The ectopic overexpression of a citrus gibberellin 20-oxidase enhances the non-13-hydroxylation pathway of gibberellin biosynthesis and induces an extremely elongated phenotype in tobacco

Transgenic plants of Nicotiana tabacum overexpressing a gibberellin (GA) 20-oxidase cDNA ( CcGA20ox1 ) from citrus, under the control of the 35S promoter, were taller (up to twice) and had larger inflorescences and longer flower peduncles than those of control plants. Hypocotyls of transgenic seedlings were also longer (up to 4 times), and neither the seedlings nor the growing plants elongated further after application of GA₃. Hypocotyl and stem lengths were reduced by application of paclobutrazol, and this inhibition was reversed by exogenous GA₃. The ectopic overexpression of CcGA20ox1 enhanced the non-13-hydroxylation pathway of GA biosynthesis leading to GA₄, apparently at the expense of the early-13-hydroxylation pathway. The level of GA₄ (the active GA from the non-13-hydroxylation pathway) in the shoot of transgenic plants was 3–4 times higher than in control plants, whereas that of GA₁, formed via the early-13-hydroxylation pathway (the main GA biosynthesis pathway in tobacco), decreased or was not affected. GA₄ applied to the culture medium or to the expanding leaves was found to be at least equally active as GA₁ on stimulating hypocotyl and stem elongation of tobacco plants. The results suggest that the tall phenotype of tobacco transgenic plants was due to their higher content of GA₄, and that the GA response was saturated by the presence of the transgene.     

Effects of exogenous gibberellins and paclobutrazol on floral stalk growth of tulip sprouts isolated from cooled and non-cooled tulip bulbs

The involvement of gibberellins (GAs) in the regulation of floral stalk elongation and flower development has been studied in tulip. The biological activity of GA₄ and GA₉, both endogenous in tulip bulb sprouts, and GA₁, was tested in vitro on sprouts of cooled and non-cooled tulip bulbs ( Tulipa gesneriana L. cv. Apeldoorn), in the presence or absence of the GA biosynthesis inhibitor paclobutrazol. At early starting dates of incubation, floral stalks from both cooled and non-cooled bulbs hardly showed any elongation in the absence of exogenous GA. Paclobutrazol had no effect on floral stalk elongation, and the response to GAs of sprouts from cooled bulbs was greater than that of sprouts from non-cooled bulbs. At later starts of incubation, considerable floral stalk elongation occurred without GA application. Paclobutrazol inhibited this floral stalk elongation, and the growth of sprouts from both cooled and non-cooled bulbs was stimulated by GA application. The effect of paclobutrazol was reversed by simultaneous application of GA₄ or GA₉. Application of GA with and without paclobutrazol resulted in the same elongation of the floral stalk, indicating the absence of substantial side effects of the inhibitor. The isolated sprouts did not develop a full-grown flower without the addition of GA. GA₄ was more effective than GA₉ in stimulating this flower development. The results demonstrate that both sprouts from cooled and non-cooled bulbs are responsive to exogenous GAs in vitro, and may be a site of GA biosynthesis.     

Gibberellin substitution for long day secondary induction of flowering in Poa pratensis

Plants of Poa pratensis cv. Holt initiate inflorescence primordia when exposed to short days (SD) and low temperature, but require a secondary induction by at least 4 long days (LD) for further inflorescence development and stem elongation. Single or double applications of 10 µg per plant of gibberellins A₁, A₃, A₅ and 16,17‐dihydro A₅ (DHGA₅) induced inflorescence development in a high proportion of plants in SD, but only if the plants were detillered to a single stem. Exposure to 2 LD cycles did not cause heading and flowering alone but enhanced the effect of exogenous gibberellins (GAs), bringing flowering to 100%. GA₅ and DHGA₅ were less effective than GA₁ and GA₃ in SD, especially with double applications, but were more effective than GA₁ and GA₃ when given together with 2 LD. The GAs had differential effects on vegetative growth and flowering, GA₅ and DHGA₅ causing much less leaf and stem growth than the other two GAs. Marginal induction, whether by LD or GA application, resulted in a high proportion of spikelets with viviparous proliferation. Thus, whereas GAs are inhibitory to the primary induction by SD, they can replace secondary induction by LD when vegetative growth is limited.     

