HOME :: CHAPTER 25 :: Essay 25.1

Essay 25.1

The Role of Gibberellins in Floral Evocation of the Grass Lolium temulentum

Rod W. King and Lloyd T. Evans, CSIRO, Plant Industry, GPO Box 1600 Canberra, ACT 2601, Australia

September, 2006

Introduction

Almost 70 years ago it was recognized that the leaf is the site of the initial response in daylength-regulated flowering. However, the nature of the factors transported from the leaf to the shoot apex where the flowers form has remained elusive. Here we summarize the evidence from studies with the grass Lolium temulentum, which shows that specific gibberellins (GAs) act as its "floral stimulus." This evidence satisfies five requirements, namely:

The production of florigenic GAs in the leaf when exposed to a photoinductive long day
Their transport from the leaf to the shoot apex
Their sufficient increase in the shoot apex when floral evocation occurs
Morphological and molecular changes at the apex, especially involving floral organ identity genes
Replacement of the long-day requirement by specific GAs

Our studies over the last 40 years, reviewed by King and Evans (2003), have established a continuous trail of evidence for the gibberellin (GA) class of plant hormone as a floral stimulus in the long-day regulated flowering of the grass, Lolium temulentum. Its long day (LD) photoresponse in the leaf blade involves far-red rich phytochrome-active light, which enhances the expression of a GA biosynthetic enzyme and alters GA precursor and product levels. Most significantly, the florigenic GAs increase in the shoot apex not long after they rise in the leaf, and at the time of the earliest molecular and morphological changes at the shoot apex. Finding the final pieces of this scientific jigsaw has involved GA application studies. Not only was it necessary to show that applied GAs could replace the need for a LD, but by utilizing a range of structurally related synthetic and natural GAs as well as inhibitors of GA synthesis, we have established that some GAs promote inflorescence initiation while others are more active for stem elongation or may participate in the later stages of flowering of L. temulentum.

Gibberellins for Flowering

In 1956, Anton Lang reported for the first time that applied gibberellic acid (GA₃), like vernalization or exposure to long days (LDs), could cause plants of Hyoscyamus niger first to bolt, and subsequently, to flower. His finding was soon confirmed both for other dicot species, for grasses and for GAs other than GA₃ (Lang 1957).

Flowering of many temperate grasses is induced by LD or by treatment with GA₃. For example, the grass Lolium temulentum requires a single LD for inflorescence initiation, and this requirement can be met in noninductive short days (SD) by a single application of some GAs, either to the leaf blade or to the shoot apex (Evans 1964a, 1969; Evans et al. 1990). However, there are differences between some GAs and LD in this response. Over at least the first three weeks of inflorescence development, there is little or no stem elongation/bolting associated with the LD-induced flowering of L. temulentum and perhaps other grasses—a response duplicated by application of some GAs—but many GAs that induce flowering of grasses also cause early and excessive stem elongation.

The responses to GAs may be similar across dicot and monocot species even though the effects on stem elongation and flowering may occur in a different order, presumably reflecting a different order of gene activation associated with different GAs. For example, GAs that stimulate stem elongation may not be involved in the process of floral evocation in grasses, although they could be involved in subsequent inflorescence development. Conversely, other GAs may be florally effective because they cause inflorescence initiation with little or no effect on stem elongation.

The Search for GAs that are Florigenic but Ineffective for Stem Elongation

Beginning in the mid 1980s, and in collaboration with Dick Pharis and Lew Mander, we compared the effects of many different gibberellins on floral evocation and stem elongation in L. temulentum. Some, like GA₈ and GA₉, influence neither process. 16,17-dihydro GA₅ promotes flowering while inhibiting stem growth (Evans et al. 1994b), while GA₁ and GA₃ enhance both and are therefore unlikely to be the floral stimulus in L. temulentum (Evans et al. 1990). However, exogenous GA₅ and GA₆ are both able to cause floral evocation and induce the early stages of inflorescence development in SD at doses that have no effect on stem elongation (King et al. 1993, 2003). Two questions arise from such findings. Firstly, are either or both GA₅ and GA₆ agents of floral evocation by one LD in L. temulentum? Secondly, are the more growth-effective GAs such as GA₁ and GA₄ involved in the subsequent processes of inflorescence development in this grass? To answer these questions, in the following sections we consider evidence based on endogenous gibberellin contents of the leaf and apex of L. temulentum along with information on metabolic pathways involved in GA synthesis and degradation.

