High-resolution mass spectrometry for detecting Acetylcholine in Arabidopsis

Abstract

Acetylcholine (ACh) was first identified a century ago, and has long been known as a neurotransmitter in animals. However, it has been shown recently that the occurrence of ACh is widespread among various non-animal species including higher plants. Although previous reports suggest that various plant species are capable of responding to exogenously applied ACh, the molecular basis for ACh biosynthesis and regulatory mechanisms mediated by endogenous ACh are largely unclear. This is partly because of the lack of conclusive data on the occurrence and the tissue specificity of ACh in plants. To this end, we performed various analyses including liquid chromatography electro-chemical detection (LC-ECD), liquid chromatography tandem mass spectrometry (LC-MS/MS) analysis. The results, together with electrospray ionization-orbitrap Fourier transform mass spectrometry (ESI-orbitrap FT-MS) analysis provide strong evidence that ACh exists in Arabidopsis thaliana tissues. The results also showed that the level of ACh is highest in seed, followed by root and cotyledon. Moreover, exogenously applied ACh inhibited the elongation of Arabidopsis root hairs. These results collectively indicate that ACh exists primarily in seed and root in Arabidopsis seedlings, and plays a pivotal role during the initial stages of seedling development by controlling root hair elongation in Arabidopsis.

Introduction

The biological role of acetylcholine (ACh) as one of the major neurotransmitters in the central and the peripheral nervous system in animals has been established for over a century, and the molecular basis of the biosynthesis, hydrolysis, and recognition of ACh has been well studied. However, recent reports show that the occurrence of ACh is not limited to the central and the peripheral nervous system in animals, but is widespread among various non-neuronal tissues in animals as well as various organisms including bacteria and plants that lack apparent neuronal networks.^1,2 For example, the accumulation of ACh was detected in rat ventricular myocardium and mouse organoids that are derived from epithelial cells of small intestine, both of which do not seem to be directly associated with neuronal cells.^3,4 In the former case, ACh was shown to be biosynthesized in situ by a mechanism that is positively regulated by cholinergic stimuli, suggesting that ACh contributes to the amplification of cholinergic signals in rat ventricular myocardium.³ Likewise, mouse intestine organoids are capable of biosynthesizing ACh and undergo self-regeneration processes governed by intestinal stem cells, the viability of which is affected by cholinergic stimuli.⁴ These results clearly indicate that ACh has biological roles beyond a neurotransmitter in mammals. Moreover, in the case of a cnidarian Hydra magnipapillata, the expression of ACh esterase (AChE) gene was localized to the ectodermal and endodermal epithelial cells except for the tips in the head or the foot end where the rate of tissue regeneration was significantly higher than other parts of the body.⁵ Given that exogenous application of AChE inhibitors to Hydra has been shown to inhibit its regeneration, the results suggest that ACh has an indispensable role in mediating the regeneration of Hydra in the outmost layer of the body. Emerging evidence in these reports suggests that ACh exhibits various novel biological activities in non-neuronal tissues and organisms other than as a neurotransmitter.

In the case of plants, accumulation of ACh and the enzymatic activities that biosynthesize and hydrolyze ACh have been detected in various plant species.^1,6 Moreover, previous pharmacological studies indicate that ACh seems to be involved in various physiological processes including, but not limited to, the control of stomatal aperture, gravitropism, and root development.^7,8,9 However, despite the general occurrence of ACh across kingdoms, genes that are orthologous to known animal ChAT, AChE, or genes encoding muscarinic or nicotinic ACh receptors have not been successfully identified from plants or bacteria to date. Instead, AChE was purified to homogeneity from Zea mays and Salicornia europaea, and the corresponding genes were identified as members of the GDS(L) esterase/lipase gene family, which shows no obvious sequence homology to animal AChE.^10,11 Nevertheless, it is still unclear whether plants possess common AChE genes, since the orthologous AChE gene products in Arabidopsis did not exhibit apparent activity of AChE.¹² Moreover, there do not seem to be conclusive evidences on the existence of ACh in plant tissues, since the detection of ACh from plants has been performed primarily by radioimmunoassays using antibodies raised against ACh or high-performance liquid chromatography with electrochemical detection (HPLC-ECD), both of which detect ACh indirectly.¹ Alternatively, ACh has been detected by gas chromatography mass spectrometry (GC-MS) on ACh that was derivatized into an N-demethylated form, which provide indirect evidence for the existence of ACh as well.¹³ Recognition of small molecules by radioimmunoassays solely depends on the specificity of epitope-antibody reaction, while HPLC-ECD detects H₂O₂ that is generated by choline oxidase-immobilized column. Although the methods for detecing acetylcholine by mass spectrometry has been improved considerably in their specificity as well as sensitivity,¹⁴ detection of ACh from plant tissues would require further improvements because the level of ACh in plants is generally much lower than in animals.¹