Expression of two gibberellin-regulated cDNAs during early flower development in tomato (Solanum lycopersicon). Effect of grafting and paclobutrazol

To study the role of translocation of gibberellin (GA) intermediates or bioactive GAs from other plant parts to buds during early flower development in tomato ( Solanum lycopersicon ), the effect of grafting and paclobutrazol (PAC) treatment on the expression of tgas100 and tgas118 , two GA-regulated mRNAs, was analysed. Both mRNAs accumulated in a dose-dependent fashion. Application of 0.5 ng GA₃ per bud to developmentally arrested flower buds of a GA-deficient mutant of tomato ( gib-1 ) induced tgas100 expression, while the tgas118 abundance increased. For obtaining normal flower development through anthesis in the mutant, a single GA₃ treatment was required of at least 5 ng GA₃ per bud. In wild-type flower buds, PAC decreased the abundance of tgas100 and tgas118 mRNAs either when PAC was sprayed on whole plants or directly applied to buds. When only the wild-type buds were treated with PAC, the expression profiles characteristic for untreated buds were not restored by translocation of endogenous GAs. Grafting of gib-1 scions onto wild-type donor plants did not result in normal flower development or expression profiles like in wild-type buds. We conclude that the role of GA transport in early flower development of tomato is negligible and that the GAs required for development have to be synthesized in the flower bud itself.     

Immunomodulation of gibberellin biosynthesis using an anti-precursor gibberellin antibody confers gibberellin-deficient phenotypes

Immunomodulation is a means to modulate an organism's function by antibody production to capture either endogenous or exogenous antigens. We have recently succeeded in obtaining gibberellin (GA)-deficient phenotypes in Arabidopsis thaliana by using anti-bioactive GA antibodies. In this study, a single-chain antibody (scFv) against GA(24), a precursor GA, was utilized to repress the biosynthesis of bioactive gibberellins. Stable accumulation of the scFv in endoplasmic reticulum (ER) was achieved by being produced as a fusion with GFP as well as KDEL ER-retention signal. The transgenic plants showed GFP fluorescence in the reticulate cortical ER network in epidermal cells. The GFP-scFv fusion produced in plants maintained its binding activity. The transgenic plants showed GA-deficient phenotypes, including reduced rosette leaf development, delayed flower induction and reduced stem elongation of the main culm, especially in the early stage of inflorescence growth. Contrarily, stem elongation of the main culm at a later stage, or that of lateral shoots was much less affected by scFv production. These phenotypes were different from anti-bioactive GA scFv-producing lines, whose stem elongation was continuously repressed throughout the inflorescence development. The GA-deficient phenotypes were recovered by treatment with GA(24) and bioactive GA(4), the latter being more effective. The transgenic lines contained conspicuously higher endogenous GA(24) and clearly less GA(4) than wild-type plants. The expression of GA 20-oxidase and GA 3-oxidase genes, which are feedback-regulated by GA signaling, were up-regulated in those plants. These results demonstrate that the scFv trapped GA(24) in ER and inhibited metabolism of GA(24) to bioactive GA(4).     

Biological activity, identification and quantification of gibberellins in seedlings of Norway spruce (Picea abies) grown under different photoperiods

Short photoperiod induces growth cessation in seedlings of Norway spruce ( Picea abies (L.] Karst.). Application of different gibberellins (GAS) to seedlings growing under a short photoperiod show that GA₉ and GA₂₀ can not induce growth. In contrast application of GA, and GA₄ induced shoot elongation. The results indicate that 3β-hydroxylation of GA₉ to GA₄ and of GA₂₀ to GA₁ is under photoperiodic control. To confirm that conclusion, both qualitative and quantitative analyses of endogenous GAs were performed. GA₁, GA₃, GA₄, GA₇, GA₉, GA₁₂, GA₁₅, GA₁₅, GA₂₀, GA₂₉, GA₃₄ and GA₅₁ were identified by combined gas chromatography-mass spectrometry in shoots of Norway spruce seedlings. The effect of photoperiod on GA levels was determined by using deuterated and ¹⁴C-labelled GAs as intermal standards. In short days, the amounts of GA₉, GA₄ and GA₁ are less than in plants grown in continuous light. There is no significant difference in the amounts of GA₃, GA₁₂, and GA₂₀ between the different photoperiods. The lack of accumulation of GA₉ and GA₂₀ under short days is discussed.     

