in GA 23-treated plants compared
to controls. GA18 once again had no observed effect on
the time to emergence of the inflorescence shoot, while
GA38 slightly promoted development of the shoot by
approximately 4 days. Treatment with GA1 brought
forward the emergence of the inflorescence shoot by
approximately 13 days compared to control plants. In
contrast, treatment with paclobutrazol retarded the
development of the shoot by approximately 25 days
compared to control plants.
In studies on other vegetative clones (<C and ,D)
GA38 had a slight promotive effect on inflorescence
emergence in the males and inflorescences were detected
approximately 2 days earlier in treated compared to
control plants. However it had a negligible effect on
females (Fig. 2C and D). Again GA1 showed the great-
est biological activity, advancing inflorescence emer-
gence by 6 days in male plants and by 3 days in females
compared to control plants.
Fig. 2. Time to emergence of the primary inflorescence of males (A
and C) and females (B and D) treated with GAs or the GA-biosynth-
esis inhibitor paclobutrazol (Paclo). Each graph shows data from
vegetative clones of a single individual. Plants were vegetatively pro-
pagated in September and overwintered outdoors. They were moved
into an unheated greenhouse in February the following year. Plants
were treated with approximately 6 ml of a 10 mM solution of a single
GA (GA18, GA38, GA23 or GA1) applied as a spray with 0.01%
Tween, or 50 ml of 0.1 mM paclobutrazol applied as a soil drench.
Control plants were treated with 0.01% Tween. After treatment plants
were observed daily until a bulbous inflorescence shoot was detected
with the unaided eye. Data for each treatment or control show the
mean (ꢀSE) of four males and four females, except for the GA38
-
treated female plants (B) where the mean of three plants is shown. One
diseased plant was excluded; it eventually flowered 51 days after
treatment.
When data were pooled, irrespective of the clonal
populations or the sex of the clones, GA1 advanced the
time to emergence of the inflorescence shoot by an
average of 9 days and this was highly significant
(Table 2) (P<0.0001). A promotive effect on inflores-
cence emergence was also detected in GA38-treated
plants compared to control plants (P<0.05). When the
data for male and female plants are pooled, the retar-
dation by paclobutrazol of approximately 19 days com-
pared to control plants is also significant (P<0.05).
Elongation of the inflorescence stem was considerably
inhibited in paclobutrazol-treated plants (data not
included) and flowering was always positively associated
with stem elongation. Similarly in GA1- and GA38-
treated plants there was no indication in any plants of
stem growth without inflorescence development.

170
T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
Table 2
Effects of exogenously applied GAs on days to inflorescence shoot
emergence following treatment
position in the early 3b,13-dihydroxylation pathway
(Fig. 1) it was not unexpected that GA38 was more
active than GA18 in R. acetosa. However, the lackof
activity of GA23 is surprising. If the activity of GA38 is
due to its eventual conversion to GA1 it must be first
metabolised to GA23. This would suggest that GA23
would be at least as active as GA38. However, it is pos-
sible that in these studies the uptake and translocation
of applied GA38 was more effective than GA23. When
deuterated or radiolabelled GA18, GA38 and GA23
become available, feeding with these labelled GAs
should reveal whether conversion to labelled GA1
occurs in R. acetosa tissues and whether the putative
early 3b,13-hydroxylation pathway occurs in this
species.
Treatment
Days to visible
inflorescence initiation
GA1
Control
GA38
19.0ꢀ1.1
28.2ꢀ1.6
22.2ꢀ1.1
26.1ꢀ1.5
46.4ꢀ7.4
27.3ꢀ3.4
Control
Paclobutrazol
Control
Data shown are the means (ꢀSE) of plants clonally propagated from
several individuals and are irrespective of the sex of the plants. GA1
significantly advanced flowering in plants compared to control plants
(P<0.0001, n=23), as did GA38 (P<0.05, n=17) while paclobutrazol
significantly retarded flowering (P<0.05, n=8). Samples were ana-
lysed using two-tailed student t-tests.
