2025 Volume 94 Issue 1 Pages 33-39
Japanese pear [Pyrus pyrifolia (Burm. F.) Nakai] has a S-RNase-based self-incompatibility system and inactivation of S-RNase causes self-incompatibility breakdown in this system. Copper ions (Cu++) strongly inhibited the RNase activity of crude stylar proteins, but did not overcome self-incompatibility in the Japanese pear ‘Kosui’ (S4S5). To elucidate the cause of this phenomenon, the present study aimed to separate stylar RNases and clarify their responses to Cu++ in ‘Kosui’. Using an anion exchange chromatograph equipped with a Mono-Q column, one strong non-S-RNase was separated, and at least six non-S-RNases together with 2 S-RNases were isolated, by cation exchange chromatography with the Mono-S column. S-RNases were identified by dot blot analysis using S-RNase antiserum, and hereafter, the seven non-S-RNases isolated are designated as NS1 to NS7 in eluting order. NS1 and NS2 were abundant RNases with strong activities, and their activities were reduced to 6.6 and 4.5% of the control by 1 mM CuSO4, respectively. NS3, NS4, NS5, NS6, and NS7 were weak RNases, and NS7 showed intermediate inhibition (31%) with Cu++, whereas inhibition of other RNases was weak, from 44 to 74% of the control. Meanwhile, the degree of inhibition was quite low in S4- and S5-RNase (86 to 89%). Thus, RNase inhibition of crude stylar proteins by Cu++ is due to strong repression of NS1 and NS2 activities. These results indicate that active S-RNase can cause a self-incompatibility reaction despite large decreases in non-S-RNase activities in the pear style.
Since Japanese pear has S-RNase-based self-incompatibility systems (Sassa et al., 1992), artificial pollination and/or pollinizers are necessary to ensure fruit set except for several self-compatible cultivars, and excess fruit thinning is subsequently inevitable. Therefore, about 20% fruit set by self-pollination is desirable, because flowers used for fruit production account for only about 5% of all flowers. The pistils of plant species exhibiting S-RNase-based self-incompatibility produce T2-type RNase that is responsible for self-incompatibility, S-RNase, and other RNases, non-S-RNase. In Japanese pear, S-RNase is a basic protein with a molecular weight of about 30 kDa (Hiratsuka et al., 1995; Norioka et al., 2007; Sassa et al., 1992), while a non-S-RNase reported by Norioka et al. (2007) is an acidic protein with a molecular weight of about 25 kDa. S-RNase activity is essential for rejecting incompatible pollen (Huang et al., 1994; Kowyama et al., 1994), and this rejection ability appears to decrease with decreasing S-RNase activity (Qin et al., 2006).
It is well known that the activity of enzymes varies with the presence of organic compounds, heavy metals and substrate analogues, and the inhibition of various enzymes such as urease, aspartate aminotransferase, peroxidase, and catalase, by heavy metal ions has been studied to develop biosensors, medicines, and so on (Fopase et al., 2019; Jia et al., 2017; Moyo et al., 2014; Pan et al., 2016). In RNase inhibition, Maruyama et al. (2007) reported gold(III) ions to be a strong inhibitor against commercial RNase A, and Kim et al. (2001) found copper ions (Cu++) and zinc ions (Zn++) were effective in terms of stylar RNase inhibition in tomato. We also recognized that Cu++ and Zn++ strongly inhibit both stylar RNase of Japanese pear and commercial RNase A, and moderately inhibit commercial RNase S and T1 (Hayashida et al., 2013a).
CuSO4 and ZnSO4 strongly inhibited crude stylar RNase activity and caused about 20% fruit set when they were sprayed on ‘Kosui’ flowers before anthesis (Hayashida et al., 2016). However, the fruit was parthenocarpic and gibberellin paste treatment was necessary to produce fruit of a harvestable size (Hayashida et al., 2015). It is unclear why self-incompatibility is not overcome despite the large decrease in total RNase activity and why parthenocarpy is induced by heavy metal ions. RNases are widely distributed enzymes in organisms and play important roles in many cellular functions including RNA catalysis, defense against microorganisms, DNA replication, and control of gene expression (Macintosh, 2011). Although only one non-S-RNase has been reported in ‘Kosui’ styles (Norioka et al., 2007), we hypothesized that the style involves multiple non-S-RNases that are sensitive to heavy metal ions, whereas S-RNase is insensitive. This idea is based on the facts that RNases are widely distributed enzymes in plants and that some of them are inhibited by heavy metal ions as described above, while S-RNase activity is necessary to reject incompatible pollen.
