The Horticulture Journal
Online ISSN : 2189-0110
Print ISSN : 2189-0102
ISSN-L : 2189-0102
ORIGINAL ARTICLES
Florigenic Effect of Gibberellin on Flowering According to Period of Chilling Treatment in Lavandula × intermedia
Masaji KoshiokaTaiga HorimotoYoshiyuki MuramatsuSatoshi KubotaTamotsu Hisamatsu
Author information
JOURNALS OPEN ACCESS FULL-TEXT HTML

2016 Volume 85 Issue 2 Pages 169-176

Details
Abstract

The effect of gibberellin (GA) on chilling-induced stem elongation and flowering in Lavandula × intermedia was investigated using several GAs and GA biosynthesis inhibitors related to chilling temperature and its period. Identification of GA1, GA19, GA20, and GA53 as endogenous GAs by GC/MS suggests the early C-13 hydroxylation pathway (→GA53→GA44→GA19→GA20→GA1→) is functioning, and GA1 is the biologically active GA in L. × intermedia. GA3 increased stem elongation with or without chilling treatment (CT), but did not induce flowering without CT. There was little difference in the effects on stem elongation and flowering among GA1, GA3, GA5, and dimethyl-GA4. GA biosynthesis inhibitor, especially uniconazole-P, inhibited stem elongation and flowering, but the inhibition was abolished by GA3. At least several weeks of CT were necessary for flowering, for example, in 8-week CT at 5°C, 9-week CT at 6°C and 12-week CT at 7°C. This period was defined as the minimum CT period. Under the minimum CT period, GA3 enhanced the flowering. However, the effect on flowering by GA3 was not found when the CT period became longer, for example, in 12-week CT at 5°C, 12-week CT at 6°C and 15-week CT at 7°C. This longer period was defined as the maximum CT period. These results suggest that the expression of endogenous GA is necessary for flowering in L. × intermedia, that the florigenic effect of GA3 on flowering varies with the period of CT, and that GA3 enhances the effect of CT on flowering in L. × intermedia, but cannot replace CT.

Introduction

Lavandula × intermedia is a sterile hybrid between L. angustifolia and L. latifolia (McNaughton, 2000), which is hardy, and considered a shrub, but is commonly produced and marketed as an herbaceous perennial. Thus, L. × intermedia is cultivated in Japan as an ornamental plant and for its oil, which is used in perfumery, and as a dry flower (Kubota et al., 2010). Lavandula × intermedia is a vernalization-type plant that requires exposure to chilling temperature at 5°C for a period of 10 weeks for subsequent flowering (Kubota et al., 2010), as shown in L. angustifolia (Monaghan et al., 2004; Whitman et al., 1996). However, the relationship between chilling temperature and period of chilling treatment (CT) has not been clarified in L. × intermedia.

It has long been suggested that endogenous gibberellins (GAs) play a regulatory role in stem elongation and flowering of some chilling requiring plants and GA can replace CT (Hasebroek et al., 1993; Hisamatsu et al., 2000; Nishijima et al., 1997; Pharis and King, 1985; Wittwer and Bukovac, 1957; Zanewich and Rood, 1995; Zeevaart, 1983). Structural requirements for florigenic activity among GAs and GA derivatives have been investigated. In Lolium temulentum, 2,2-dimethyl-GA4 had stronger florigenic activity than GA1, GA3, and GA5 (Evans et al., 1990). In Matthiola incana, 2,2-dimethyl-GA4 and GA4 showed florigenic activity, but GA1 and GA3 did not (Hisamatsu et al., 2000). These findings indicate that the effect of GA on flowering depends on the GA structure and/or plant species. However, the effect of GA on flowering in L. × intermedia has not been investigated.