There is evidence that some GAs d-lactones can be
oxidised in vegetative tissues (Gilmour et al., 1986;
Kobayashi et al., 1996). For example a radiolabelled
feeding study has demonstrated that the d-lactone GA15
may be oxidized at C-20 in maize seedlings to yield a
range of products including 20-carbon and 19-carbon
GAs of the early 13-hydroxylation pathway (Davis et
al., 1999). However, it has been previously suggested
(Reeve and Crozier, 1974; Kamiya et al., 1991) that
GA38 may have activity per se because the d-lactone
ring in the molecule has functional similarity to the g-
lactone ring of GA1. If this is so, it is interesting as it
indicates intrinsic biological activity in a C20-GA. The
findings of early studies which suggested possible activ-
ity of C20-GAs have been neglected, perhaps due to the
difficulty in obtaining relevant radiolabelled GAs.
However, as this study has also indicated, this is an area
which warrants further investigation.
Treatment with paclobutrazol retarded inflorescence
emergence, tentatively indicating a requirement for a
threshold level of GAs for the normal development
of the inflorescence shoot. GA1 is primarily impli-
cated in stem elongation. It is not known whether
the early emergence of the inflorescences in plants
treated with GA1, ‘emergence’ being recorded as the
time at which the inflorescence shoot was percep-
tible to the unaided eye, is due to the growth-promot-
ing activities of GA1 on the inflorescence shoot, or is
due to GA1 acting as a florigenic signal and promoting
the vegetative to reproductive phase transition. These
considerations may also apply to GA38 since the condi-
tions under which the experiments were carried out were
conducive to the eventual flowering of all plants, male
and female.
Of the C20-GAs of the putative early 3b,13-dihydroxy-
lation pathway, only GA38 showed any evidence of
activity in advancing the emergence of the inflor-
escences. A survey of the relative activities of GAs in a
range of bioassays found that the activities of the early
3b,13-hydroxylated GAs depended on the plant system
in which they were assayed, but GA38 showed some
activity in four of the five bioassays investigated (Reeve
and Crozier, 1974). For example, in the cucumber
hypocotyl bioassay, GA38 was moderately active and
GA18 and GA23 were inactive, while in the dwarf pea
assay, GA38 had high activity, with GA18 and GA23
proving only moderately active. Biochemical studies on
the chemical groups that confer activity found that the
presence of a C-19–C-20 d-lactone bridge (as found in
GA38) or aldehyde group (as for GA23) resulted in GAs
with higher activity than those possessing a C-20 methyl
group (as found in GA18) (Hoad, 1983). However, it
was not known whether the activity of these GAs was
intrinsic, or due to their conversion to other biologically
active GAs (Kamiya et al., 1991). Bearing in mind its
3. Experimental
3.1. General experimental procedures
Inorganic chemicals were from Sigma Chemical Co.,
Poole, UK, unless stated otherwise. The solvents used
were of the highest purity: HPLC grade from BDH
Chemicals Ltd., Poole, UK. Water used in all solutions
was ‘polished water’ from an Elga system. Infrared
spectra were recorded on a Perkin-Elmer 1800 Fourier
Transform Infrared spectrometer. 1H and 13C NMR
spectra were recorded on a Varian Gemini 300 instru-
ment. Chemical shifts are reported as values in parts per
million. For proton spectra recorded in chloroform, the
residual peakof CHCl was used as the internal refer-
3
ence (7.26 ppm) for H NMR spectra, while the central
1
peakof CDCl 3 (77.0 ppm) was used as the reference for
13C spectra. EI mass spectra (EIMS, 70 eV) were recor-
ded on a VG Fisons Autospec M mass spectrometer.
Flash chromatography was carried out on a Merck
Kieselgel 60.

T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
171
3.2. GA analysis
cartridge and the eluate passed to waste. Lastly, 10 ml
of 80% of methanol was used to elute the bound frac-
tion and this was reduced under vacuum to near dry-
ness. The sample was resuspended in 1 ml of water. The
sample was purified further by shaking it in a column
containing a slurry of polyvinyl polypyrrolidone on an
orbital shaker (R100 Luckham, Sussex, UK) at 1 Hz for
30 min and centrifuged at 900 ꢃ g for 15 min. The
supernatant was reduced under vacuum and resus-
pended in 1 ml of water.