Interaction of gibberellins and phytochrome in the control of cowpea epicotyl elongation

The physiological response of cowpea ( Vigna sinensis L.) epicotyl explants to far‐red light (FR) and its interaction with gibberellins (GAs) have been investigated. The effect of FR and GA₁ varied with the age of the seedlings from which the explants were made: for FR, it decreased progressively with age (though the sensitivity of the epicotyls to FR did not change significantly until at least day 11), whereas it remained essentially constant for applied GA₁ between days 5 and 9 after sowing. This indicates that the loss of response to FR may be due to a decrease in endogenous GA levels in the epicotyl. For a range of GA₁ and GA₂₀ (0.01–1 µg explant⁻¹), both hormones were more active in FR than in R irradiated epicotyls, suggesting that phytochrome may affect GA sensitivity besides GA metabolism. The location of the epicotyl region most sensitive to FR (between 5 and 20 mm below the apex) was different from that to GAs (the upper 10 mm). Nevertheless, FR extended the region responsive to applied GAs, even in paclobutrazol‐treated epicotyls where elongation was due entirely to exogenous GAs. This means that modulation of epicotyl elongation by phytochrome, that occurs in a zone different from though overlapping with the GA‐sensitive subapical zone, is also mediated by GAs. Growth in the most FR‐sensitive region of the epicotyl stimulated by FR or GA₁ was due to cell elongation, and in the most GA‐sensitive region to both cell division and elongation. The effect of FR and GA₁ was negated by colchicine, indicating that microtubules may be involved in the response to both factors.     

Internode length in Pisum. The lrs mutation reduces gibberellin response and levels

We describe a new mutation, lrs , which reduces internode length in Pisum sativum L. The mutation appears to act by reducing both GA synthesis and the response to GA₁. The levels of the 13‐hydroxylated GAs, GA₅₃, GA₄₄, GA₁₉, GA₂₀, GA₁, and GA₈ in the lrs mutant were greatly reduced compared with the wild‐type. The extent of the reduction in GA₁ content in the apical tissues would, at least in part, account for the dwarf phenotype of the mutant. The reduced GA responsiveness of the new mutant was indicated by the inability of applied GA₁ to remove the difference in elongation between lrs and LRS plants. The lrs mutant appears to be unique amongst internode length genotypes, possessing characteristics of both GA synthesis and GA response mutants.     

The plant-growth-promoting rhizobacteria Bacillus pumilus and Bacillus licheniformis produce high amounts of physiologically active gibberellins

The plant-growth-promoting rhizobacteria (PGPR), Bacillus pumilus and Bacillus licheniformis, isolated from the rhizosphere of alder ( Alnus glutinosa [L.] Gaertn.) have a strong growth-promoting activity. Bioassay data showed that the dwarf phenotype induced in alder seedlings by paclobutrazol (an inhibitor of gibberellin [GA] biosynthesis) was effectively reversed by applications of extracts from media incubated with both bacteria and also by exogenous GA₃. Full-scan gas chromatography-mass spectrometry analyses on extracts of these media showed the presence of GA₁, GA₃, GA₄and GA₂₀, in addition to the isomers 3- epi -GA₁ and iso -GA₃. Isotope dilution analysis indicated that epi -GA₁ was an artefact. Likewise, iso -GA₃ is also probably an artifact spontaneously formed during extraction and/or analysis. In both culture media, GA₁ was present in higher concentrations (130–150 ng ml⁻¹) than GA₃ (50–60 ng ml⁻¹), GA₄ (8–12 ng ml⁻¹) and GA₂₀ (2–3 ng ml⁻¹). The data indicated that culture of both bacteria accumulate bioactive C19-gibberellins in relative high amounts and that these GAs appear to be physiologically active in the host plant. The evidence suggests that the promotion of stem elongation induced by the PGPR could be mediated by bacterial GAs.     

Photoperiodic control of endogenous gibberellins in seedlings of Salix pentandra

In the temperate-zone woody species Salix pentandra elongation growth is regulated by the photoperiod. Long days sustain active growth, whereas short days induce cessation of apical growth, which is a prerequisite for winter hardening. It is shown that this is correlated to quantitative changes in levels of endogenous GA₁₉ GA₂₀, and GA₁. Within two short days the amount of the active GA₁ and its immediate precursor GA₂₀, decreased markedly in young leaves us well as in stem tissue. Also, the amount of GA¹⁹, declined, but the decrease was delayed relative to that of GA₁ and GA₂₀. The ability of S. pentandra seedlings to respond to exogenous GA₁₉, decreased with increasing numbers of short days. Observations that support the hypothesis that the level of GA₁ in S. pentandra is regulated by the photoperiod in a quantitative mode with conversion of GA₁₉, to GA₂₀, being one target for control.
Different distribution of GAs in various plant parts was observed. The level of GA was higher in young leaves than in other plant parts, and the amount of GA₁₉ was 5–10 times higher in stem tissue than in leaves and roots. The ratios of GA₈ to GA¹ and GA₂₀, were higher in roots as compared with other parts, as rods contained very low levels of GA₁ and GA₂₀, but amounts of GA₂₀ comparable with other parts.     