Critical Steps in GA Metabolism

The relevant steps in GA metabolism as summarized by Hedden and Phillips (2000) are shown below:

Figure 1 (Click image to enlarge.)

The key to biological activity among many gibberellins is the closure of the lactone ring with the conversion of GA₁₉ to GA₂₀ and GA₂₄ to GA₉. This step involves a 20-oxidase whose activity increases in LD, as shown for spinach and Arabidopsis (Wu et al. 1996; Xu et al. 1997). A multifunctional 3-oxidase completes the conversion of the biologically inactive GA₂₀ or GA₉ to bioactive GAs—including GA₅ and GA₆—and to the 3-hydroxylated GA₁, GA₃, and GA₄. A key inactivation step involves the 2-oxidase responsible for adding a hydroxyl at C-2 to give GA₈, GA₂₉, or GA₃₄.

GAs for Floral Evocation: Production in the LD Leaf and Transport to the Shoot Apex

In L. temulentum, when the leaf is exposed to a single photoinductive LD, expression of the messenger RNA for a 20-oxidase increases dramatically after 16 hours of light and peaks 4 hours later (Figure 2). Most significantly, the timing of the increase in the 20-oxidase in LD matches the minimum duration of light required for floral induction of L. temulentum by one LD. In SD, the expression level of mRNA for this key enzyme is much weaker and peaks 12 hours later during the next 8-hour-daylight period.

Figure 2 Expression of a L. temulentum 20-oxidase GA biosynthesis gene. Compared with an 8h short day, LD exposure of the leaf (8h SD extended by 9.5h using incandescent lamps) led to a large increase in gene expression. (Semi-quantitative RT PCR assays from Blundell, MacMillan, and King; unpublished).

The consequence of increased 20-oxidase activity in L. temulentum leaves is a fall in GA₁₉ levels when the critical daylength is reached around midnight, and an equivalent rise in GA₂₀ (GA₁₉ falls from a peak of 35 ng g^-1 DW to a minimum of 12 ng g^-1 around midnight, while GA₂₀ rises from 3 ng g^-1 to 23 ng g^-1; King et al. in preparation). GA₅, an immediate product of GA₂₀, also increases in LD but not in SD leaves so that the difference by midnight (i.e., after 16 hours of light) is fourfold. GA₁ could also be expected to rise in the LD leaves, but this is not apparent until the following day.

The next step, the export of GA₅ from the leaf blades and down the leaf sheath has not been directly documented. The difficulty is to isolate specific vascular tissue and to sense potentially small changes in GAs. As an alternative, we have assessed movement from the leaf to the apex of labelled GA₅. In LD, ²H₄-GA₅ applied to the leaf blade was transported intact and quantitatively to the shoot apex (King et al. 2001). Earlier experiments in which the blade and sheath of the single LD leaf were cut off at various positions and times suggest that the LD stimulus is translocated to the shoot apex at a speed of 1–2.4 cm h^-1, compared with the simultaneous transport of sugars at about 80–100 cm h^-1 (Evans and Wardlaw 1966). Furthermore, evocation can be maximal in shoot apices, which throughout the low intensity daylength extension, show no increase in their sucrose content over the 2% found in the vegetative plants exposed to SD (King and Evans 1991). The LD florigenic stimulus in L. temulentum is clearly not sucrose, and an increase in the content of sucrose in the shoot apex is neither necessary nor sufficient for its floral evocation.

Based on its apparent speed of translocation, the LD floral stimulus should begin to arrive at the shoot apex and floral evocation begin on the morning after the LD. Such timing of floral evocation was confirmed by excising shoot apices at various times after the LD, onto media supporting apex development (McDaniel et al. 1991). King et al. (1993) confirmed these results but also found with plants of various ages that shoot apices excised from vegetative plants in SD were induced to flower by supplying GA₃ in the medium and could reach a similar stage of flowering to those from plants given one LD but with no GA₃ in the medium.