The GAs were separated using reverse-phase HPLC
using a 150 mm ꢃ 4.6 mm, 5 mm octadecylsilica column
(Anachem, Rainin Instruments, Massachusetts, USA)
with a SP8700 HPLC solvent delivery system (Spectra
Physics, St. Albans, UK) and a SP4100 computing
integrator (Spectra Physics, St. Albans, UK). Authentic
GAs were detected using a LC 871 UV–Visible detector
(Pye Unicam Ltd., Cambridge, UK). The mobile phase
used in the separation was based on a gradient that
allows detection of authentic GA standards at UV
wavelengths at 206 nm (Barendse, 1980). The gradient
ran from 18% (v/v) methanol/2 mM acetic acid to 60%
(v/v) methanol/2 mM acetic acid in 20 min, then iso-
cratically at 60% (v/v) methanol/2 mM acetic acid for
15 min. The column was then washed by running 100%
methanol for 15 min. The flow rate was set to deliver
1 ml/min and the fraction collector was set to collect
fractions at 1 min intervals.
The plants used for analysis were F1 generation pro-
geny of a single cross of native R. acetosa plants col-
lected from coastal sites in Cornwall, UK, by Dr.
Anthony Lowe, University of Cambridge. The F1 gen-
eration plants were grown in the Cambridge University
Botanic Garden in outdoor experimental plots. The
plants were sexed by Dr. Anthony Lowe using cytoge-
netic studies and subsequently the sex of plants was
confirmed by morphological observation of sex organs
in flowers. Analysis was carried out on young inflor-
escences approximately 2–2.5 cm in length. Inflores-
cence stem tissues were not included in the analysis. At
the time of sampling the majority of flowers that com-
posed the male inflorescences were pre-anthesis and the
majority of flowers that composed the female inflor-
escences were not yet open to reveal the styles.
The inflorescences were snap frozen in liquid nitrogen
and ground to a powder in liquiꢂd nitrogen. They were
freeze dried and stored at ꢁ20 C until analysis. For
each sample 0.5 g DW of tissues was used for analysis.
The protocols used for extraction and purification of
GAs, and derivatisation of GAs for analysis by GC–MS
are described in Croker et al. (1990), except that
methylation of samples was carried out prior to reverse-
phase HPLC. The composition of the gradient used for
HPLC was as previously published, except that 2 mM
acetic acid was omitted because the GAs were methy-
lated. GAs were analysed as methyl ester trimethylsilyl
ethers by full-scan GC–MS using a Finnigan GCQ mass
spectrometer, as described in Coles et al. (1999). GAs
were identified by comparison of sample KRI and mass
spectra with those of authentic standards or library
spectra (Hedden, 1986).
The fractions which corresponded to the retention
time of authentic GA23, resulting from the separation of
several inflorescence samples were pooled and the com-
bined fractions were taken to dryness. The sample was
taken up in 2 ml of methanol and methylated, tri-
methylsilylated and analysed using GC–MS according
to protocols in Croker et al. (1990) and Coles et al.
(1999). GA23 was confirmed with MS/MS using 552 as
the parent ion.
Although GA23 was tentatively identified by initial
qualitative GC–MS following this procedure, the spec-
tra were contaminated. Therefore a different protocol,
which is outlined below, was followed for the extraction
and purification of samples and this enabled confirma-
tion of the presence of GA23.
3.3. Synthesis of GA18, GA38 and GA23
GA23 methyl ester has been prepared from gibberellic
acid (GA3) via the 3,13-bis(tetrahydropyranyl) ether
(‘‘bis THP’’) (Jiang and Pan, 1989) following proce-
dures reported earlier for the synthesis of GA36 (Dawe
et al., 1985). A repeat of the GA23-Me synthesis, fol-
lowed by alkaline hydrolysis of the bis THP–GA23 ester,
then removal of the THP ether protecting groups affor-
ded the parent GA23, in fair yield. Because of equili-
brium between the cyclic lactol and aldehyde forms,
GA23 affords poor NMR spectra and was therefore
additionally characterised as its dimethyl ester. GA38
was prepared by sodium borohydride reduction of GA23
(Koshimizu et al., 1968a), while Wolff-Kishner reduc-
tion (Mander et al., 1996) of the bis THP ether of GA23
followed by hydrolysis of the THP ether functions
furnished GA18.