Internode length in Lathyrus odoratus. The involvement of gibberellins

Evidence was obtained by gas chromatography-mass spectrometry and gas chromatography-selected ion monitoring for the presence of gibberellin A₂₀), GA₁, GA₂₉, GA₈ and 2-epiGA₂₉ in vegetative shoots of tall sweet pea, Lathyrus odoratus L. Both tall (genotype L –) and dwarf (genotype II ) sweet peas elongated markedly in response to exogenous GA₁ attaining similar internode lengths at the highest dose levels. Likewise internode length in both genotypes was reduced by application of the GA biosynthesis inhibitor, PP333. The ratio of leaflet length to width was reduced by application of PP333 to tall plants and this effect was reversed by GA₁. When applied to plants previously treated with PP333, GA₂₀ promoted internode elongation of L – plants as effectively as GA₁, but GA₂₉ was not as effective as GA₁ when applied to II plants. In contrast, GA₂₀ and GA₁ were equally effective when applied to the semidwarf lb mutant but GA-treated lblb plants did not attain the same internode length as comparable GA-treated Lb – plants. The difference in stature between the tall and dwarf types persisted in dark-grown plants. It is concluded that GA₁ may be important for internode elongation and leaf growth in sweet pea. Mutant l may influence GA₁ synthesis by reducing 3β-hydroxylation of GA₂₀ whereas mutant lb appears to affect GA sensitivity.     

The end-of-day far-red irradiation increases gibberellin A1 content in cowpea (Vigna sinensis) epicotyls by reducing its inactivation

It has been shown previously that gibberellins (GAs) mediate the phytochrome (Phy) control of cowpea ( Vigna sinensis L.) epicotyl elongation induced by end-of-day (EOD)-far-red light (FR). In the present work, the EOD-FR effect on GA metabolism and GA levels in cowpea has been investigated. GA₁, GA₈, GA₁₉ and GA₂₀ were identified in epicotyls, and GA₁, GA₁₉, GA₂₀ and GA₂₉-catabolite in leaves of 6-day-old cowpea seedlings. The content of GA₁ in the epicotyl paralleled the decrease of its growth rate, supporting the hypothesis that this is the GA bioactive in controlling cowpea epicotyl elongation. FR enhanced both the amount of [3H]GA₁ in the epicotyl produced from applied [3H]GA₂₀, and that of applied [3H]GA₁ that remained unmetabolized in epicotyl explants, suggesting that Phy may regulate the inactivation of GA₁. In agreement with this effect of light on GA₁ metabolism, the contents of GA₁ in the epicotyl remained higher in FR-treated than in R-treated explants. Moreover, in intact seedlings EOD-FR treatment increased both epicotyl elongation and GA₁ content in the responsive epicotyl, whereas it was not altered in the leaves. These results show, for the first time, that photostable Phys modulate the stem elongation in light-grown plants by locally controlling the GA₁ levels through regulation of its inactivation.     

Gibberellins in relation to flowering in Polianthes tuberosa

Endogenous gibberellins (GAs) in corms of Polianthes tuberosa L. (cv. Double) were isolated and identified by high performance liquid chromatography, bioassay and combined capillary gas chromatography-mass spectrometry (GC-MS). Gibberellins A₁, A₁₉, A₂₀ and A₅₃ were quantified at the vegetative, early floral initiation and flower development stages. The identification of 13-hydroxylated GAs indicates the presence of the early 13-hydroxylation pathway in P. tuberosa corms. An increase in GA₁ and GA₂₀, and a decrease in GA₁₉ levels, coincided with the transition from the vegetative phase to the stages of early floral initiation and flower development. GA₅₃ stayed at constant levels at the 3 different growth stages. The absence of GA₁ in vegetative corms and its presence in corms at early floral initiation and flower development stages suggest that GA₁ is a causal factor in inducing floral initiation in P. tuberosa . When GA₁, GA₃, GA₄, GA₂₀ and GA₃₂ were applied to corms at the vegetative stage (plants about 5 cm in height), floral initiation was promoted by all of the GAs used, GA₃₂ being the most active. In contrast with the other GAs, GA₃₂ had no effect on stem elongation. Therefore, it is suggested that hydroxylated C-19 GAs play an important role in flower induction in P. tuberosa .     