Floral evocation in plants of L. temulentum is associated with an increase on the day after the LD in ³²P incorporation into RNA, and of ³⁵S into protein in shoot apices (Rijven and Evans 1967; Evans and Rijven 1967). These increases occur in the dome of the apex and in the sites of future spikelets, down both sides (Knox and Evans 1968). Thus, the timing of these changes fits with the estimated time of arrival of the LD photoperiodic stimulus in the apex (Evans and Wardlaw 1966; McDaniel et al. 1991; King et al. 1993). However, to complete the evidence relating GAs to floral evocation in L. temulentum, it was essential to measure their content in the shoot apex.

The minute size of the shoot apex (<3 µg dry weight) had made it difficult to measure its GA content, but this became feasible through a recent collaboration with Thomas Moritz in Sweden. Using highly sensitive GCMS techniques, he analysed GAs at subpicogram levels in the shoot apex of L. temulentum. As shown in Table 1, by the end of the daylight period following the overnight LD, the content of the highly florigenic GA₅ and GA₆ at least doubles in the shoot apex (King et al. 2001, 2003). Moreover, at this time the maximum GA₅ concentration reached in the shoot apices is 3 × 10^-7M which is close to that needed in an agar medium if shoot apices excised from plants in SD are to flower (about 5 × 10^-7M GA₅; King et al. 1993). It is highly probable, therefore, that translocation of these two GAs from leaf blades in LD causes floral evocation of L. temulentum.

Table 1

Changing the Guard: Inflorescence Development and Growth-Active GAs

A number of bioactive GAs and precursors including GA₁, GA₄, and GA₉ are notably absent from the shoot apex or minor in SD and on the day after the inductive LD, and remain so for 6–10 days. After exposure of the leaf to 2 or 3 LD, they all increase in content at the apex (King et al. 2001), whereas the content of GA₅, GA₁₉, and GA₂₄ falls more rapidly with additional LD. Such additional LDs accelerate inflorescence development (Evans 1958, 1960; Evans and Blundell 1996) and, as well, increase the content of GA₁, GA₄, and GA₉ in the LD leaf (Gocal et al. 1999). Thus, as for floral evocation, for inflorescence development there are increases in GAs in LD but now for a new group of GAs that are highly active for stem elongation.

During inflorescence development, the manyfold increase in the shoot apex of the "new guard" GAs—GA₁ and GA₄—along with evidence that their application can now promote flowering (King et al. 2001), shows their importance for inflorescence development. That a GA input is effective during floral development is also evident from studies with apices induced to flower by one LD, and then excised. Progress to flowering is weak unless GAs are supplied by about six days after the end of the LD (King et al. 1993). Which GAs are important at this time is indicated by the response to application of two growth retardants, Trinexapac Ethyl and LAB 198 999. These retardants block synthesis of GA₁ and GA₄ by 3-oxidases and inhibit inflorescence development but show no inhibition when applied earlier during floral evocation (Evans et al. 1994a). Thus, whereas GA₅ and GA₆ are the gibberellins most active in floral evocation of L. temulentum, it is the elongation-active, C-3-hydroxylated GAs such as GA₁ and GA₄ that play a role in subsequent inflorescence development.

Since GA₁ and GA₄ can be detected in leaves of L. temulentum—whether from vegetative or floral plants—there must be some mechanism first to exclude them from the shoot apex and then allow them to access the shoot apex late in inflorescence development. As discussed below, this exclusion mechanism may involve: (i) degradation of specific GAs by a 2-oxidase; (ii) localization of a 2-oxidase just below the apex of vegetative plants and; (iii) disappearance of this 2-oxidase during inflorescence development. The implication that some GAs are more readily degraded than others (e.g., GA₁ or GA₄ vs GA₅ or GA₆) also provides a focus in the following section for understanding how GA structure affects response.