GA23 was analysed in samples which had been
extracted and purified using the following protocols.
The young inflorescences from several plants were snap
frozen in liquid nitrogen. Each sample (1–3 g FW) was
ground to a fine powder using liquid nitrogen. The
samples were allowed to extract for 2 h in 30 ml of 80%
methanol (v/v) and the extract was centrifuged at 900 ꢃ
ꢂ
g, 6 C, for 10 min. The supernatant was reduced to
near dryness (approximately 500 ml) by rotary evapor-
ation.The sample was resuspended in approximately 9.5
ml of water and centrifuged as before for 5 min.
The extract was then partially purified on C-18 Sep-
Pakcartridges, which were activated by passing 10 ml of
100% methanol through them and washed with 10 ml of
2 mM acetic acid. The sample was passed through the

172
T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
3.3.1. ent-3ꢀ,13,20,20-Tetrahydroxy-gibberell-16-ene-
7,19-dioic acid 19,20-lactone (GA23)
(hexane/EtOAc/HOAc, 1:2:0.05) afforded a mixture of
16-ene (major) and 15-ene (minor) acids (46 mg, 58%
yield). Dowex 50W-X2 resin (H+ form) (16 mg) was
added to a solution of the crude GA18 bis THP ether (46
mg, 0.20 mmol) in methanol (2.5ml) and water (0.50
A stirred solution of bis THP dimethyl ester (88 mg) in
MeOH (1.0 ml) was treated with 2M NaOH (5.0 ml) and
ꢂ
heated (bath temperature 125 C) for 15 h. Acidification
ꢂ
and extraction into EtOAc, fractionation and chromato-
graphy (EtOAc/hexane/AcOH, 2:1:01) on silica gel affor-
ded the corresponding dicarboxylic acid (58 mg) and 7-
methy ester (12 mg). Dowex resin (19 mg, H+ form) was
added to a solution of the diacid in MeOH (2.5 ml) and
water (0.5 ml), and heated under reflux for 40 min.
Extraction into EtOAc followed by drying over Na2SO4
then chromatography over silica gel (EtOAc/MeOH/
AcOꢂH, 6:0.1:0.05) afforded GA23 (25 mg) m.p. 184–
ml). The reaction was then heated to 60 C for 1 h 20
ꢂ
min, then at 70 C for 20 min. The cooled solution was
diluted with EtOAc (20 ml) and filtered through Celite,
NaOAc added (to pH 6). The MeOH was removed
under a vacuum. Then the residue acidified with H3PO4
(10%, 1 ml) and diluted with ice cold EtOAc/20% 2-
butanol (20 ml). The layers were separated and the
aqueous phase was extracted with EtOAc/20% 2-buta-
nol (2ꢃ10 ml). The combined organic phases were
washed to pH=4 with ice cold brine. The combined
organic phases were dried over sodium sulfate, filtered
and the solvent removed in vacuo. Chromatography on
silica gel (hexane/EtOAc/AcOH, 1:1:0.05–3:1:0.05)
yielded GA18 (24 mg) as white crystals, m.p. 239–242 ꢂC
[lit. 240–242 ꢂC (Koshimizu et al., 1966)]. IR 3400–3100,
ꢂ
190 C [lit. 185–189 C (Koshimizu et al., 1968a)]. IR
1
3400–3100, 2500 (br), 1720, 1703, 1673 cmꢁ1. H NMR
(300 MHz, CD3OD): ꢁ 0.97 (3H, s, H18), 2.02 (1H, d,
partly obscured, H5), 2.02 (1H, d, J=10.8Hz), 3.45 (1H,
br s, H3), 4.63 (1H, d, J=3.0 Hz, H17), 4.69 (2H, OH),
4.91 (1H, d, J=3.0 Hz, H017), 5.38 (very br envelope,
1
H20). Dimethyl ester: H NMR (300 MHz, CDCl3 ): ꢁ
2500 (br), 1703, 1673 cmꢁ1 1H NMR (300 MHz,
.