These reports collectively demonstrate that, although plants seem to be capable of responding to ACh, the molecular basis of how plants biosynthesize, sense, and degrade ACh is poorly understood. The reports also show that the existence of ACh in higher plants still needs to be carefully examined with alternative techniques with higher resolution and accuracy. In this study, we identified ACh from Arabidopsis seed through combined use of various mass spectrometers at high resolution and accuracy. The analysis by electrospray ionization orbitrap Fourier transform mass spectrometry (ESI orbitrap FT-MS) as well as liquid chromatography triple-quadruple mass spectrometry (LC-MS/MS) confirmed that ACh does exist in plants.

Results

High-resolution MS and LC-MS/MS identified ACh from Arabidopsis seeds

In order to uncover the biological significance of ACh within plant tissues, we first performed LC-MS/MS using multiple reaction monitoring (MRM) (the mass transition: m/z 146 > 87) and tested whether ACh can be detected in Arabidopsis (Col-0) seed extract.¹⁵ The result showed that the MRM chromatogram peak at 6.6 min obtained from the extract was identical to that of the authentic ACh with respect to the retention time of the mass chromatogram, thus ACh exists in Arabidopsis (Fig. 1A and B). On the other hand, 3 other peaks detected in the MRM chromatogram were derived from unknown compounds that were contained in the Arabidopsis seed. The data show that, in addition to ACh, Arabidopsis seeds accumulate various different compounds with the parental ion around m/z 146, which generate the fragment ions around m/z 87. In order to further confirm the accumulation of ACh in Arabidopsis seed, we performed high-resolution accurate mass MS analysis with ESI-orbitrap FT-MS on a crude extract from Arabidopsis seeds. The result obtained with EST-orbitrap FT-MS also showed that the Arabidopsis seed extracts contained a couple of molecules with the masses around m/z 146 (Fig. 2A). The well resolved MS peak of ACh from Arabidopsis seeds was detected at m/z 146.11769 as the monoisotopic ion, indicating the chemical composition formula is C₇H₁₆NO₂ (Fig. 2A and B). This observed molecular mass on Arabidopsis seed extract differs by only 0.2 and 0.9 ppm from the values of the authentic ACh (m/z 146.11766) and the theoretical monoisotopic mass of ACh (m/z 146.11756), respectively (Fig. 2A-D). Furthermore, the product fragment ions at m/z 87.04392, which was detected in the MS/MS spectrum on the seed extracts, differed by less than 0.2 and 1.6 ppm from authentic ACh (m/z 87.04390) and theoretical value (m/z 87.04406) generated from the acetyl cation moiety, respectively (Fig. 3A-C).¹⁵ This is the first report of the detection of ACh from plant tissue by high-resolution mass spectrometry, and provides conclusive evidence that ACh occurs in plants. It is notable that ESI-orbitrap FT-MS analysis detected multiple peaks of unknown compounds other than ACh. Among them were 2 major peaks detected at m/z 146.03639 and 146.09254, which was only separable from ACh by using high-resolution and accurate mass spectrometry (Fig. 2A). These results clearly indicate that plants accumulate compounds with m/z that are quite similar to ACh and highlight the importance of the use of high-resolution and accurate mass spectrometry in detecting ACh from plants.

Figure 1.

LC-MS/MS product ion MRM chromatogram (m/z 146 > 87) corresponding to ACh molecule for (A) Arabidopsis seed extract and (B) Authentic ACh (30 fmol).

Figure 2.

High-resolution accurate mass spectra of Arabidopsis seed extract (A), the expanded spectrum of (A) (B), authentic ACh (C) and the theoretical chemical composition of ACh [C7H₁₆NO₂]⁺ (D) in ESI-orbitrap FT-MS. The difference between the theoretical value and the measured value obtained from Arabidopsis seed extract was less than 0.9 ppm.

Figure 3.