Therefore, this study aimed to separate stylar RNases by ion exchange chromatography and clarify their responses to Cu++.
Adult trees of Japanese pear ‘Kosui’ (S4S5) and ‘Chojuro’ (S2S3) [Pyrus pyrifolia (Burm. F.) Nakai] planted at the Experimental Farm of Mie University, Tsu, Mie, Japan, were used.
Determination of fruit set and seed formation following Cu++ applicationThis experiment was conducted to reconfirm the Cu++ effects on fruit set and seed formation in ‘Kosui’. One week before anthesis, a branch possessing about 10 clusters with 3 to 4 buds was sprayed with 1 mM CuSO4 (Wako Pure Chemicals, Japan) solution containing 0.1% Tween-20. As controls (self-, cross-, and non-pollination), 0.1% Tween-20 solution was sprayed. Then, the branch was covered with a net bag to prevent contamination, and the bag was shaken to promote self-pollination at anthesis. In cross-pollination, ‘Chojuro’ pollen was supplied to each flower at anthesis by using cotton buds. Three branches were used for each treatment. Fruit set was checked four weeks after anthesis, fruit was harvested and seeds in the fruit were counted at the mature stage.
Preparation of stylar proteinsJust before anthesis, ‘Kosui’ styles were collected and proteins were extracted according to Hiratsuka et al. (1995) with slight modification. Briefly, styles were ground in liquid nitrogen using a mortar and pestle, and an extraction buffer [50 mM Tris-HCl buffer (pH 8.4) containing 150 mM NaCl, 10 mM L-cysteine, 1 mM ascorbic acid, and polyclar-AT (0.3 g·g−1 FW)] was added. After stirring the homogenate, it was kept for 30 min, centrifuged at 15,000 × g for 10 min, and then the supernatant was passed through a Sephadex G-25 column to remove polyphenols. After the proteins were saturated with ammonium sulphate at 100%, it was centrifuged again at 15,000 × g for 10 min to precipitate proteins. The resulting protein pellet was dissolved with a small amount of the Tris-HCl buffer (pH 8.4), dialyzed against the same buffer overnight, and subjected to an RNase assay as described below. Protein concentration was determined by the method of Bradford (1976).
Ion exchange chromatographyProteins precipitated with ammonium sulphate at 100% were used. To partially purify the samples, proteins salted out by 60% ammonium sulphate were discarded because no RNase activity was detected in this fraction (data not shown). Then, the 500 μg of resulting proteins dissolved in 50 mM sodium acetate buffer (pH 5.0) were subjected to HPLC analysis using apparatus equipped with a Mono-Q anion exchange column (5 mm × 50 mm; Sigma-Aldrich, USA). HPLC apparatus set up was as follows; pump = LC-10AT vp (Shimadzu, Japan), detector = SPD-10A vp (Shimadzu). After the proteins were applied to the column equilibrated with 50 mM sodium acetate buffer (pH 5.0), the column was washed with the same buffer, and adsorbed proteins were eluted with an NaCl gradient at a flow rate of 0.5 mL·min−1 at 25°C. The NaCl gradient was from 0 to 1.0 M during first 60 min and 1.0 M NaCl was continued until no protein peak was detected. Elution profiles of proteins were monitored at 280 nm and recorded. The elution was fractionated every 30 s (about 250 μL) and 140 fractions were obtained. Then, each fraction was subjected to an RNase assay, and RNase positive fractions were further subjected to dot blot analysis to identify S-RNases as described below. Separation by a Mono-S cation exchange column (5 mm × 50 mm; Sigma-Aldrich) was conducted using the same procedures as for the Mono-Q column except for the NaCl gradient. The gradient was from 0 to 0.2 M during first 10 min and a slow gradient (0.2–0.5 M) up to 75 min, then a rapid gradient (0.5–1.0 M) from 75 to 80 min, and finally 1.0 M NaCl was continued for 25 min. Elution was fractionated every 30 s (about 250 μL) and 180 fractions were obtained. The NaCl gradients used here were determined by checking the protein elution profile in preliminary experiments.