1.21 (3H, s, H18), 2.76 (1H, d, J=12.6Hz, H5), 3.66 (3H,
s, OMe), 3.74 (3H, s, OMe), 3.88 (1H, d, J=12.6Hz, H6),
4.11 (1H, br s, H3), 4.93 (1H, d, J=3.0 Hz, H17), 5.18
(1H, d, J=3.0 Hz, H017), 9.70 (1H, s, H20).
CDCl3-CD3OD, 4:1) ꢁ 0.77 (3H, s, H20), 1.18 (3H, s,
H18), 0.80–2.30 (18H, m), 2.19 (1H, d, J=12.6 Hz, H5),
3.32 (1H, d, J=12.6 Hz, H6), 3.98 (1H, br s, H3), 4.84
(1H, br s, H17), 5.04 (1H, br s, H017). 13C NMR (75
MHz, CD3OD) ꢁ 14.2 (C20), 18.1 (C11), 23.3 (C18),
26.6 (C1), 33.3 (C3), 38.8 (C12), 42.9 (C10), 43.8, 44.5,
47.7, 48.1 (C4, C8, C14, C15), 47), 50.1 (C6), 50.3 (C5),
55.9 (C9), 70.6 (C3), 78.2 (C13), 105.7 (C17), 155.6
(C16), 178.2, 179.8 (–CO2H).
3.3.2. ent-3ꢀ,13,20-Trihydroxy-gibberell-16-ene-7,19-
dioic Acid 19,20-Lactone (GA38)
GA38 was prepared by NaBH4 reduction of GA23
according to the procedure of Koshimizu et al. (1968a),
affording the desired product, m.p. 240–242 ꢂC (lit. 237–
ꢂ
1
239 C) in 35% yield. H NMR (300 MHz, CDCl3): ꢁ
1.12 (3H, s, H18), 2.60, 2.61 (2ꢃ1H, 2ꢃABd, J=12.9
Hz, H6, H5), 3.59 (1H, br s, H3), 4.02 (1H, d,
Jgem=12.3 Hz, 20-pro-S-H), 4.36 (1H, d, Jgem=12.3 Hz,
20-pro-R-H), 4.78 (1H, s, H17), 5.08 (1H, d, J=2.4 Hz,
H017).
3.4. Bioassays
To minimise natural variations in flowering times
plants used for each of the flowering experiments were
genetically identical individuals. Plants that were vege-
tatively propagated from the same individual and
therefore were genetically identical are referred to as
vegetative clones. The following notation is used to refer
to plants: clones of a male individual ‘A’ are referred to
as <A and clones raised from a female individual ‘B’ are
referred to as ,B and so on.
In October, two mature male (A and C) and two
mature female plants (B and D), that were F1 genera-
tion siblings, were divided longitudinally through the
base of the plant into sections which comprised roots,
leaves and apical tissues. These were potted individually
in John Innes No. 1 (composition: 60% peat and 40%
loam). The plants were kept in an unheated greenhouse
where they were watered every day until new leaves
were produced. The plants were then moved outdoors
to experimental plots to undergo vernalization during
the winter months.
3.3.3. ent-3ꢀ,13-Dihydroxygibberell-16-ene-7,19-dioic
acid) (GA18)
Hydrazine hydrate (0.48 ml) was added to a stirred
solution of GA23 bis THP ether (79 mg) in ethanediol (4
ꢂ
ml) and the reaction mixture heated at 120 C for 1 h
50 min. Sodium hydroxide pellets (1.7 g) were added
ꢂ
and the temperature maintained at 120 C for 3 h, at
which time a precipitate began to form. After a further
ꢂ
30 min the temperature was raised to 175 C and the
ꢂ
reaction maintained at 175–180 C for 17 h. The cooled
mixture was diluted with EtOAc/20% 2-butanol (50 ml)
and was acidified to pH 2–3 with 2N HCl. The layers
were separated and the aqueous phase was extracted
with the EtOAc/2-butanol mixture (2ꢃ20 ml). The
combined organic phases were washed with brine until
the washings were pH 4. The organic phase was dried
over sodium sulfate, filtered and the solvent removed
under reduced pressure. Chromatography on silica gel
In February the following year the plants were moved
indoors to an unheated greenhouse. The plants were still
in the vegetative phase of development. Lighting was

T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
173
natural in the greenhouse and the plants experienced
lengthening photoperiods with the onset of spring. The
plants treated with a GA were sprayed on two occa-
sions, 1 weekapart. Plants received the first treatment
on the day they were moved indoors into the green-
house and they were treated with one of the following:
GA18, GA38, GA23 and GA1. Each plant received a
total of approximately 6 ml of 10 mM GA with 0.1 ml
lꢁ1 Tween, sprayed on the leaves. Control plants were
sprayed with an equal volume of 0.1 ml lꢁ1 Tween.
