In the present investigation, transparent glass bottles, 62 mm in diameter, 110 mm in height and 48 mm in bottle mouth diameter (Figure 1), served throughout for culturing seedlings and performing experiments in cap-loose and cap-tight modes. The cap-loose mode denotes the conditions that an ordinary plastic cap was screwed loosely on the bottle mouth so as to allow air flow freely, and the cap-tight mode means that the ordinary cap was replaced by another type of a cap equipped with inlet and outlet tubes and other necessities, and screwed tightly so as not to allow air to pass (Figure 1).

In assembling the cap for the cap-tight mode, special care was spent to prevent air leak through the cap assembly. The contact surfaces between the packing sheet (c) and the Teflon bolt (d) and between the Teflon bolt (d) and the inlet- and outlet-tubes (f) were pasted with fluorochemical lubricating grease (DEMUNUM GREASE, Daikin Industries, Ltd., Osaka, Japan). When screwing the cap (b) to the bottle (a), the bottle mouth edge was moistened with a drop of water to secure airtightness. Bottles thus prepared were examined for ethylene preservation for 4 days, and only cap-bottle combinations which preserved ethylene more than 95% were used.

Seeds of tomato (Solanum lycopersicum L.), cvs. Ponte-Rosa, Sekaiichi and Seikoh No.17, were purchased, respectively, from Ishihara Seed Co. (Sakai, Japan), Marutane Seed Co. (Kyoto, Japan) and Watanabe Seed Co. (Misatomachi, Miyagi, Japan). The three cultivars behaved similarly in phytochrome-mediated hook exaggeration (Shichijo et al., 2010a,b; Shichijo and Hashimoto, 2013), and were used arbitrarily in the present experiments. Seeds were sterilized for 10–20 min with 25-times diluted solution of commercial bleach, followed by thorough washing with running tap water for 30–90 min, and soaked in sterilized distilled water until full imbibition. Thirty to 40 seeds thus treated were germinated on a layer of 60-mm filter paper moistened with 3.5 or 4 ml of distilled water in culture bottles (Figure 1) and seedlings were grown at 25.5 ± 0.5°C throughout. In experiments where ethylene was quantified, the same number of seeds were sown in each culture bottle. Reagent solutions were applied to the filter paper instead of distilled water before sowing or sprayed to seedlings after grown for 5 days in culture bottles.

The red light (R) sources used were R_LED (λ_max, 658 nm; half-bandwidth, 26 nm) and R_FL (λ_max, 660 nm; half-bandwidth, 20 nm); far-red light (FR) sources were FR_LED (λ_max, 746 nm; half-bandwidth, 28 nm) and FR_FL (λ_max, 766 nm; half-bandwidth, 72 nm). The details of these light sources were already described (Shichijo et al., 1993, 2001). Photon fluence rate was determined with an Optical-Power-Meter (TQ8210; Advantest, Tokyo, Japan). Irradiations were made from above, unilaterally or bilaterally. Since a single pulse of R or FR is enough to cause hook exaggerations, mediated by very low and low fluence responses of phytochrome, and allows free selection of an appropriate time-point for irradiation according to its purpose, a single pulse irradiation has been adopted as the standard, and where needed, the irradiation period was extended accordingly.

Ethylene gas (99.5%) was purchased from GL Sciences, Inc. Japan. 1-Aminocyclopropane-1-carboxylic acid (ACC) and aminoethoxyvinylglycine (AVG) were from Sigma–Aldrich through Hirose kagaku, Ltd, Kobe, Japan; cobalt chloride (CoCl₂) and N-(1-naphthyl)phthalamic acid (NPA), from Wako Pure Chemical Industries, Osaka, Japan and Tokyo Chemical Industry, Tokyo, Japan, respectively. Ethylene was diluted with air, and the other reagents were dissolved in distilled water.

Hook curvature was quantified by the angle formed by extended lines of the main hypocotyl axis and the apical axis, defining as zero degree when the apical hook is straightened up. The hypocotyl height designates the length from the highest point of hook to the base of the hypocotyl. These are in accordance with Shichijo et al. (2010a).

For applying and quantifying ethylene, culture bottles in the cap-tight mode were used (Figure 1). When growing seedlings in culture bottles of the cap-tight mode, the evolving CO₂ was absorbed by the CO₂-absorbent container (h in Figure 1) where 1.5 ml of 1 M NaOH is added to 0.1 g absorbent cotton.

To fill a definite concentration of ethylene in a culture bottle of the cap-tight mode, 1 ml of prescribed higher concentrations of ethylene/air mixture was injected so as to result in the required concentration when it dispersed within the bottle. A 1-ml empty syringe with a needle (od = 0.5 mm) was inserted into the outlet tubes (f in Figure 1) as the air receiver, and another same sized syringe filled with the prescribed concentration of ethylene/air mixture was similarly inserted in the inlet tube (f in Figure 1) as the donor syringe. As the receiver syringe sucked up air from the bottle, the donor syringe pressed the ethylene/air mixture into the bottle. After the ethylene injection was completed, the outlet and inlet tubes were closed with stopping pins (i in Figure 1).