One hundred and thirty six GAs have now been endogenously identified (MacMillan, 2002). Of those, GAs such as GA1, GA3, GA4, GA7, GA35, and GA80, are considered as biologically active GAs. Each of them has at least a hydroxyl group on carbon skeleton C-3β position, a carboxyl group on C-6 position, and a lactone between C-4 and C-10 positions. These structural characteristics are required for biological activities. Although several biosynthetic pathways to these biologically active GAs are found in higher plants, the major biosynthetic pathways are thought to be the early C-13 hydroxylation pathway (→GA53→GA44→GA19→GA20→GA1→) and the non-hydroxylation pathway (→GA12→GA15→GA24→GA9→GA4→) (Yamaguchi, 2008). However, it is unknown which GA biosynthetic pathway is mainly functioning in L. × intermedia.

This study examined the relationship between chilling temperature and period of CT on flowering in L. × intermedia. Then, we identified endogenous GA and its metabolic pathway in L. × intermedia, and investigated the promotional effect of GA on stem elongation and flowering in L. × intermedia related to chilling temperature and period of CT.

Materials and Methods

Plant material and cultivation

Mother stock plants in L. × intermedia ‘Super Sevillian Blue’, which were maintained in a greenhouse at College of Bioresource Sciences, Nihon University, were used for experiments. Cuttings were collected from the mother plants every spring, and transplanted to 9 cm plastic pots as one plant per pot. Then, plants were grown in a greenhouse maintained below 25°C in day and above 15°C in night (25°C/15°C, day/night) until the experiments. Plants were top-watered as necessary with a water-soluble fertilizer (Hyponex; HYPONeX Japan Co. Ltd., Osaka, Japan) in moderate dilution.

Experiment 1. Qualitative analysis of endogenous GAs

Combined vegetative and inflorescence shoots (315.3 g in dry weight), which were collected from the mother plants, were extracted with 80% methanol. The extract was purified and fractionated with C18-HPLC by routine methods as described in Oikawa et al. (2004). As HPLC conditions, ODS-425-D column (250 mm × 10 mm i.d.; Senshu Scientific Co. Ltd., Tokyo, Japan) and a linear gradient of 1.0% acetic acidic water-methanol (30% methanol for 2 min, followed by 30 min from 30 to 100% methanol; and finally 18 min with 100% methanol) were used. Column oven temperature was 35°C. Flow rate was 3 mL·min−1. GA-like activity of each C18-HPLC fraction was assayed by rice micro-drop bioassay (Nishijima et al., 1992). Fractions showing biological activities were further purified on N(CH3)2-HPLC eluted with 0.1% acetic acid in methanol at a flow rate of 3 mL·min−1 at room temperature. As an HPLC column, N(CH3)2-4151-N (150 mm × 10 mm i.d.; Senshu Scientific Co. Ltd.) was used. GA-like activity of each N(CH3)2-HPLC fraction was assayed as described above. Fractions showing biological activity were dissolved in a small amount of methanol, and methylated with ethereal diazomethane. They were then dried and trimethylsilylated with N,O-Bis (trimethylsilyl) trifluoroacetamide at 75°C for 30 min. The methylated and trimethylsilylated (MeTMSi) fractions were co-injected along with hydrocarbons for obtaining Kovats Retention Indices (KRIs; Kovats, 1958) onto a DB-1MS microcapillary column (30 m × 0.25 mm i.d., 0.25 μm film thickness; Agilent Technologies Japan, Tokyo, Japan) in a GC/MS system (5975C mass spectrometer equipped with a 7890A GC; Agilent Technologies Japan) by modifying the method of Tanaka-Ueguchi et al. (1998). Identification of endogenous GAs was performed based on HPLC retention time, KRIs, and mass spectra compared with those of authentic protio GAs.