The CID-MS/MS product ion accurate mass spectra from the precursor ion at m/z 146.1 of ACh detected from Arabidopsis seed extract (A), authentic ACh (B) and the calculated MS spectrum of [C₄H₇O₂]⁺. Note that the mass values obtained from the Arabidopsis seed extract differs less than 1.7 ppm.

Electrochemical detection method reveals that ACh is accumulated in seeds in Arabidopsis

Crude extracts prepared from leaf, root, and seed of Arabidopsis were subjected to quantitation of ACh by HPLC-ED (Table 1). The results showed that the concentration of ACh in seed was the highest among tissues tested (20.9 ± 5.3 nM) (Table 1). Root contained 1.5 ± 0.5 nM of ACh, while the concentration in leaf was not detected. These results clearly demonstrate that ACh is present within Arabidopsis tissues, and is preferentially localized in seeds rather than in leaf or root. Our findings suggest that ACh might play significant biological roles during seed development and/or germination.

Table 1.

ACh concentration (nM) in Arabidopsis tissues as measured by HPLC-ED.

Leaf	Root	Seed
ND	1.5 ± 0.5	20.9 ± 5.3

ND not detected.

Concentration given in units of nM, n = 3.

ACh does not significantly affect germination rates of Arabidopsis seeds

Since seed accumulates the highest levels of ACh among organs tested, we speculate that seed germination might be affected by exogenous application of ACh. To test this hypothesis, we treated Arabidopsis seeds in agar medium containing varying concentrations of ACh and compared the germination rates. The result demonstrated that seed germination rate does not seem to be significantly affected by the exogenous application of ACh when 100 nM to 10 mM ACh was applied (Fig. 4)

Figure 4.

Effects of ACh on germination rate of Arabidopsis. Arabidopsis seeds were sown on half-strength Murashige and Skoog agar medium supplemented with different concentrations of ACh. Germination rate was analyzed 5 days after germination. The value was presented as mean values of from 3 independent experiments ± SEM counted 5 days after germination.

ACh is capable of inhibiting seminal root elongation and root hair formation of Arabidopsis

To elucidate in vivo function of ACh in Arabidopsis in root, ACh was exogenously applied to Arabidopsis seeds that were sowed on agar medium. Arabidopsis seeds were grown for 5 days and the lengths of seminal roots and root hairs were measured. The result showed that compared with a no-ACh control, ACh inhibited the average length of seminal roots by 30–33% when 10 nM to 1 μM of ACh was supplemented to the medium (Fig. 5). Similarly, the length of root hairs was significantly reduced when ACh concentration ranged from 10 nM to 1 μM (Fig. 6). These results indicate that ACh can affect seminal root elongation and root hair formation. In turn, the data also imply that Arabidopsis root has a mechanism to sense and respond to exogenously applied ACh both at the tissue (seminal root) and the single cell (root hairs) levels. From these observations, it is conceivable that a decrease in the level of endogenous ACh that is localized in mature seed is associated with the elongation of seminal roots and root hairs. Notably, it was previously shown that the application of AChE inhibitors neostigmine and physostigmine to barley seeds repressed germination.⁶ Thus, it appears that genes responsible for the biosynthesis, sensing, and degradation of ACh in Arabidopsis is localized primarily to the seed in Arabidopsis, while root is also the site that is capable of sensing exogenously applied ACh.

Figure 5.

Effects of ACh on seminal root elongation. Arabidopsis seeds were sown on half-strength Murashige and Skoog agar medium supplemented with different concentrations of ACh, and incubated. Figure shows mean values of seminal root length measured after 5 days of incubation (n = 13). Asterisks indicate significant difference to no ACh treatment control (0 mM) (*P < 0.05, **P < 0.01, Student's t-test).

Figure 6.

Effects of ACh on root hair formation. Root hair length (cm) is plotted on the x-axis, and frequency is plotted on the y-axis. σ represents standard deviation (n > 36). Arabidopsis seeds were sown on half-strength Murashige and Skoog agar medium supplemented with different concentrations of ACh. Root hair lengths were measured 3 days after germination by Image J software and plotted as histograms. Representative microscopic images of root hairs are shown in insets. White bars; 300 μm.