To determine the ethylene concentrations in culture bottles, gas-sampling was made similarly by concerted motions of two syringes, one of which was filled with ethylene-free air of the same volume as that of the sample air to be taken with the other empty syringe. In this way, sample air of 1 or 2 ml was taken out and injected into a gas chromatograph (GC-14A; Shimadzu Co., Kyoto) equipped with a 100 cm × 0.26 cm active alumina (60–80 mesh) column (Gasukuro Kogyo Inc., Tokyo, Japan) and a flame ionization detector. In some experiments (Figures 7 and 8) ethylene emission rate was calculated with the equation (ethylene concentration × bottle capacity)/(Fresh weight of seedlings × incubation period).

Statistical significance was determined by the post hoc analysis, especially Tukey’s test, after one-way ANOVA by means of the software KaleidaGraph (Synergy Software, Inc., Reading, PA, USA), and has been indicated by asterisks, pound marks or alphabets in the vicinity of relevant data points or on the top of histogram bars to be compared in Figures.

To examine whether ethylene causes the exaggeration of the apical hook instead of light, dark-grown 5-day-old tomato seedlings were exposed to various concentrations of ambient ethylene in culture bottles in the dark. As the standard of hook exaggerations mediated by very low and low fluence responses as well as high irradiance response of phytochrome (Shichijo et al., 2010a), seedlings were, without addition of exogenous ethylene, irradiated with continuous red (Rc) and far-red light (FRc) in place of pulsed light. In parallel, effects of ethylene on hook under such irradiated conditions were also examined (Figure 2A). In the present study, such low concentrations of ethylene as endogenous ethylene also matter, the concentrations in the culture bottles were determined at the end of experiments and used as ambient ethylene concentration to plot the hook response in this Figure and others. In the dark, ethylene suppressed (i.e., opened) hook curvature in the neighborhood of 1.0 μL L^-1, and as its concentration increased beyond 1.0 μL L^-1, the gas enhanced (i.e., closed) the hook, but even at 10 μL L^-1, it could not intensify the curvature to the level of the hooks exaggerated by either Rc or FRc in the absence of supplementary ethylene (left end of the curves). As seen in the photos at 10 μL L^-1 (Figure 2C), the toxicity of high concentration ethylene, i.e., the severe growth inhibition of the hypocotyl and the swelling at the part just below the hook, implies that no further enhancement of the hook was possible even by further raising ethylene concentration.

The hook exaggerated by Rc was suppressed by ethylene more markedly than the hook suppression in the dark. Similar suppression was also observed under FRc to lesser extents (Figure 2A). In both cases under light the enhancement of the hook angle at the higher ethylene concentration ranging from 3.0 to 10 μL L^-1 was small in ratios compared with that observed in the dark. The suppression of hypocotyl height (Figure 2B) as well as the appearance of the seedlings (Figure 2C) shows that the ambient ethylene worked normally.

Endogenous ethylene may give some different effect by evolving differently among the target tissues of seedlings; for example, between the concave and convex sides of the hook. To examine this possibility, seedlings were sprayed with various concentrations of ACC, the immediate biosynthetic precursor of ethylene, and kept in the dark for 72 h. To minimize the accumulation of emitted ethylene, the bottles were kept in the cap-loose mode. As the standard for comparison of the expected effects of ACC, Rp or FRp was given to induce hook exaggeration not only in the absence of ACC, but also in the presence (Figure 3A). The hook angles presented themselves comparatively small due to the pulse of light here compared with those of Rc and FRc in Figure 2A. All the three sorts of hook curvatures including the one kept in the dark throughout were similarly suppressed by ACC concentration-dependently, and excluded the above-stated assumption that endogenously evolving ethylene may cause hook exaggeration instead of light. The effects of the ACC on hypocotyl height (Figure 3B) did not differentiate the treatments Dark, Rp and FRp, supporting that ethylene evolved from ACC equally regardless of the light pulse given and operated normally in this test system.

Although it was suggested in Figure 3B that R and FR do not affect ethylene evolution from ACC, it was indirect proof with a single pulse of light. Accordingly, the same question was examined by direct quantification of ethylene evolved from ACC under Rc, FRc and in darkness. How high a concentration the evolved ethylene reached in the bottles was another important question to be answered.

Seedlings were sprayed with ACC solution or water and cultured in the cap-tight mode for 48 h with or without R_{24 h} or FR_{24 h}. To magnify the effects of light, the irradiation continued for 24 h starting at the ACC spray, and as the result, marked light-induced hook exaggeration (LIHE) and hypocotyl growth inhibition were caused (Figures 4A,B). But the ethylene evolution did not differ among the above three conditions, giving the almost identical final concentrations in the neighborhood of 1.0 μL L^-1 (Figure 4C). The concentration range corresponded with that which suppressed the hook curvatures most effectively not only in the dark, but also under Rc or FRc (Figure 2).

