Phaseolus vulgaris, L., L.

Enomoto, Hirofumi & Miyamoto, Koji, 2021, Unique localization of jasmonic acid-related compounds in developing Phaseolus vulgaris L. (common bean) seeds revealed through desorption electrospray ionization-mass spectrometry imaging, Phytochemistry (112812) 188, pp. 1-9 : 3-5

2.4. DESI-MS/MSI analysis of developing P. vulgaris seeds

DESI-MS/MSI analysis was performed to estimate the contamination of OPDA and OPC-8:0 isomers based on their ion images obtained through DESI-MSI analysis. The fragment ion generated by dehydration, [M–H 3 O] –, was not used for reconstructing the ion images because the fragment ion was possibly detected not only from analytes but also from their isomers ( Fig. 2a and b ). Fig. 5a shows the optical image of the section of a developing seed used for OPDA analysis. In the MS/MS spectrum of the precursor ion at m/z 291.1966 obtained at the target-enhanced mode at m/z 165.1, a fragment ion at m/z 165.1309, corresponding to the specific fragment ion of the OPDA standard [M–C 7 H 11 O 2] –, was detected ( Figs. 2a and 5b ). The fragment ion was mainly localized in the seed coat and consistent with the ion image at m/ z 291.1953 obtained by DESI-MSI analysis ( Figs. 1d and 5c ). This result suggested that accurate localization of OPDA in developing P. vulgaris seeds can be determined through DESI-MSI analysis. This was consistent with the results of our previous study ( Enomoto et al., 2017), suggesting that the effect of the isomer on the ion image of OPDA was small.

Fig. 5d shows an optical image of the section of a developing seed used for OPC-8 :0 analysis. In the MS/MS spectrum for the precursor ion at m/z 293.2122 obtained at the target-enhanced mode at m/z 225.1, specific fragment ions for OPC-8 :0 corresponding to [M– C 5 H 9 ] – and [M– CHO 2 ] – ions were not detected ( Figs. 2b and 5e ). Moreover, fragment ions at m/z 223.1400 and 231.2142 corresponding to the characteristic fragment ions of the isomers were detected ( Fig. 4g and h , 5e ). They showed localization patterns similar to the ion image at m/z 293.2117 obtained by DESI-MSI analysis ( Figs. 1e and 5f, g ), indicating that the ion image at m/z 293.2117 obtained by DESI-MSI analysis was affected by the ion images of the isomers ( Fig. 1e ). Furthermore, these results suggest that DESI-MS/MSI analysis using specific fragment ions (i.e., [M–C 5 H 9] – and [M–CHO 2] –) is necessary for the accurate visualization of OPC-8:0 in the radicle and seed coat of developing P. vulgaris seeds .

In the present study, we identified and visualized three JA-related compounds, namely free αLA, OPDA, and OPC-8:0, in developing P. vulgaris seeds using DESI-MSI and LC-ESI-MS analyses. The results showed the localization and abundance of these free JA-related compounds in developing P. vulgaris seeds . To the best of our knowledge, this is the first report of the unique localization of free JA-related compounds in developing P. vulgaris seeds .

Seed germination and maturation depend on the breakdown of stored triacylglycerols, which are metabolized to provide carbon and energy ( Yu et al., 2014). At the beginning of lipid breakdown, the main storage lipids are first hydrolyzed to free fatty acids by lipases. Long-chain acyl-CoA synthetase converts free fatty acids into acyl-CoA thioesters that serve as substrates for β- oxidation ( Watkins and Ellis, 2012). Hence, accumulated free αLA in the radicle and seed coat of developing P. vulgaris seeds likely plays a role in providing carbon and energy for seed maturation and germination. A 4-coumarate:CoA ligase-like protein, OPC-8:CoA ligase 1, activates OPC-8:0 to its CoA ester, which then undergoes β- oxidations to form JA ( Wasternack et al., 2013, 2018). The presence of JA in the radicle and seed coat indicates that β- oxidation is active in these tissues ( Table 3 ). Therefore, it was speculated that the long-chain acyl-CoA synthetase and OPC-8:CoA ligase 1 activities are low in the radicle and seed coat of developing P. vulgaris seeds .