Discussion

The apparent low levels of ACh in leaf and root compared to that of seed also suggest that there is an enzyme that degrades ACh, which is yet to be identified from Arabidopsis, in root and leaf primarily after germination, and that plants have a mechanism to maintain ACh levels in the respective tissues during the course of development. The use of LC-MS/MS and ESI-orbitrap MS for high-resolution and accurate mass detection of ACh from Arabidopsis seed in this study provides strong evidence that plants indeed accumulate ACh, which is complementary to the results from HPLC-ECD where ACh can be detected only indirectly (Fig. 1A and B, Fig. 2A-D, Fig. 3A-C). High-resolution mass spectral analysis of ACh has been applied to either mammals or in vitro enzyme assays previously.^16,17 However, there have been no reports on the detection of ACh from plant tissues using mass spectrometry at high resolution where the level of ACh is generally much lower compared to animal tissues. In this study, we detected various peaks which have m/z values quite similar, but clearly distinct from that of ACh by ESI-orbitrap FT-MS (Fig. 2). The differences in their m/z values of ACh (m/z 146.11769) and an unknown peak (m/z 146.03639) that showed highest abundance among peaks between 145.7 and 146.6 in their m/z values was less than 81.3 thousandth mass units. These results provide strong evidence that ACh exists in Arabidopsis seeds. The data also show that Arabidopsis seeds accumulate metabolites with m/z values similar to that of ACh. Collectively, our results show that careful examination of plant extracts using high accuracy mass spectrometry is required to distinguish ACh from other unrelated molecules as detected in our experiment.

While Arabidopsis seed accumulates highest levels of ACh compared to cotyledon or roots, their germination rates were not considerably affected through exogenous application of ACh (Fig. 4). It is plausible that the embryos at the onset of germination do not have direct access to the exogenously supplied ACh, or alternatively, endogenous ACh within seed functions primarily in seed development rather than germination.

HPLC-ECD analysis in our study demonstrated that seed accumulates higher level of ACh than cotyledon and root in Arabidopsis (Table 1). In addition, exogenously applied ACh inhibits the development of seminal root as well as root hairs of Arabidopsis seedlings (Figs. 5 and 6). Since the cotton root was previously shown to be rather elongated upon exposure to salinity stress from sodium chloride ¹⁸, the inhibitory effect of ACh upon root elongation is likely to be from ACh, but not from chloride ion that is derived from ACh chloride that was used to prepare various concentrations of ACh solution. Therefore, we conclude that ACh plays a pivotal role in root elongation in Arabidopsis. The concentration of exogenously applied ACh that significantly inhibits the seminal root growth was between 10 nM to 1 μM (Fig. 5). While these values are much lower than mammalian neuronal tissues, this level of concentration was equivalent to that of seeds (Table 1), showing that the amount of ACh that is required for the inhibition of seminal root elongation and root hair formation falls within the range of the physiological level of ACh in Arabidopsis.

Altogether, our data demonstrate conclusive evidence that Arabidopsis accumulates ACh. Higher levels of ACh in seed than in leaf or root suggest that there is an enzyme that degrades ACh, which is yet to be identified from Arabidopsis, in root and leaf primarily after germination, and that plants have a mechanism to maintain ACh levels in respective tissues during the course of development. Therefore, seed is likely to be a promising source for identifying molecular components that are involved in the biosynthesis, and if any, the degradation and sensing of ACh in Arabidopsis. Further characterization of the molecular basis for the biosynthesis and the recognition of ACh, together with the biological roles that ACh might play in plants would lead to the understanding of the functional diversification of ACh from evolutionary perspective.

Materials and Methods

Plant material

Arabidopsis thaliana (Col-0) seeds were surface-sterilized with hypochlorite, sowed on half-strength Murashige and Skoog agar medium containing 0.8% Phytoblend (Caisson Laboratories, North Logan, UT, USA), and cultivated in a growth chamber under a long-day photoperiod (16-h light/8-h dark) at 22°C for 5 days. The agar plates (9 × 15 cm) were placed vertically in the growth chamber for easy visual inspection and measurement of seminal root length.

Extraction of ACh from Arabidopsis (Col-0) seeds

10 mg of Arabidopsis (Col-0) seed was frozen in liquid nitrogen, homogenized using a TissueLyser (Qiagen, Germany) at the vibration rate of 60 times/s for 1 min, and dissolved in 200 µl of 0.1 M perchloric acid supplemented with 100 μM EDTA-2Na. After a 5-min incubation at 4°C, the samples were centrifuged at 15,000 rpm for 15 min at 4°C. The supernatant was transferred to a new 1.5-ml tube and mixed with 5 µl of 1 M K₂CO₃.