Experiment 2-1. Effect of two CT periods at 8°C and GA3, 2,2-dimethyl-GA4, and GA5 on flowering

Rooted cuttings of ‘Super Sevillian Blue’ were pruned as each plant had five lateral shoots on 6th April, 2007. Plants were maintained in a greenhouse until 26th June, 2007. Then, 40 plants were transferred to a chilling room maintained at 8°C under an 8-h photoperiod by 10-watt fluorescent lamps (National ball-type fluorescent lamp; EFA15EN). The averaged PPFD level was about 10 μmol·m−2·s−1, which was almost the same as Whitman et al. (1996) and Nishijima et al. (1997). Each set of 5 plants was moved to a phytotron from the chilling room after 5- and 10-week CTs. The phytotron (FR-535AS6; Koito Industries Ltd., Yokohama, Japan) was maintained at 23°C/15°C (day/night) under a 16-h photoperiod. The PPFD level was 490 μmol·m−2·s−1, supplied from metal halide lamps (MT400DL/BUD; Iwasaki denki, Tokyo, Japan). Acetone solutions (10%) containing 0.025% Tween 20 of GA3, 2,2-dimethyl-GA4, and GA5 were prepared. After CT, 10 μL of each GA solution (100 ng·μL−1) was applied to shoot tips twice a week until flower buds were observed. Control plants were treated with 10% acetone solution. Watering and fertilization were done as the above. Days to visible buds from the end of CT (DVB) and days to the first open flower from the end of CT (DFLW), rate of flowering plants (RFP), rate of flowering shoots (RFS) and stem elongation were recorded for 19 weeks in 5-week CT plants and 14 weeks in 10-week CT plants, after the corresponding CT.

Experiment 2-2. Effect of four CT periods at 6°C and GA1, GA3, and GA5 on flowering

Rooted cuttings of ‘Super Sevillian Blue’ were pruned as each plant had five lateral shoots on 20th April, 2008. Plants were maintained in a greenhouse until 10th June, 2008. Then, 240 plants were transferred to a chilling room maintained at 6°C under an 8-h photoperiod by 10-watt fluorescent lamps. Each set of 15 plants was moved to the same phytotron under the same conditions as described in Exp. 2-1 after 0-, 6-, 9-, and 12-week CTs. Solution of GA1, GA3, and GA5 (100 ng·μL−1) were prepared, and applied as described in Exp. 2-1. Control plants were treated with 10% acetone solution. Watering and investigation of flowering were done as the above. The duration of investigation was 24 weeks for 0-week CT plants, 18 weeks for 6-week CT plants, 15 weeks for 9-week CT plants, and 12 weeks for 12-week CT plants, after the corresponding CT.

Experiment 2-3. Effect of five CT periods at 7°C and GA3 on flowering

Rooted cuttings of ‘Super Sevillian Blue’ were pruned as each plant had five lateral shoots on 8th May, 2009. Plants were maintained in a greenhouse until 2nd August, 2009. Then, 180 plants were transferred to a chilling room maintained at 7°C under an 8-h photoperiod by 10-watt fluorescent lamps on 2nd August, 2009. Each set of 10 plants was moved to the same phytotron under the same conditions as described in Exp. 2-1 after 0-, 6-, 9-, 12-, and 15-week CTs. After CT, 10 μL of GA3 solution (100 ng·μL−1) was applied as described in Exp. 2-1. Control plants were treated with 10% acetone solution. Watering and investigation of flowering were done as the above. The duration of investigation was 24 weeks for 0-week CT plants, 18 weeks for 6-week CT plants, 15 weeks for 9-week CT plants, 12 weeks for 12-week plants, and 9 weeks for 15-week CT plants, after the corresponding CT.

Experiment 3-1. Effect of four CT periods at 5°C, GA3 and GA inhibitors on flowering