Structural Considerations: The Lord of the Rings

The functional groups contributing to GA activity are indicated in Figure 3, and from our comparisons of many GAs (Evans et al. 1990, 1994 a,b; King et al. 2003; Mander et al. 1998 a,b), it is clear that the structural requirements for florigenicity are quite different from those for stem elongation. As an example, for four GAs—2,2-dimethyl GA₄, GA₃₂, GA₁, and GA₄—although not greatly different in their effectiveness in promoting stem elongation, there is up to a ten thousandfold range in their florigenic activity.

Figure 3 Gibberellin structure and carbon numbering.

Differences in activity of an applied GA are likely to be determined not only by their ability to resist degradation (see above) but also by structural features altering uptake and transport, their inherent bioactivity, their potential as a biosynthetic substrate, and their capacity to interfere with endogenous GA synthesis and degradation. GA uptake and transport is unlikely to be critical to florigenicity, as GAs could variously promote either or both stem elongation and flowering. Those structural elements favouring florigenicity in particular include:

(i) Formation of the lactone bridge between carbons 4 and 10, so adding a fifth ring to the 4-ringed structure of the biologically inactive GAs. This requirement seems to apply to all the LD grasses whose flowering is induced by gibberellins. For example, GA₁₉ (a 20-carbon precursor GA) is wholly inactive, whereas GA₃, GA₅, GA₇, and several other 19-carbon GAs have reported florigenic activity in other species besides L. temulentum. However, formation of the lactone ring itself is insufficient, and a further step involving activity of a 3-oxidase enzyme is also essential (see iii).

(ii) A free carbon-7 carboxy group is an absolute requirement for all plant GA responses and may reflect ability to bind to a GA receptor.

(iii) Hydroxylation at carbons 3, 12, 13, and 15 and/or the presence of a C-1, 2 double (C-C) bond (as in GA₃), or a 2,3 double bond (as in GA₅), or a 2,3 epoxide (as in GA₆).

(iv) Absence of a C-2 hydroxyl.

While features listed under (iii) may enhance the stability of a GA against inactivation by 2β-hydroxylation, structural elements at C-2 are the most significant in this regard. A good comparison involves the responses to applications of three closely related GAs: the highly florigenic and growth active 2,2-dimethyl GA₄, the modestly florigenic but reasonably growth-active 2α-methyl GA₄, and the nonflorigenic, growth-active GA₄ (Evans et al. 1990). Because it lacks any C-2 structural elements, 2-hydroxylation of GA₄ is highly likely and this apparently rings the death knell for any florigenic activity. GA₄ is taken up by—and transported in—the plant as it is highly active for stem elongation but quite clearly it is not transported intact into the shoot apex. The most likely explanation, for this observation would involve the presence of a concentrated zone of 2-oxidase just below the vegetative apex, and while this has yet to be demonstrated for L. temulentum, such a highly localized zone of 2-oxidase RNA expression has been reported recently for rice (Sakamoto et al. 2001).

The concept for rice of the exclusion of 2-hydroxylation-sensitive GAs from the shoot apex fits well with the structures of the GAs that are naturally found in the shoot apex of L. temulentum. Neither of the readily 2-hydroxylated GAs we examined, GA₁ and GA₄, was detectable in the shoot apex of vegetative plants or for about the first week after floral evocation (King et al. 2001) although their content increased in LD leaves (Gocal et al. 1999). By contrast, we detected highly florigenic but weakly growth-active GAs—such as GA₅ and GA₆—these GAs probably being protected from C-2 hydroxylation by the C-2,3 double bond in GA₅ and the C-2,3 epoxide in GA₆. High florigenicity with limited growth activity for GA₃₂ also fits with its having a C-1,2 double bond. Later, as floral development proceeds GA₅ and GA₆ decrease in the apex while there is a dramatic increase in GA₁ and GA₄ (King et al. 2001) and it is at this stage of flower development that the localized band of 2-oxidase disappears from just below the rice inflorescence (Sakamoto et al. 2001).

Apparently, the localized band of 2-oxidase activity protects the shoot apex from "growth-active" GAs, so guaranteeing the integrity of the apex during vegetative growth, floral evocation, and early floral differentiation. Later, during inflorescence development, disappearance of the 2-oxidase barrier allows the influx of highly growth-active GAs.