MRM chromatogram analysis and LC-MS/MS

The LC-MS/MS analysis was conducted with LCMS-8030 (Shimadzu, Japan). Chromatographic separation was performed using a 50 mm × 3 mm i.d., 3 μm, Intrada Amino Acid column (Imtakt, Japan) at 40°C. The injection volume was 1 μl. The mobile phase consisted of solvent A containing 0.2% (v/v) formic acid and 100 mM ammonium formate (pH 4.0) and solvent B containing acetonitrile at a flow rate of 0.15 ml/min. The linear gradient used was as follows: 0−5 min, 90−60% B; 5−12 min, 60% B; 12−15 min, 60-30% B; 15–16 min, 30–90%. The electrospray ionization (positive ionization mode) mass spectrometer was operated in MRM mode to observe the transition of m/z 146 to 87 (146 > 87) for ACh quantitation, at a collision energy of 20 eV.

High-resolution accurate MS analysis

High-resolution accurate MS spectra were acquired by a linear ion-trap Fourier transform Orbitrap mass spectrometer (LTQ orbitrap Elite) (ThermoFisher Scientific, Germany). The positive ion mass spectrum was obtained with an ESI source. ESI was performed under the conditions: source voltage 3.3 kV, capillary temperature 275 °C, scan range m/z 100–300, flow rate 3 μl/min. The collision-induced dissociation (CID) experiment was performed for the ACh precursor ion at m/z 146.1 with 0.5 Da mass window in linear ion trap, and the normalized collision energy was 30%. The product ion was detected by orbitrap FT-MS.

Acetylcholine detection by HPLC-ECD

Roots and aerial parts of Arabidopsis seedlings that were grown for 5 days were harvested (100 mg) in a 1.5-ml tube and immediately frozen in liquid nitrogen. For seed extraction, dried seeds of Arabidopsis were weighed (10 mg), transferred to a 1.5-ml tube, and frozen in liquid nitrogen. The frozen tissues were homogenized using a TissueLyser at a vibration rate of 60 times/s for 1 min, and dissolved in 200 µl of 0.1 M perchloric acid containing 100 µM EDTA-2Na. After a 5-min incubation at 4°C, the samples were centrifuged at 15,000 rpm for 15 min at 4°C. The supernatant was transferred to a new 1.5-ml tube, mixed with 10 µl of 1 M K₂CO₃, and frozen at −80°C until analysis. The samples were thawed just before the analysis, centrifuged at 15,000 rpm for 3 min at 4°C. The supernatant was filtrated to remove debris at 8,000 rpm for 15 min at 4°C. For HPLC-ECD analysis of ACh, a HPLC system with a reversed-phase column (Eicompak AC-GEL, ϕ2.0 × 150 mm; Eicom, Japan) and a pre-column (Eicompak CH-GEL, ϕ 3.0 × 4.0 mm; Eicom) was coupled to a column to which AChE and choline oxidase were immobilized (AC-Enzymepak II, ϕ1.0 × 4.0 mm; Eicom). The choline concentration produced by AChE was standardized against the concentration of an authentic ethylhomocholine. ACh in the column effiuent was enzymatically converted to hydrogen peroxide at 33°C by passage through the AC-Enzympak II column at a fiow rate of 0.15 ml/min with 50 mM potassium bicarbonate buffer (pH 8.2) containing 300 mg/l of sodium dodecyl sulfate and 50 mg/l of EDTA-2Na. The electrode potential was set at +300 mV against an Ag/AgCl reference electrode for the detection of hydrogen peroxide. Only single peaks were detected from Arabidopsis extracts.

Bioassay

Seeds of Arabidopsis (Col-0) were sown on half-strength Murashige and Skoog agar medium with or without the supplementation of various concentrations (10 nM to 100 μM) of ACh. The seeds were cultivated for 3 days in a growth chamber (Sanyo, Japan) under a long-day photoperiod (16-h light/8-h dark) at 23°C before analyzing morphological phenotypes. The microscopic images of root hairs were obtained using a stereo microscope (M205 FA; Leica). Root hair lengths were measured and analyzed using Image J software¹⁹. The seminal root length data were analyzed using a 2-tailed t-test. Data for germination rates are presented as mean values of germination rate from 3 independent experiments ± SEM counted 5 days after germination.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Drs. Makoto Suematsu and Honoo Satake (Suntory Foundation for Life Sciences) for valuable discussions, and Kazuko Ikimura (Eicom) for technical assistance with HPLC-ECD analysis.