Determination of RNase activityRNase activity was determined by the modified method of Brown and Ho (1986). The reaction mixture contained 50 μg of crude stylar proteins or 45 μL of fractionated elution, 500 μL of 0.4% torula yeast RNA (Sigma-Aldrich) dissolved in 50 mM sodium acetate buffer (pH 5.0), and 10 mM KCl. The total volume was adjusted to 750 μL with the sodium acetate buffer. After incubating for 30 min at 55°C, 250 μL of 20% trichloroacetic acid was added to stop the reaction. The mixture was centrifuged at 15,000 × g for 10 min, and the supernatant was diluted 100-fold with distilled water. Then, the absorbance at 260 nm was measured using a UV-1800 spectrophotometer (Shimadzu). To determine the inhibitory effect of Cu++, a final concentration of 1 mM CuSO4 was added to the reaction mixture before incubation.
Dot blot analysis of fractionated RNasesThe objective of this analysis was to identify S4- and S5-RNase. Fractions containing RNase activity were reacted with S4-RNase antiserum which cross-reacts with S5-RNase (Hiratsuka et al., 2012). Briefly, onto the nitrocellulose membrane (Amersham, UK), 1.4 μg of fractionated protein was spotted, dried, and dipped in phosphate buffered saline (PBS) containing 18% H2O2 for 1 h to inactivate endogenous peroxidase. Then, the membrane was dipped in blocking solution (ECL Blocking agent; Amersham) with shaking for 1 h, and incubated in 8,000-fold diluted S-RNase antiserum for 1 h. After washing with PBS-T (PBS containing 0.1% Tween 20), the membrane was incubated in a secondary antiserum (peroxidase-linked anti-rabbit IgG; Amersham) solution. The membrane was then treated with a detection reagent (ECL Prime western blotting detection reagent; Amersham), exposed to Hyperfilm ECL, and the film was developed according to the manufacture’s protocol.
Statistical analysisData were calculated using Microsoft Excel (ver. 16.0.17726.20078), and standard errors (SE) were set at each data point. The significance of the data was analyzed by the Tukey-Kramer test at a 5% level. Regarding the % data, each value was transformed to arc-sign data using Excel before subjecting it to the Tukey-Kramer test.
Effects of CuSO4 on total RNase inhibition, fruit set, and seed number in the fruit are shown in Table 1. Addition of 1 mM CuSO4 to the reaction mixture reduced RNase activity to about 8% of the control. When 1 mM CuSO4 was sprayed on flowers one week before anthesis, about 17% fruit set was obtained. However, no seeds were observed in the harvested fruit, i.e. the fruit was parthenocarpic. Thus, we reconfirmed the function of CuSO4 on RNase inhibition, fruit set, and parthenocarpy induction in ‘Kosui’.
Effects of Cu++ on total RNase inhibition, fruit set, and seed number in ‘Kosui’.
The elution profile of stylar proteins through a Mono-Q anion exchange column is shown in Figure 1A, and RNase activity in each fraction is shown in Figure 1B. One strong RNase peak, NS1, was detected at 13 min and the approximate pI (isoelectric point) of this RNase was 4.5. The approximate pI was determined by comparing the locations of 11 pI markers (pI 3.5 to 9.3) with isoelectric-focusing polyacrylamide gel electrophoresis, to map the locations of major and specific stylar protein bands on the gel, including S-RNase (Hiratsuka et al., 2001), and generate the protein elution profile. Non-S-RNases isolated were designated NS1 to NS7 in eluting order through Mono-Q and Mono-S columns. The protein peak corresponding to NS1 was also extremely high. Although very faint activities were detected in several fractions from 24 to 55 min, they were not treated as RNases, because they were extremely minor with activities below 0.01.
(A) Elution profile of ‘Kosui’ stylar proteins through a Mono-Q anion exchange column, and (B) RNase activity in each eluted fraction. Five hundred micrograms of proteins were loaded on the column and the elution was fractionated every 30 s (about 250 μL). Protein levels were monitored by absorbance at 280 nm, and RNase activity was assayed by using 45 μL of fractionated elution as described in the text.
In cation exchange chromatography with a Mono-S column, many proteins were eluted from 7 to 50 min (Fig. 2A), and multiple RNases, NS2 to NS6, were present in these fractions (Fig. 2B). Two S-RNases and NS7 were eluted at about 80 min. NS2 possessed the strongest activity among the stylar RNases of ‘Kosui’ and NS3, NS4, NS5, and NS6 showed intermediate activities. In contrast, S4-, S5-RNases, and NS7 activities were much weaker. The approximate pIs of these RNases were as follows; NS2 = 5.7, NS3 = 5.9, NS4 = 6.0, NS5 = 6.1, NS6 = 6.2, S5-RNase = 9.4, S4-RNase = 9.6, NS7 = 9.7.