Plants treated with paclobutrazol received one treat-
ment on the day the plants were moved indoors. Paclo-
butrazol was applied as a soil drench at a concentration
of 0.1 mM with one application of 50 ml per plant.
Plants were monitored on a daily basis and the time
from the first treatment of plants to the emergence of the
first inflorescence shoot was recorded. Plants were con-
sidered to have developed inflorescences when it was evi-
dent from the slightly bulbous appearance of the young
shoot that the enveloping sheath-like tissue did not con-
tain only leaf primordia, but also enclosed the young
inflorescence (Stokes, 2001). In preliminary studies dis-
section of similar shoots confirmed that a developing
inflorescence was enclosed. However, in subsequent non-
destructive experiments the presumed young inflores-
cence were tagged for confirmation at a later stage of
development when the inflorescence bearing the flowers
was revealed. Each plant could bear several inflorescence
stems, but all experiments were carried out on the first
inflorescence developed. Data from the flowering experi-
ments were analysed using two-tailed student t-tests.
development in Arabidopsis by sense and antisense expression of
gibberellin 20-oxidase genes. The Plant Journal 17, 547–556.
Croker, S.J., Hedden, P., Lenton, J.R., Stoddart, J.L., 1990. Compar-
ison of gibberellins in normal and slender barley seedlings. Plant
Physiology 94, 194–200.
Davis, G., Kobayashi, M., Phinney, B.O., Lange, T., Croker, S.J.,
Gaskin, P., MacMillan, J., 1999. Gibberellin biosynthesis in maize.
Metabolic studies with GA15, GA24, GA25, GA7 and 2, 3-dehydro
GA9. Plant Physiology 121, 1037–1045.
Dawe, R.D., Mander, L.N., Turner, J.V., Pan, X., 1985. Synthesis of
C-20 Gibberellin A36 and A37 methyl esters from gibberellic acid.
Tetrahedron Lett. 26, 5725–5728.
Durley, R.C., Railton, I.D., Pharis, R.P., 1974. Conversion of gibber-
ellin A14 to other gibberellins in seedlings of dwarf Pisum sativum.
Phytochemistry 13, 547–551.
Evans, L.T., King, R.W., Chu, A., Mander, L.N., Pharis, R.P., 1990.
Gibberellin structure and florigenic activity in Lolium temulenteum,
a long-day plant. Planta 182, 97–106.
Fukui, H., Ishii, H., Koshimizu, K., Katsumi, M., Ogawa, Y., Mitsui,
T., 1972. The structure of gibberellin A23 and the biological prop-
erties of 3, 13-dihydroxy C20-gibberellins. Agricultural and Biologi-
cal Chemistry 36, 1103–1112.
Gilmour, S.J., Zeevaart, J.A.D., Schwenen, L., Graebe, J.E., 1986.
Gibberellin metabolism in cell-free extracts from spinach leaves in
relation to photoperiod. Plant Physiology 82, 190–195.
Hedden, P., 1986. The use of combined gas chromatography-mass
spectrometry in the analysis of plant growth substances. In:
Linskens, H.F., Jackson, J.F. (Eds.), Gas Chromatography/
Mass Spectrometry. Springer Verlag, Heidelberg, Germany, pp.
1–22.
Hiraga, K., Kawabe, S., Yokota, T., Murofushi, N., Takahashi, N.,
1974. Isolation and characterization of plant growth substances in
immature seeds and etiolated seedlings of Phaseolus vulgaris. Agri-
cultural and Biological Chemistry 38, 2521–2527.