Thus, the possibility is excluded that endogenous ethylene evolving from its immediate precursor ACC by a possible light action might cause LIHE. It was also confirmed that ethylene suppresses hook curvature in the neighborhood of 1.0 μL L^-1.

What will happen to hook curvatures, when endogenous ethylene level is reduced below normal? To reduce endogenous ethylene level, the ethylene biosynthesis inhibitor CoCl₂ was firstly tested. Seedlings were grown from seeds on CoCl₂ solutions of various concentrations including null concentration in the cap-tight mode for 5 days, and an Rp or FRp was given to induce hook exaggeration (Figure 5A). As the result, hook curvatures of the irradiated and non-irradiated seedlings were equally enhanced by the inhibitor at 500 μM, and the ethylene levels were conversely suppressed at the same concentration. There was no significant difference in the reduced ethylene levels among the three different light treatments (Figure 5B).

Secondly, another ethylene biosynthesis inhibitor AVG was sprayed to seedlings at null, 10, 100, and 500 μM. At 100 and 500 μM it enhanced the hook curvatures under the three light conditions to similar extents (data not shown). Next, seedlings were grown from seeds on 100 μM AVG for 5 days, and then supplemented with various concentrations of ethylene and irradiated with Rp or FRp, or not irradiated, followed by culture for 48 h (Figure 6A). Another similar experiment made without AVG gave the results as shown in Figure 6B. In both Figures the left ends of the curves indicate the concentration of only ethylene emitted by seedlings. Comparison of the left ends of corresponding curves between the two Figures indicates that the AVG treatment reduced endogenous ethylene from about 0.2 μL L^-1 (also seen in Figure 2A) to about 0.08 μL L^-1. For easy comparison of the effect of endogenous ethylene at reduced concentrations, excerpts from the three curves in Figure 6A are overlaid as the thick lines on the corresponding curves in Figure 6B. The excerpts fell, respectively, on the extensions of the corresponding curves obtained in the absence of AVG. Thus it is clear that AVG suppressed ethylene evolution, and, in turn, resulted in the enhancement of the hook curvatures, whether exaggerated or non-exaggerated, at the low concentration range below 1.0 μL L^-1, and the enhancement was reversed by supplementing ethylene. These data confirm that ethylene suppresses all of the three kinds of hook curvature at the low concentration range.

At the high ethylene concentration range above about 1.0 μL L^-1, on the other hand, where it tended to enhance the hook curvatures, the hook enhancement is obviously greater when fed with AVG (Figure 6A, compared with Figure 6B, also Figure 2A). This tendency seems more marked when irradiated with Rp or FRp than when kept in the dark, shifting the valleys of the curves for Rp and FRp to a lower ethylene concentration. Since the action of AVG as ethylene synthesis inhibitor is canceled by the supplemented ethylene, the hook enhancement found here is considered to be due to another unknown action of AVG.

Literature shows that an apical hook is formed by lateral localization of auxin at the apical part of the hypocotyl or epicotyl, and ethylene exerts its hook-exaggerating action through enhancing the localization of auxin (Vandenbussche et al., 2010; Žádníková et al., 2010, 2015; Abbas et al., 2013; Mazzella et al., 2014; Willige and Chory, 2015). It has also been known that auxin induces ethylene evolution in tissues (Kang et al., 1967; Kang and Ray, 1969; Yip et al., 1992; Schwark and Bopp, 1993; Abel et al., 1995; Peck and Kende, 1995; Coenen et al., 2003; Tsuchisaka and Theologis, 2004). Hence the present authors assumed that LIHE may also involve auxin localization intensified by light. To examine this possibility, effects of the auxin polar transport inhibitor, NPA, on hook curvature and ethylene emission rate were examined.

As shown in Figure 7, the inhibitor suppressed markedly and uniformly the hook curvatures in all lots, irrespective of Rp, FRp or no pulse, at the concentrations over 1 μM, half inhibition being at 5 μM, and at 30 μM the hooks were almost straightened up (Figure 7A). The emission rate of ethylene, however, was not affected at all the concentrations tested (Figure 7B). These results exclude the possibility that ethylene evolution is controlled by auxin and regulates LIHE in tomato. At the same time this experiment with NPA suggests that LIHE is controlled by localization of auxin, as already shown with the apical hook curvature formed in the dark (Abbas et al., 2013; Mazzella et al., 2014; Willige and Chory, 2015; Žádníková et al., 2015).

If LIHE involves any variation of endogenous ethylene, the emission rate of ethylene by seedlings should be altered by R and FR. Some negative results to this assumption have already been described (Figures 4 and 5), but with more extensive coverage of irradiation periods it was examined whether ethylene emission rate is affected by R and FR according to the enhancement of hook curvatures (Figures 8A–D). Since the Figures represent the experiments on different dates, comparison of absolute values is possible only within each Figure, but for comparison among the Figures their ratios in each Figure should be used. Although the hook curvatures increased with the increasing duration of irradiation, the ethylene levels did not show any appreciable change, indicating that the emission rate of ethylene was not altered by the irradiations. This result is another piece of negative evidence against the possible involvement of ethylene as a limiting factor of LIHE.