Rooted cuttings of ‘Super Sevillian Blue’ were pruned as each plant had five lateral shoots on 26th December, 2005. Plants were maintained in a greenhouse until 3rd August, 2006. Then, 180 plants were transferred to a chilling room maintained at 5°C under an 8-h photoperiod by 10-watt fluorescent lamps on 3rd August, 2006. Each set of 10 plants was moved to the same phytotron under the same conditions as described in Exp. 2-1 from the chilling room after 0-, 6-, 8-, and 10-week CTs. Uniconazole-P (UCZ; Sumitomo Chemical Co. Ltd., Tokyo, Japan) and prohexadione calcium salt (PCa; Kumiai Chemical Industry Co. Ltd., Tokyo, Japan), inhibitors of endogenous GA biosynthesis, were used to suppress the effect of endogenous GAs before GA3 application to examine GA3 response. At three days before CT, UCZ was applied to the soil in a 20-mL water solution containing 50 mg·L−1 of active ingredient. Acetone solutions (10%) containing 0.025% Tween 20 of GA3 and PCa were prepared. After CT, 10 μL of GA3 solution (0, 1, 10, 100 ng·μL−1) and PCa solution (100 ng·μL−1) were applied to shoot tips once a week until flower buds were observed. Control plants were treated with 10% acetone solution. Watering and investigation of flowering were done as the above. The duration of investigation was 24 weeks for 0-week CT plants, 18 weeks for 6-week CT plants, 16 weeks for 8-week CT plants, and 14 weeks for 10-week CT plants, after the corresponding CT.

Experiment 3-2. Effect of three CT periods at 5°C, GA3 and GA inhibitor on flowering

Rooted cuttings of ‘Super Sevillian Blue’ were pruned as each plant had five lateral shoots on 9th May, 2008. Plants were maintained in a greenhouse until 10th June, 2008. Then, 180 plants were transferred to a chilling room maintained at 5°C under an 8-h photoperiod by 10-watt fluorescent lamps on 10th June, 2008. Each set of 10 plants was moved to the same phytotron under the same conditions as described in Exp. 2-1 from the chilling room after 0-, 8-, and 12-week CTs. UCZ water solutions containing 0, 6.25, 12.5, or 25 mg·L−1 of active ingredient were prepared. UCZ solution (100 mL) was applied to the soil once a week for 2 weeks before CT and once every 3 days after CT started. After CT, 10 μL of GA3 solution (0, 100 ng·μL−1) was applied to shoot tips once a week until flower buds were observed. Control plants were treated with 10% acetone solution. Watering and investigation of flowering were done as the above. The duration of investigation was 24 weeks for 0-week CT plants, 16 weeks for 8-week CT plants, and 12 weeks for 12-week CT plants, after the corresponding CT.

Results

Experiment 1. Qualitative analysis of endogenous GAs

Several biologically active fractions that were found by rice plant bioassay after two consecutive different HPLC separations were analyzed by GC/MS. Comparing full scan modes of mass spectrometry, HPLC retention times, and KRIs on gas-chromatography of GA-like peaks in the biologically active fractions with those of authentic protio GAs, the presence of GA1, GA19, GA20, and GA53 was confirmed (Table 1).

Table 1

Identified gibberellins (GAs) and their Kovats retention indices (KRIs) obtained from full scan GC/MS analysis of the methylated and trimethylsilylated derivatives of endogenous GAs in Lavandula × intermedia and authentic protio GAs.

Experiment 2-1. Effect of two CT periods at 8°C and GA3, 2,2-dimethyl-GA4, and GA5 on flowering

Five-week CT did not induce flowering of plants with or without GA (GA3, dimethyl-GA4, and GA5) in a subsequent warm temperature condition after CT as shown in Table 2. Ten-week CT induced flowering of both plants with and without GA. Both RFP and RFS of GA applied plants were higher than those of untreated plants. DVB in 2,2-dimethyl GA4 applied plants was the shortest in 10-week CT, but there was no significant difference in DFLW among treatments. Stem elongation of GA applied plants was significantly longer than those of untreated plants both in 5- and 10-week CTs. Stem elongation by GA was greater in 10-week CT than in 5-week CT.

Table 2

Effect of chilling treatment (CT) at 8°C and gibberellin A3 (GA3), 2,2-dimethyl-GA4, and GA5 on flowering in Lavandula × intermedia ‘Super Sevillian Blue’.