(A) Elution profile of ‘Kosui’ stylar proteins through a Mono-S cation exchange column, and (B) RNase activity in each eluted fraction. Five hundred micrograms of proteins were loaded on the column and the elution was fractionated every 30 s (about 250 μL). Protein levels were monitored by absorbance at 280 nm, and RNase activity was assayed by using 45 μL of fractionated elution as described in the text.
To identify S-RNase, RNase positive fractions were subjected to dot blot analysis using S-RNase antiserum (Fig. 3). NS5, NS6, and NS7 faintly reacted to S-RNase antiserum, and fractions at 76 and 79 min were strongly positive to the antiserum. Therefore, the fraction at 76 min was identified as S5-RNase, and the fraction at 79 min as S4-RNase respectively, because S4-RNase is a more basic protein than S5-RNase (Hiratsuka et al., 1995).
Dot blot analysis of fractionated RNases using S-RNase antiserum. Some of the RNase active fraction (1.4 μg protein) was reacted with S-RNase antiserum on a nitrocellulose membrane as described in the text. Black spot shows a positive reaction to the antiserum.
The degree of inhibition of each RNase by Cu++ is shown in Table 2, together with its retention time and response to S-RNase antiserum. NS1 and NS2 had very high activities that were strongly reduced by Cu++ to 6.6 and 4.5% of the control respectively, while the degree of inhibition of NS7 was intermediate (31%). Inhibition of NS3, NS4, NS5, and NS6 was weak, ranging from 44 to 73% of the control. On the contrary, S-RNase inhibition was quite low; S-RNase activity was maintained at more than 86% even with Cu++. Thus, Cu++ mainly inhibits NS1 and NS2 which are extremely abundant RNases in ‘Kosui’ styles.
Inhibitory effect of Cu++ on stylar RNase activity shown by separation by ion exchange chromatography.
In this study, we separated seven non-S-RNases, NS1 to NS7, by using anion and cation exchange chromatographs. In cation exchange chromatography, RNase peaks from 12–18 and 20–23 min (Fig. 2B) may also have been non-S-RNases, but we treated them as one RNase here because NS2 at 19 min had extremely high activity and protein separation was not well near this NS2 peak (Fig. 2A). Therefore, we considered that these fractions contained NS2 as a contaminant. However, when more precise NaCl gradient was set from 12–23 min, additional new non-S-RNases could be isolated in these fractions. Meanwhile, the NS5 and NS6 fractions showed a faintly positive reaction to S-RNase antiserum (Fig. 3), suggesting that these RNases have epitopes that reacted with S-RNase antiserum. However, since these fractions probably contained proteins other than RNase, the antiserum may have reacted with these contaminants. Although the NS7 fraction also faintly reacted with the S-RNase antiserum, this may have been due to contamination of S4-RNase, because the S4-RNase fraction had an extremely strong reaction to the antiserum and NS7 had a very similar retention time to this S4-RNase. Although stylar RNases were adsorbed on ion exchange columns at pH 5.0, uncharged and very weakly charged RNases could not adsorb on the columns at this pH: therefore, they were not reflected in the present results. For definite identification of non-S-RNases in ‘Kosui’ styles, more precise setting of experimental conditions is necessary, including the NaCl gradient and pH used for protein adsorption.
To date, one non-S-RNase has been reported in ‘Kosui’ styles (Norioka et al., 2007). They purified and characterized seven basic S-RNases, S1- to S7-RNase, and one acidic non-S-RNase, by using CM-cellulose, Phenyl Superose, Hydroxyapatite, and Mono-S column chromatographs. The NS1 detected in this study may have been this acidic non-S-RNase, because the RNase is acidic protein that has no affinity for the cation exchanger (Norioka et al., 2007). However, if the pI of this acidic RNase was near 5.0, it would not have been adsorbed by the anion exchanger used in this study, and in this case, NS1 would be another acidic RNase in ‘Kosui’ styles. Regardless, NS1 activity was restricted to 6.6% of the control by adding Cu++, and NS2, which had the strongest activity among the stylar RNases detected, also showed decreased activity to just 4% due to Cu++. Since activity of NS1 plus NS2 may account for the majority of total stylar RNases, strong inhibition of total RNase activity by Cu++ (Table 1) is likely due to repression of these two enzymes. However, since S4- and S5-RNase were almost insensitive to Cu++, the self-incompatibility system would be functional. This result indicates that active S-RNase can cause a self-incompatibility reaction despite a strong reduction in non-S-RNase activities in the style.