Detection and identification of gibberellins in Sitka spruce (Picea sitchensis) of different ages and coning ability by bioassay, radioimmunoassay and gas chromatography – mass spectrometry

Gibberellins Al (GA₁), GA₃, GA₄, GA₉, and after enzymatic hydrolysis of GA-conjugate-like fractions, GA₉ and GA₁₅, were identified in shoots of Sitka spruce [ Picea sitchensis (Bong.) Carr.] of different ages by combined gas chromatography-mass spectrometry (GC-MS). The purification and separation of the GAs involved the use of reverse phase and normal phase high performance liquid chromatography (HPLC). The Tan-ginbozu dwarf rice bioassay and binding to antibodies raised against GA₁, GA₄ and GA₉ were used for detection of GA-like substances. The qualitative differences between the three ages of plant material were the presence of GA₃ and GA₁ in the 48-year-old material and the absence of detectable amounts of GA₄ in the same material. This indicates a difference in GA metabolism which may reflect the difference in ability to form reproductive buds.     

Gravistimulation leads to asymmetry of both auxin and gibberellin levels in barley pulvini

The auxin indole-3-acetic acid (IAA) is known to promote the biosynthesis of active gibberellins (GAs) in barley ( Hordeum vulgare ). We therefore investigated the possibility that this interaction might contribute to the gravitropic response of barley leaf sheath pulvini. Barley plants at the inflorescence stage were gravistimulated for varying times, and the pulvini were then separated into upper and lower halves for quantification of IAA and GAs by GC-MS. Consistent with the Cholodny–Went theory, the lower portion contained more IAA than did the upper portion. This difference was detected as early as 2.5 h after the start of gravistimulation, and bending was also observed at this stage. At later time points tested (6 h and 24 h), but not at 2.5 h or 3 h, the higher auxin content of the lower half was associated with a higher level of GA₁, the main bioactive GA in barley. Consistent with that result, the expression of Hv3ox2 , which encodes a key enzyme for the conversion of GA₂₀ to GA₁, was higher in the lower side than in the upper, after 6 h. It is suggested that in gravistimulated leaf sheath pulvini, auxin accumulates in the lower side, leading to a higher level of GA₁, which contributes to the bending response. Further evidence that GAs play a role in the gravitropic response was obtained from GA-related mutants, including the elongated sln1c mutant, in which GA signalling is constitutive. Pulvinar bending in the sln1c mutant was greater than in the wild-type. This result indicates that in the lower side of the gravistimulated pulvinus, the relatively high level of bioactive GA facilitates, but does not mediate, the bending response.     

Role of gibberellins in parthenocarpic fruit development induced by the genetic system pat-3/pat-4 in tomato

The role of gibberellins (GAs) in the induction of parthenocarpic fruit-set and growth by the pat-3/pat-4 genetic system in tomato ( Lycopersicon esculentum Mill.) was investigated using wild type (WT; Cuarenteno) and a near-isogenic line derived from the German line RP75/59 (the source of pat-3/pat-4 parthenocarpy). Unpollinated WT ovaries degenerated but GA₃ application induced parthenocarpic fruit growth. On the contrary, parthenocarpic growth of pat-3/pat-4 fruits, which occurs in the absence of pollination and hormone treatment, was not affected by applied GA₃. Unpollinated pat-3/pat-4 fruit growth was negated by paclobutrazol, an inhibitor of ent -kaurene oxidase, and this inhibitory effect was negated by GA₃. The quantification of the main GAs of the early 13-hydroxylation pathway (GA₁, GA₈, GA₁₉, GA₂₀, GA₂₉ and GA₄₄) in unpollinated ovaries at 3 developmental stages (flower bud, FB; pre-anthesis, PR; and anthesis, AN), by gas chromatography-selected ion monitoring, showed that the concentration of most of them was higher in pat-3/pat-4 than in WT ovaries at PR and AN stages. The concentration of GA₁, suggested previously to be the active GA in tomate, was 2–4 times higher. Unpollinated pat-3/pat-4 ovaries at FB, PR and AN stages also contained relatively high amounts (5–12 ng g⁻¹) of GA₃, a GA found at less than 0.5 ng g⁻¹ in WT ovaries. It is concluded that the mutations pat-3/pat-4 may induce natural facultative parthenocarpy capacity in tomato by increasing the concentration of GA₁ and GA₃ in the ovaries before pollination.     