Hisamatsu, T., Koshioka, M., Kubota, S., King, R.W., 1998. Effect of
gibberellin A4 and GA biosynthesis inhibitors on growth and flow-
ering of stock[ Matthiola incana]. Journal of the Japanese Society of
Horticultural Science 67, 537–543.
Hisamatsu, T., Koshioka, M., Kubota, S., Fujime, Y., King, R.W.,
Mander, L.N., 2000. The role of gibberellin biosynthesis in the
control of growth and flowering in Matthiola incana. Physiologia
Plantarum 109, 97–105.
Acknowledgements
T.S.S. is grateful to the BBSRC and Advanced Tech-
nologies, Cambridge for sponsorship of part of this
work. Dr. Anthony Lowe is gratefully acknowledged
for providing the plants for sampling and propagation.
Hoad, G.V., 1983. Gibberellin bioassays and structure-activity rela-
tionships. In: Crozier, A. (Ed.), The Biochemistry and Physiology of
Gibberellins, vol. 2. Praeger Publishers, New York, USA, pp. 57–
94.
Jiang, B., Pan, X., 1989. Studies on the synthesis of gibberellins—
stereoselective synthesis of GA23 methyl ester. Chin. Sci. Bull 34,
2020–2022.
References
Kamiya, Y., Kobayashi, M., Fujioka, S., Yamane, H., Nakayama, I.,
Sakurai, A., 1991. Effects of a plant growth regulator, Prohexadione
calcium (BX-112), on the elongation of rice shoots caused by exo-
genously applied gibberellins and helminthosporol, part II. Plant
and Cell Physiology 32, 1205–1210.
Ainsworth, C.C., Lu, J., Winfield, M., Parker, J., 1999. Sex determi-
nation by X:autosome dosage: Rumex acetosa (sorrel). In: Ains-
worth, C.C. (Ed.), Sex Determination in Plant. Bios Scientific
Publishers, Oxford, UK, pp. 121–136.
Kobayashi, M., Spray, C.R., Phinney, B.O., Gaskin, P., MacMillan,
J., 1996. Gibberellin metabolism in maize-the stepwise conversion of
gibberellin A12-aldehyde to gibberellin A20. Plant Physiology 110,
413–418.
Barendse, G.W.M., Van de Werken, P.H., Takahashi, N., 1980. High
performance liquid chromatography of gibberellins. Journal of
Chromatography 198, 449–455.
Bavrina, T.V., Eulafie, L., Chailakhyan, M.Kh., 1991. Influence of
daylength and phytohormones on flowering and sex expression in
dioecious plants of sheep sorrel (Rumex acetosella L.). Doklady
Akademii Nauk SSSR 317, 1510–1514.
Korpelainen, H., 1992. Patterns of resource allocation in male and
female plants of Rumex acetosa and R. acetosella. Oecologia 89,
133–139.
Koshimizu, K., Fukui, H., Inui, M., Ogawa, Y., Mitsui, T., 1968a.
Gibberellin A23 in immature seeds of Lupinus luteus. Tetrahedron
Lett. 1143–1147.
Blazquez, M.A., Green, R., Nilsson, O., Sussman, M.R., Weigel, D.,
1998. Gibberellins promote flowering of Arabidopsis by activating
the Leafy Promoter. The Plant Cell 10, 791–800.
Koshimizu, K., Fukui, H., Kusaki, T., Mitsui, T., Ogawa, Y., 1966. A
new gibberellin in immature seeds of Lupinus luteus. Tetrahedron
Lett. 2459–2463.
Coles, J.P., Phillips, A.L., Croker, S.J., Garcıa-Lepe, R., Lewis, M.J.,
Hedden, P., 1999. Modification of gibberellin production and plant

174
T.S. Stokes et al. / Phytochemistry 62 (2003) 165–174
Koshimizu, K., Fukui, H., Kusaki, T., Ogawa, Y., Mitsui, T., 1968b.
Isolation and structure of gibberellin A18 from immature seeds of
Lupinus luteus. Agricultural and Biological Chemistry 32, 1135–1140.
Lange, T., 1998. Molecular biology of gibberellin biosynthesis. Planta
204, 409–419.