Enzyme activity is a function of active site structure where the substrate interacts with the enzyme. When the enzyme has copper-binding domains, Cu++ can bind with the domain and change the enzyme activity by causing conformational alteration of the protein (Irie, 1999; Smith et al., 2004). Jia et al. (2017) constructed a copper-binding proteome of rat livers and detected 97 putative Cu-binding proteins including aspartate aminotransferase, malate dehydrogenase, catalase and so on. They also reported that Cu-binding aspartate aminotransferase activity decreased greatly, while Cu-binding malate dehydrogenase and catalase activities decreased weakly. Therefore, if Cu++ can bind to the enzyme, the degree of inhibition will be different among enzymes; conformational changes may be different depending on the domain sites in the peptide sequence. The Cu-binding domains contain histidine, methionine, and cysteine in peptide sequences (Smith et al., 2004), and S-RNases of ‘Kosui’ have these domains (Norioka et al., 2007). Therefore, Cu++ can bind with the S4- and S5-RNase of ‘Kosui’, but sufficient conformational changes may not be induced in these enzymes; the active site of the enzyme will not be fully blocked. Since the Cu-protein complex can enter cells and further affects intercellular enzyme activities, copper-binding has a strong influence on cellular metabolism (Jia et al., 2017).
We used 1 mM CuSO4 to both spray flower clusters and for the RNase assay in vitro, because thicker CuSO4 (> 3 mM) caused chemical injuries to flower buds and thinner CuSO4 (< 0.5 mM) did not show enough inhibition of stylar RNase activity (Hayashida et al., 2013a). When Cu++ was measured in the pistil one week after spraying 1 mM CuSO4 (63 μg Cu++·mL−1), absorbed Cu++ was about 10 μg·g−1 FW (0.16 mM CuSO4) (Hayashida et al., 2013b), while 0.16 mM CuSO4 reduced stylar RNase activity to only about 90% of the control (Hayashida et al., 2013a). Thus, the functional Cu++ concentration was different between parthenocarpy induction in vivo and RNase inhibition in vitro in this study. Although the detailed mechanism is unclear, RNase inhibition may not have any connection with parthenocarpy induction, and the effective concentration of Cu++ to induce parthenocarpy may be around 10 μg·g−1 FW in the pistil.
The relation of Cu++ to parthenocarpy induction is still unclear. We expected that Cu++ restricts ethylene production by inhibiting ACC synthase, because exogeneous application of an ethylene generator (ethephon) promoted fruit drop and an ethylene repressor (AVG; aminoethoxyvinyl glycine) caused parthenocarpy of ‘Kosui’ (Fig. S1). However, Cu++ application before anthesis did not repress ethylene production from flowers (Fig. S2). Therefore, parthenocarpy induction by Cu++ may not be mediated through ethylene metabolism in pears, but by other plant hormones such as auxins, gibberellins, and cytokinins as reported in other plants (Ogawa and Takisawa, 2022). Since Cu++ can bind with many kinds of enzymes and control their activities (Jia et al., 2017; Smith et al., 2004), it is possible that Cu++ binds with enzymes involving the biosynthetic pathways of auxins, gibberellins, and/or cytokinins, affecting their metabolism.
The final objective of this study was to establish a cultivation system without the need for artificial pollination and fruit thinning practices, allowing a major reduction in labor required by pear growers. To achieve 20% seeded fruit setting in ‘Kosui’, partial breakdown of self-incompatibility is desirable and screening and/or developing inhibitors against S-RNase activity may be useful. Our study indicates that crude stylar protein cannot be used to screen S-RNase inhibitors, and that development of a simple purification method for S-RNase and/or construction of S-RNase analogs are necessary to further study partial breakdown of self-incompatibility in Japanese pear.
ConclusionThe styles of ‘Kosui’ contain at least seven non-S-RNases together with S4- and S5-RNase, and the activities of two strong non-S-RNases were extremely inhibited by Cu++, but S-RNases were almost insensitive to the Cu ions. Thus, active S-RNase can cause a self-incompatibility reaction despite a great decrease in non-S-RNase activities in the style. To induce partial setting of seeded fruit in ‘Kosui’, screening and/or developing inhibitors against S-RNase activity may be useful. Since about 60% inhibition of S-RNase activity may be necessary to ensure 20–30% fruit set (Qin et al., 2006), an efficient inhibitor against pear S-RNase is required to establish a cultivation system without the need for artificial pollination and excess fruit thinning. The present data, together with conformational data on S-RNases (Norioka et al., 2007), contribute not only to the progress of sequential studies in this field, but also towards the establishment of labor-saving cultivation systems for the Japanese pear ‘Kosui’.