Explaining a Paradox: Flowering Promotion by GA Biosynthesis Inhibitors

Classic studies by Baldev and Lang (1965) with the rosette LDP Samolus parviflorus showed that flowering in LD is inhibited when GA synthesis is blocked using either of the GA biosynthesis inhibitors, Amo-1618 and Cycocel (CCC). Such results supported a role for GAs in the LD response. Paradoxically, however, with L. temulentum, several "anti-gibberellins"—which might be expected to inhibit floral induction—actually promote it, and act synergistically with some GAs. For example, CCC alone did not change the flowering response to one LD, but greatly enhanced the promotive effect of GA₃, although other "anti-gibberellins" such as B9 and Amo 1618 gave the expected inhibiting effect on flowering (Evans 1969). With the acylcyclohexanedione inhibitors, LAB 198 999 and Trinexapac Ethyl, promotion of flowering could reflect an interference with the 2-oxidase. Whether or not this enzyme is localized below the vegetative shoot apex, endogenous or applied bioactive GAs would be spared from inactivation. Another explanation involves either or both inhibition of the 3-oxidase(s) responsible for conversion of GA₂₀ to GA₁ and inhibition of the 2-oxidase involved in converting GA₂₀ to inactive GA₂₉. By inhibiting the conversion of GA₂₀ to either GA₁ or GA₂₉, LAB 198 999 and Trinexapac Ethyl treatments could promote GA₂₀ conversion to GA₅, thereby accounting for their strongly promotive effects on floral evocation in L. temulentum. Their subsequently inhibitory effects on inflorescence development (Evans et al 1994a) when synthesis of GA₁ and GA₄ becomes important, fits with their known action as 3-oxidase competitors. The greater inhibition of the 2-oxidase by the derivative 16,17-dihydro GA₅ than by LAB 198 999 (Junttila et al. 1997), indicates further interesting effects of "antigibberellins," which preclude drawing too simple an interpretation of how they may alter GA metabolism.

The Genes Involved at the Shoot Apex

Of the various genes known to be involved in floral determination and differentiation, the earliest to be expressed in the shoot apex of L. temulentum is LtCDC2, which greatly increases in activity in the future spikelet sites by the afternoon after the long day, just when floral evocation is completed (Gocal 1997). By the next day, the API-like gene LtMADS2 and a related gene LtMADS1 have also increased their expression at the spikelet sites and subsequently at the floret sites (Gocal et al. 2001).

Whereas the LEAFY (LFY) gene is expressed early in the floral apices of Arabidopsis, Gocal et al. (2001) found it not to be expressed until about 12 days after LD induction in L. temulentum—possibly associated with the great difference between the two species in the timing of stem elongation vis-à-vis floral evocation, and, as well, in the GAs involved.

Conclusion

For no other species is there such a complete and consistent trail of evidence on the identity of the LD floral stimulus as in L. temulentum, from the daylength-sensitive, physiologically mobile GA₅, and probably GA₆, in the leaf blades, translocated intact and at appropriate velocity to reach the shoot apex in sufficient concentration to effect floral evocation at the clearly identified time.

Moreover, several previously puzzling features of this investigation, especially those relating to the great range in florigenicity among the gibberellins and to the synergistic effects of several inhibitors of GA synthesis, can now be resolved. If, as in rice, the vegetative shoot apex in L. temulentum is protected by a ring of 2-oxidase, a very coherent picture of the control of flowering in L. temulentum by gibberellins emerges. In addition, such localized action of the 2-oxidase explains much of the variation among bioactive GAs in their relative promotion of floral evocation vis-à-vis stem elongation.

Our findings also highlight a feature of the growth habit of temperate grasses, which may have been crucial to their evolutionary success when close-grazed by ungulates. Except for the relatively brief stage when the inflorescences must grow upwards for wind pollination and seed dispersal, the terminal meristems are kept close to the ground by the absence of stem elongation until the later stages of inflorescence development. The initial involvement at floral evocation of GAs such as GA₅ and GA₆, which do not cause stem elongation at concentrations that are florigenic, and the initial exclusion from the apex of GAs such as GA₁ and GA₄, which do cause stem elongation, presumably aids survival under both close-grazing and adverse environmental conditions. Such an evolutionary explanation is required, because stem elongation per se is not antagonistic to flowering. Applied 2,2-dimethyl GA₄, for example, not only induces flowering but also causes massive stem elongation.