Rijnders, J.G.H.M., Yang, Y.-Y., Kamiya, Y., Takahashi, N., Bare-
ndse, G.W.M., Blom, C.W.P.M., Voesenek, L.A.C.J., 1997. Ethyl-
ene enhances gibberellin levels and petiole sensitivity in flooding-
tolerant Rumex palustris but not in flooding-intolerant R. acetosa.
Planta 203, 20–25.
Lange, T., Hedden, P., Graebe, J.E., 1993. Gibberellin biosynthesis in
cell-free extracts from developing Cucurbita maxima embryos and the
identification of new endogenous gibberellins. Planta 189, 350–358.
Langridge, J., 1957. Effect of day-length and gibberellic acid on the
flowering of Arabidopsis. Nature 6, 36–37.
Sponsel, V.M., 1995. Gibberellin biosynthesis and metabolism. In:
Davies, P.J. (Ed.), Plant Hormones: Physiology, Biochemistry and
Molecular Biology. Kluwer Academic Publishers, Dordrecht, The
Netherlands, pp. 66–97.
Spray, C.R., Kobayashi, M., Suzuki, Y., Phinney, B.O., Gaskin,
P., MacMillan, J., 1996. The dwarf-1 (d1) mutant of Zea mays
blocks three steps in the gibberellin-biosynthetic pathway. Pro-
ceedings of the National Academy of Sciences of the USA 93,
10515–10518.
Lee, I.-J., Foster, K.R., Morgan, P.W., 1998. Effect of gibberellin
biosynthesis inhibitors on native gibberellin content, growth and
floral initiation in Sorghum bicolor. Journal of Plant Growth Reg-
ulation 17, 185–195.
MacMillan, J., 1997. Biosynthesis of the gibberellin plant hormones.
Natural Products Reports 14, 221–243.
Stokes, T. S., 2001. Gibberellins and Cytokinins in Rumex acetosa L.
PhD thesis, University of Cambridge, UK.
Mandel, R.M., Rood, S.B., Pharis, R.P., 1992. Bolting and floral
induction in annual and cold-requiring biennial Brassica spp.:
effects of photoperiod and exogenous gibberellin. In: Karssen,
C.M., Van Loon, L.C., Vreugdenhil, D. (Eds.), Progress in Plant
Growth Regulation. Kluwer Academic Publishers, Dordrecht, The
Netherlands, pp. 371–379.
Sun, T.-p., Kamiya, Y., 1994. The Arabidopsis GA1 locus encodes the
cyclase ent-kaurene synthetase A of gibberellin biosynthesis. Plant
Cell 6, 1509–1518.
Talon, M., Zeevaart, J.A.D., 1990. Gibberellins and stem growth as
related to photoperiod in Silene armeria L. Plant Physiology 92,
1094–1100.
Mander, L.N., Owen, D.J., Twitchin, B., 1996. Synthesis of gibberellin
GA53 and its 17,17-d2-Derivative, Aust. J. Chem. 49, 249–253.
Michniewicz, M., Lang, A., 1962. Effect of nine different gibberellins
on stem elongation and flower formation in cold-requiring and
photoperiodic plants grown under non-inductive conditions. Planta
58, 549–563.
Van Roggen, P.M., Debenham, B., Hedden, P., Phillips, A.L., Tho-
mas, S.G., 1998. A model for control of bolting and flowering in
sugar beet and the involvement of gibberellins. Flowering News-
letter 25, 45–49.
Wilson, R.N., Heckman, J.W., Somerville, C.R., 1992. Gibberellin is
required for flowering in Arabidopsis thaliana under short days.
Plant Physiology 100, 403–408.
Reeve, D.R., Crozier, A., 1974. An assessment of gibberellin structure-
activity relationships. Journal of Experimental Botany 25, 431–445.
Reeves, P.H., Coupland, G., 2001. Analysis of flowering time control
in Arabidopsis by comparison of double and triple mutants. Plant
Physiology 126, 1085–1091.
Zeevaart, J.A.D., Talon, M., 1992. Gibberellin mutants in Arabidopsis
thaliana. In: Karssen, C.M., Van Loon, L.C., Vreugdenhil, D.
(Eds.), Progress in Plant Growth Regulation. Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp. 34–42.