Growth and development of Brassica genotypes differing in endogenous gibberellin content. II. Gibberellin content, growth analyses and cell size

Three rapid cycling Brassica rapa genotypes were grown in greenhouse conditions to investigate the possible relationships between endogenous gibberellin (GA) content and shoot growth. Endogenous GA₁ GA₃ and GA₂₀ were extracted from stem samples harvested at 3 weekly intervals and analyzed by gas chromatography-mass spectrometry with selected ion monitoring, using [²H₂]-GA₁ and [²H₂]-GA₂₀ as quantitative internal standards. During the first 2 weeks, GA levels of the dwarf, rosette ( ros ), averaged 36% of levels in normal plants (on a per stem basis). Levels in the tall mutant, elongated internode (ein) , were consistently higher, averaging 305% of levels in normal plants.
Differences in shoot height across the genotypes resulted from varying internode length which resulted from epidermal cell length and number being increased in ein and decreased in ros relative to the normal genotype. The exogenous application of GA₃ to normal plants increased cell length while the application of paclobutrazol (PP333), a triazole plant growth retardant, reduced cell size. Thus, exogenous GA manipulations mimicked the influence of the mutant genes ros and ein. The dwarf, ros , had reduced shoot dry weights and relative growth rates compared to the other genotypes. Total dry weights were similar in ein and the normal genotype but stem weights were increased in ein , compensating for decreased leaf weights. Thus, the gibberellin-deficiency of ros resulted in generally reduced shoot growth. The overproduction of endogenous GA by ein did not result in enhanced shoot growth but rather a specific enhancement of internode elongation and stem growth at the expense of leaf size.     

Germination of annual celery (Apium graveolens) seeds: Inhibition by paclobutrazol and its reversal by gibberellins and benzyladenine

Plants of annual celery ( Apium graveolens L.) were treated with paclobutrazol during anthesis. Seeds collected from the treated plants showed a marked reduction in germination in light and failed to germinate in the dark. Application of GA_4/7 to the imbibition solution reversed the inhibitory effect of paclobutrazol while gibberellic acid (GA₃) was ineffective. Benzyladenine (BA) interaction with GA_4/7 was light and concentration-dependent. At relatively low concentrations in the dark there was a synergistic effect, but at higher concentrations, especially in the light, BA, antagonized the GA_4/7 effect. Seedlings emerging from the seeds from paclobutrazol-treated plants were only slightly shortened. It is suggested that paclobutrazol applied to the mother plants inhibited the biosynthesis of endogenous GAs, which normally enable the germination of annual seeds under unfavorable conditions. Exogenously applied GA_4/7 fulfills the function of the absent endogenous GAs.     

Gibberellic acid decreases anthocyanin accumulation in wild carrot cell suspension cultures but does not alter 3'-nucleotidase activity

Gibberellic acid (GA₃) applied at different times during the growth of wild carrot ( Daucus carota ssp. Carota ) cell suspension cultures inhibited anthocyanin accumulation. Application of 3 × 10^–6 M GA₃ to cultures on day 0 or day 4 gave, respectively, 10 or 35% of anthocyanin accumulation relative to levels occurring when GA₃ was applied at the end of the growth period. Endogenous GAs were separated by high pressure liquid chromatography, and identified and quantified by gas chromatography-selected ion monitoring. Gibberellins GA₁, GA₃ and traces of GA₈. GA₁₉ and GA₂₀ were identified in carrot cell suspension cultures of both high and low anthocyanin-accumulating clones. The concentrations of GA₁. GA₃ and GA₈ in the two clones were similar and were not significantly different after the application of uniconazole which promoted anthocyanin accumulation. This suggests that these endogenous GAs are not the sole factors controlling the accumulation of anthocyanin in these different clones. Exogenous GA₃ and uniconazole had no effect on 3'-nucleotidase and 5'-nucleotidase activity in the carrot cell suspension cultures. Thus 3'-nucleotidase does not appear to play a role in the inhibition of anthocyanin accumulation by exogenous GA₃.