For a number of reasons, our findings that specific GAs are florigens in grasses, cannot be generalized to all species. Based on their evolutionary relatedness (see Kellogg 2001), we believe our findings are applicable to other temperate grasses and cereals, which retain a LD response—as most do (Evans 1964b; Heide 1994). However, the warm climate grasses and cereals—including rice, corn, and sugarcane—often have no photoperiodic response and if they are sensitive, they often respond to SD, which may lead to decreases, not increases, in GA content. Where a species is insensitive to photoperiod, a role for GAs appears unlikely and one of many other florigenic factors must be limiting. In temperate, LD-responsive dicots, GA increases have often been documented but it is also clear that their primary action may be on stem elongation rather than on flowering. This latter observation is especially interesting as the GAs detected in dicots have always been the growth-active ones, and the florigenic but weakly growth-active GAs have yet to be examined in such dicots. Nevertheless, there is no reason at present to suggest that the growth-inactive GAs that are florigens in grasses play any role in flowering of dicots.

References

Baldev, B., and Lang, A. (1965) Control of flower formation by growth retardants and gibberellins in Samolus parviflora, a long-day plant. Amer. J. Bot. 52: 408–417.

Evans, L. T. (1958) Lolium temulentum L., a long-day plant requiring only one inductive photocycle. Nature 182: 197–198.

Evans, L. T. (1960) The influence of environmental conditions on inflorescence development in some long-day grasses. New Phytol. 59: 163–174.

Evans, L. T. (1964a) Inflorescence initiation in Lolium temulentum L. V. The role of auxins and gibberellins. Aust. J. Biol. Sci. l7: l0–23.

Evans, L. T. (1964b) Reproduction. In Grasses and Grasslands, C. Barnard, (Ed.), pp. 126–153. MacMillan: London.

Evans, L. T. (1969) Inflorescence initiation in Lolium temulentum L. XIII. The role of gibberellins. Aust. J. Biol. Sci. 22: 773–786.

Evans, L. T., and Blundell, C. (1996) The acceleration of primordium initiation as a component of floral evocation in Lolium temulentum L. Aust. J. Plant Physiol. 23: 569–576.

Evans, L. T., and Rijven, A. H. G. C. (1967) Inflorescence initiation in Lolium temulentum L. XI. Early increases in the incorporation of ³²P and ³⁵S by shoot apices during induction. Aust. J. Biol. Sci. 20: 1033–1042.

Evans, L. T., and Wardlaw, I. F. (1966) Independent translocation of ¹⁴C-labelled assimilates and of the floral stimulus in Lolium temulentum. Planta 68: 319–326.

Evans, L. T., King, R. W., Chu, A., Mander, L. N., and Pharis, R. P. (1990) Gibberellin structure and florigenic activity in Lolium temulentum, a long day plant. Planta. 182: 97–106.

Evans, L. T., King, R. W., Mander, L. N., and Pharis, R. P. (1994a) The relative significance for stem elongation and flowering in Lolium temulentum of 3β-hydroxylation of gibberellins. Planta 192: 130–136.

Evans, L. T., King, R. W., Mander, L. N., Pharis, R. P., and Duncan, K. A. (1994b) The differential effects of C-16,17-dihydro gibberellins and related compounds on stem elongation and flowering in Lolium temulentum. Planta 193: 107–114.

Gocal, G. F. W. (1997) Molecular biology of floral evocation in Lolium temulentum. PhD dissertation. Aust. Natl. Univ., Canberra. 181 pp.

Gocal, G. F. W., Poole, A. T., Gubler, F., Watts, R. J., Blundell, C., and King, R. W. (1999) Long-day up-regulation of a GAMYB gene during Lolium temulentum inflorescence formation. Plant Physiol. 119: 1271–1278.

Gocal, G. F. W., King, R. W., Blundell, C. A., Schwartz, O. M., Andersen, C. H., and Weigel, D. (2001) Evolution of floral meristem identity genes: Analysis of Lolium temulentum genes related to APETALA1 and LEAFY of Arabidopsis. Plant Physiol. 125: 1788–1801.