Dave and Graham. (2012; 2011) demonstrated that OPDA acts alongside abscisic acid to regulate seed germination in Arabidopsis thaliana . In another study ( Dave et al., 2016), they showed the central role of OPDA in regulating seed dormancy and germination, thus underlining the complexity of the interactions between OPDA and other dormancy-promoting factors, such as abscisic acid, RGL2, and MOTHER OF FT AND TFL. In this study, OPDA was found to be an abundant JA-related oxylipin in the seed coat of developing P. vulgaris seeds ( Figs. 1d and 5c ; Table 3 ), consistent with our previous studies ( Enomoto et al., 2017, 2018b). Goetz et al. (2012) reported that OPDA was the dominant oxylipin, occurring nearly exclusively in the seed coat of developing tomato seeds. Therefore, we speculated that OPDA is localized in the seed coat of developing P. vulgaris seeds and plays an important biological role in seed dormancy and germination.

Using a mutant tomato (spr2) which is OPDA- and JA-deficient, and another mutant (acx1a) which preferentially accumulates OPDA and residual amounts of JA, Goetz et al. (2012) further suggested that OPDA or an OPDA-related compound might play a role in proper embryo development, possibly by regulating carbohydrate supply and detoxification. We found that both OPDA and OPC-8:0 were dominant JA-related oxylipins in the radicle or seed coat of developing P. vulgaris seeds ( Table 3 ). The only structural difference between OPDA and OPC-8:0 is the lack of one carbon–carbon double bond in the pentacyclic ring of the latter ( Fig. 2a and b ) ( Bao et al., 2014). Therefore, OPC-8:0 might be an OPDA-related compound that regulates carbohydrate supply and detoxification for proper embryo development in developing tomato and P. vulgaris seeds . Interestingly, the localization pattern of OPDA was similar to that of JA and differed from that of OPC-8:0 ( Table 3 ). In addition, compared to the concentrations of OPDA, the concentration of OPC-8:0 was 5.9 times lower in the seed coat and 2.5 times higher in the radicle ( Table 3 ). These results suggest that OPDA and/or JA play a biological role mainly in the seed coat, while OPC-8:0 is biologically active mainly in the radicle. Further analysis using mutants is needed to elucidate the biological significance of OPC-8:0 in developing seeds.

The effects of JA-related oxylipins on mammals have attracted much attention because of their structural similarities to mammalian oxylipins such as prostaglandins and leukotrienes. Known in vitro and in vivo effects of JA-related oxylipins include anti-inflammatory and anti-cancer effects, and anti-aging properties in the human skin ( Henriet et al., 2017; Taki-Nakano et al., 2016). Therefore, the seed coat of developing

P. vulgaris seeds may be a promising source of these JA-related oxylipins.

Figures

Fig. 2. LC-ESI-MS/MS analysis of JA-related compound standards. Spectra of (a) OPDA, (b) OPC-8:0, and (c) JA standards. Compound names are defined in Table 1.

Fig. 5. DESI-MS/MSI of OPDA and OPC-8 in the developing Phaseolus vulgaris seeds. (a) Optical image of the seed section for OPDA analysis. (b) MS/MS spectrum of precursor ion at m/z 291.1966 ±1 Da obtained at the target enhanced mode for m/z 165.1. (c) Ion image at m/z 165.1300. (d) Optical image of the seed section for OPC-8:0 analysis. (e) MS/MS spectrum of precursor ion at m/z 293.2122 ±1 Da obtained at the target enhanced mode for m/z 225.1. Ion images at m/z (f) 223.1400 and (g) 231.2142. Scale bar = 2 mm. Compound names are defined in Table 1.

Fig. 1. DESI-MSI analysis of JA-related compounds in the developing Phaseolus vulgaris seeds. (a) Optical image of the section. (b) Mass spectrum obtained from the section. Ion images of m/z (c) 277.2172, (d) 291.1953, and (e) 293.2117. Three different developing seeds were analyzed, and the results from one are shown as representative data. Scale bar = 2 mm. Compound names are defined in Table 1.

Fig. 4. LC-ESI-MS/MS analysis of JA-related compounds in the extracts from the radicle and seed coat of developing Phaseolus vulgaris seeds. MS/MS spectra of peaks at (a) 5.3 min in Fig. 3c, (b) 5.3 min in Fig. 3d, (c) 6.5 min in Fig. 3c and (d) 6.5 min in Fig. 3d and (e) 6.3 min in Fig. 3e and (f) 6.3 min in Fig. 3f and (g) 6.4 min in Fig. 3e, (h) 6.4 min in Fig. 3f, (i) 6.7 min in Fig. 3e, and (j) 6.7 min in Fig. 3f. Compound names are defined in Table 1.