Experiment 2-2. Effect of four CT periods at 6°C and GA1, GA3, and GA5 on flowering

Similarly to Experiment 2-1, none of the applied GA induced flowering in 0-week CT or 6-week CT as shown in Table 3. Both 9-week CT and 12-week CT induced flowering. In 9-week CT, both RFP and RFS of GA applied plants were slightly higher than those of untreated plants. GA reduced the DVB and DFLW, especially GA3 and GA5 reduced them more than GA1. In 12-week CT, there was no significant difference among RFP, RFS, DVB, and DFLW between GA applied plants and untreated plants. GA elongated stems in all CTs; however, no significant difference was found in either 9-week CT plants or 12-week CT plants.

Table 3

Effect of chilling treatment (CT) at 6°C and gibberellin A1 (GA1), GA3, and GA5 on flowering in Lavandula × intermedia ‘Super Sevillian Blue’.

Experiment 2-3. Effect of 5 CT periods at 7°C and GA3 on flowering

Similarly to the above results, no induction of flowering was observed in plants with or without GA3 after 0-, 6-, and 9-week CTs as shown in Table 4. Twelve-week CT and 15-week CT induced flowering of both plants with and without GA3. In 12-week CT, both RFP and RFS of GA3 applied plants were slightly higher than those of untreated plants. GA3 reduced DVB as shown in the above. In 15-week CT, there was no significant difference in RFP and DVB between untreated plants and GA3 applied plants. GA3 increased stem elongation in all CTs except 12-week CT.

Table 4

Effect of chilling treatment (CT) at 7°C and gibberellin A3 (GA3) on flowering in Lavandula × intermedia ‘Super Sevillian Blue’.

Experiment 3-1. Effect of four CT periods at 5°C, GA3 and GA inhibitors on flowering

No induction of flowering was observed in plants with or without GA3 after 0-week CT and 6-week CT except for the highest dose of GA3 in 6-week CT as shown in Table 5. Eight-week CT and 10-week CT induced flowering of plants with and without GA3 application. In 8-week CT, 75.0% of shoots showed flowering without chemicals. Although GA inhibitors decreased the RFS to 36.7%, other inhibitory effects were not observed. This inhibition was abolished by GA3, and the RFS increased with the GA3 dose. DVB and DFLW decreased inversely with the GA3 dose. The differences in RFP, DVB, and DFLW among GA3 applications in 10-week CT were much less than those in 8-week CT. GA inhibitors suppressed stem elongation of plants only in 8-week CT. The suppression was abolished by GA3. However, this phenomena was not observed in 10-week CT.

Table 5

Effect of chilling treatment (CT) at 5°C and gibberellin A3 (GA3) with/without GA inhibitor on flowering in Lavandula × intermedia ‘Super Sevillian Blue’.

Experiment 3-2. Effect of three CT periods at 5°C, GA3 and GA inhibitor on flowering

No induction of flowering was observed in plants with or without GA3 after 0-week CT as shown in Table 6. In 8-week CT, 80.0% of untreated plants showed flowering. UCZ completely inhibited flowering at both low and high concentrations. This inhibition was abolished by GA3 at a low UCZ concentration. RFP decreased inversely with the UCZ dose. RFP in 12-week CT showed the same tendency as that in 8-week CT. The inhibition of flowering by UCZ increased with the dose. This inhibition was similarly abolished by GA3. Inhibition of flowering by UCZ in 8-week CT was more marked than that in 12-week CT. DVB in 8-week CT was similar among applications. In 12-week CT, DVB were not affected by GA3 in plants not treated with UCZ, but DVB was decreased by GA3 in UCZ applied plants. Stem elongation was inhibited by UCZ, but the inhibition was abolished by GA3.

Table 6

Effect of chilling treatment (CT) at 5°C and gibberellin A3 (GA3) with/without GA inhibitor on flowering in Lavandula × intermedia ‘Super Sevillian Blue’.