Hedden, P., and Phillips, A. L. (2000) Gibberellin metabolism: New insights revealed by the genes. Trends Plant Sci. 5: 523–530.

Heide, O. M. (1994) Control of flowering and reproduction in temperate grasses. New Phytol. 128: 431–462.

Junttila, O., King, R. W., Poole, A., Kretchmer, G., Pharis, R. P., and Evans, L. T. (1997) Regulation in Lolium temulentum of the metabolism of gibberellin A₂₀ and gibberellin A₁ by 16,17-dihydro GA₅ and by the growth retardant, LAB 198 999. Aust. J. Plant Physiol. 24: 359–369.

Kellogg, E. A. (2001) Evolutionary history of the grasses. Plant Physiol. 125: 1198–1205.

King, R. W., and Evans, L. T. (1991) Shoot apex sugars in relation to long-day induction of flowering in Lolium temulentum L. Aust. J. Plant Physiol. 18: 121–135.

King, R. W., and Evans, L. T. (2003) Gibberellins and flowering of grasses and cereals: Prizing the lid of the "Florigen" black box. Ann. Rev Plant Biol. 54: 307–328.

King, R. W., Blundell, C., and Evans, L. T. (1993) The behaviour of shoot apices of Lolium temulentum in vitro as the basis of an assay system for florigenic extracts. Aust. J. Plant Physiol. 20: 337–348.

King, R. W., Evans, L. T., Mander, L. M., Moritz, T., Pharis, R. P., and Twitchin, B. (2003) Synthesis of gibberellin GA₆ and examination of its role in flowering of Lolium temulentum. Phytochem. 62: 72–82.

King, R. W., Moritz, T., Evans, L. T., Junttila, O., and Herlt, A. J. (2001) Long-day induction of flowering in Lolium temulentum involves sequential increases in specific gibberellins at the shoot apex. Plant Physiol. 127: 624–632.

Knox, R. B., and Evans, L. T. (1968) Inflorescence initiation in Lolium temulentum L. XII. An autoradiographic study of evocation in the shoot apex. Aust. J. Biol. Sci. 21: 1083–1094.

Lang, A. (1957) The effect of gibberellin on flower formation. Proc. Natl. Acad. Sci 43: 709–717.

Mander, L. N., Adamson, G., Bhaskar, V. K., Twitchin, B., Camp, D., King, R. W., and Evans, L. T. (1998a) Effects of 17-alkyl-16,17-dihydrogibberellin A₅ derivatives on growth and flowering in Lolium temulentum. Phytochemistry 49: 1509–1515.

Mander, L. N., Sherburn, M., Camp, D., King, R. W., Evans, L. T., and Pharis, R. P. (1998b) Effects of D-ring modified gibberellins on flowering and growth in Lolium temulentum. Phytochemistry 49: 2195–2206.

McDaniel, C. N., King, R. W., and Evans, L. T. (1991) Floral determination and in vitro floral differentiation in isolated shoot apices of Lolium temulentum L. Planta 185: 9–16.

Rijven, A. H. G. C., and Evans, L. T. (1967) Inflorescence initiation in Lolium temulentum L. X. Changes in ³²P incorporation into nucleic acids of the shoot apex at induction. Aust. J. Biol. Sci. 20: 13–24.

Sakamoto, T., Kobayashi, M., Itoh, H., Tagiri, A., Kayano, T., Tanaka, H., Iwahori, S., Matsuoka, M. (2001) Expression of a gibberellin 2-oxidase gene around the shoot apex is related to phase transition in rice. Plant Physiol. 125: 1508–1516.

Wu, K. S., Li, L., Gage, D. A., and Zeevaart, J. A. D. (1996) Molecular cloning and photoperiod-regulated expression iof gibberellin 20-oxidase from the long-day plant spinach. Plant Physiol. 110: 547–554.

Xu, Y., Gage, D. A., and Zeevaart, J. A. D. 1997. Gibberellins and stem growth in Arabidopsis thaliana: Effects of photoperiod on the GA₄ and GA₅ loci. Plant Physiol. 114: 1471–1476.