Discussion

Some plant species express both the early C-13 hydroxylation pathway to biologically active GA1 and the non-hydroxylation pathway to biologically active GA4 (Hisamatsu et al., 1998; Nakayama et al., 1995; Niki et al., 2001). The identification of four endogenous GAs, GA1, GA19, GA20, and GA53, by GC/MS, in vegetative and reproductive tissues in L. × intermedia suggests that only the early C-13 hydroxylation pathway is functioning, and GA1 is the biologically active GA in L. × intermedia.

GA3 increased stem elongation with or without CT, but did not induce flowering without CT. There was no significant difference in florigenic effects among GA1, GA3, and GA5 (Tables 2 and 4). The reason might be that these GAs are produced in the same biosynthesis pathway, or that GA5 is metabolized to GA1 and GA3 (Koshioka et al., 1988), although GA3 and GA5 were not found in L. × intermedia. An artificial GA, 2,2-dimethyl-GA4 which was modified from GA4, had more florigenic activity than GA1, GA3, and GA5 in Lolium temulentum (Evans et al., 1990), and also induced stem growth and flowering more markedly than GA1, GA3, and GA4 in M. incana, (Hisamatsu et al., 2000). However, the florigenic activity of 2,2-dimethyl-GA4 was similar to those of GA1, GA3, and GA5 in L. × intermedia (Table 2). The reason for the different responses to GA among plants was not elucidated in this study. It might be caused by metabolic mechanisms or GA-receptors of 2,2-dimethyl-GA4 in the plants.

GA biosynthesis inhibitors suppressed RFP, RFS, and stem elongation. However, only UCZ significantly suppressed RFP in this study (Table 6). The suppression increased with the UCZ concentration. This suppression was abolished by GA3. This suggests that the expression of endogenous GA is necessary for flowering in L. × intermedia.

Flowering in L. × intermedia without GA was observed under at least several weeks of CT in each experiment, such as 8-week CT at 5°C (Tables 5 and 6), 9-week CT at 6°C (Table 3), 12-week CT at 7°C (Table 4), and 10-week CT at 8°C (Table 2). This period was defined as the minimum CT period. When the CT period became longer, RFP and RFS increased, and DVB and DFLW decreased at every chilling temperature. GA3 increased RFP and RFS and decreased DVB and DFLW in L. × intermedia under the minimum CT period. GA3 also induced flowering of plants under 6-week CT at 5°C, in which control plants did not flower. However, the florigenic effect of GA3 was not observed when the CT period became longer, such as 10-week CT at 5°C (Table 5), 12-week CT at 5°C (Table 6), 12-week CT at 6°C (Table 3), and 15-week CT at 7°C (Table 4). This longer period was defined as the maximum CT period. Both lengths of the minimum CT period and the maximum CT period became shorter when the chilling temperature decreased.

These findings indicate that flowering in L. × intermedia is influenced by both of chilling temperature and period of CT, and that GA3 enhances the effect of CT on flowering in L. × intermedia, but GA3 cannot replace for CT, and that responses to GA3 in L. × intermedia vary according to the CT period. In R. sativus, GA application induced flowering of plants that were not exposed to chilling temperature (Nishijima et al., 1997). Thus, the necessity of chilling temperature and GA for flowering varies depending on plant species. In some chilling requirement plants, levels of endogenous biologically active GAs and their precursors increase during CT and/or after CT as well as the responses to GA for stem elongation and flowering (Hisamatsu et al., 2000; Nishijima et al., 1997). In this study, the florigenic effect of GA3 varied with the CT period. However, we could not elucidate the reason for these differences. Therefore, studies on quantitative analysis of endogenous GAs and on expression of GA receptors may help to elucidate flowering in L. × intermedia.

Acknowledgements

We thank Mrs. K. Kashiwagi, K. Sone, and H. Iezumi and Misses S. Kamoi, N. Tsukahara, T. Numagami, Y. Saito, Y. Ogoshi, T. Tamano, and S. Yonezawa, for their technical assistance.

Literature Cited
 
© 2016 The Japanese Society for Horticultural Science (JSHS), All rights reserved.
feedback
Top