Quantitative analysis of ethylene precursor ACC in plant samples by liquid chromatography-tandem mass spectrometry

Abstract

Background

Ethylene (C₂H₄) is a gaseous phytohormone that regulates various plant physiological processes and mediates the responses of the plants to various environmental stresses. 1-Aminocyclopropane-1-carboxylic acid (ACC) is the direct precursor of the phytohormone ethylene, and is also considered as a plant growth regulator. Accurate quantification of ACC is critically important in investigating its function. However, it remains challenging to accurately quantify ACC in plant tissues because it is a small, electroneutral molecule with very low concentrations.

Methods

An easy, cost-saving, and highly efficient quantitative method for ACC in plant tissues was set up by liquid-liquid micro-extraction (LLME) purification with green solvent ethyl-acetate, and the precise control of the mobile phase entering into the mass spectrometer combined with the ultra-high performance liquid chromatography electrospray ionization-triple quadrupole mass spectrometer (UHPLC-ESI-MS/MS).

Results

The contents of ACC in 10 mg of different fresh fruits were detected. The established method had limit of detection (LOD) (2.5 pg), matrix effect (ME) (92.6%), good precision (3.54%), recovery rate (95.82%), and a good linear relationship within the range of 0.5 to 1500 ng.mL⁻¹ (R² = 0.9998).

Conclusion

We have developed an easy, sensitive, and cost-saving quantitative method for ACC in plant tissues without derivatization which is useful for the precise quantification of ACC in plant tissues.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-025-06943-7.

Introduction

Ethylene (C₂H₄) is the unique gaseous phytohormone among all the nine categories of phytohormones that regulates various plants physiological processes, such as seed germination, root formation, flowering and pollination, abscission, senescence and fruit ripening [1–5]. As a ripening phytohormone [6, 7], ethylene is very effective in regulating the taste, color and texture of fruits, and has great commercial value [2]. As a stress responsive phytohormone, ethylene mediates the responses of the plant to various environmental stresses such as wounding [8, 9], flooding [10], drought [11, 12], salt stress [13–15], pathogen attack [16, 17], and hypoxia stress [18]. Up to date, major strides have been made in exploring ethylene’s metabolism, perception, signaling, and cross-talk with other phytohormones [9, 19–26].

It remains difficult to quantify ethylene accurately because it can be produced by detached organs in response to wounding and desiccation [1, 27]. The detection of ethylene can be interfered with nitrogen in the air by gas chromatography-mass spectrometry (GC-MS) because ethylene has almost the same molecular weight with nitrogen. Currently, the instrument commonly used for the detection of ethylene is a gas chromatograph (GC) equipped with a flame ionization detector (FID) or photoionization detector (PID) [3, 8]. A headspace injector and long-time headspace accumulation are required for sampling. In recent years, a new ethylene monitoring device with a laser-based photo-acoustic detector has been reported [28, 29], but the equipment is more expensive [29]. Interestingly, recent studies have showed that ACC not only serves as the unique precursor of ethylene biosynthesis, but also as a signaling molecule independent of ethylene [30–37]. Thus, people try to accurately quantify ACC instead of ethylene.

Previous quantification methods for ACC include direct quantification by chemical assay [38, 39], and indirect quantification by GC-MS [40–42] or LC-MS [43, 44] after solid phase extraction (SPE) purification and derivatization. However, these methods are complicated, and required lengthy sample pretreatment, which increases the technical difficulty and the experimental uncertainty, consequently resulting in poor repeatability and lower recovery rate. In particular, some derivatization reagents such as phenyl isothiocyanate (PITC) [42–44], acetyl chloride, pyridine [40, 41] are toxic. There are also some reports of ACC quantitative methods with simple pre-treatment coupled with LC-MS [17, 45]. However, these methods do not have purification steps, and do not use isotope labelled internal standard. As a result, these samples matrix are complex, and easily induce matrix effects (ME) in mass spectrometry detection. The complex sample matrix may interfere the ionisation process, and change the ionization efficiency of the target molecule, and consequently affect the detection signal. Especially, the matrix molecules with high content and strong ionization energy can enhance or suppress the target signal detection, and lead to erroneous measurements [46–48]. Recently, Karady et al. developed a sensitive mass spectrometry-based method that could simultaneously analyze 15 phytohormonal compounds in 10 mg of plant tissues [49]. However, ACC can be determined only after derivatization independently, and the recovery rate (88.17%) and ME (23.74%) are both low. The hydrophilic interaction chromatography (HILIC) colum was used for chromatographic separation, and [²H₄]ACC was used as an isotopic labelled internal standard to detect ACC without derivatization [50, 51]. However, these reports have shown that the limit of detections (LODs) are relatively higher. More reliable quantitative methods for ACC are urgently needed.

Deuterium isotope labelled standards are often used as internal standards to eliminate matrix effects in mass spectrometry detection [52]. [²H₄]ACC is a classic internal standard molecule in quantitative detection of ACC. Four hydrogen atoms of ACC molecule are replaced by four deuterium atoms in [²H₄]ACC, indicating that the two molecules can be broken at the same position to form characteristic ions under the action of collision energy (CE). Liquid-liquid extraction (LLE) is based on the principle that a solute or an analyte can be transferred from one solvent to another because of the difference of solubility or partition coefficient in two insoluble (or slightly soluble) solvents, so as to achieve the purpose of separation and purification [53]. LLE remains one of the most attractive separation and purification techniques due to its flexibility, simplicity, and effectiveness [54]. A high-pressure six-port rotary valve in front of the electrospray ionization (ESI) source of the mass spectrometer enables the precise control of the mobile phase entering into the ESI source to reduce the interfering substances entering into the mass spectrometer, consequently reduce the contamination of the ESI source and the mass spectrometer, and improve the stability of the detection results.

Here, we set up an easy quantitative method for ACC in plant tissues by using: (1) the green solvent ethyl-acetate for liquid-liquid micro-extraction (LLME) purification, (2) [²H₄]ACC as isotope labelled internal standard, and (3) the precise control of the mobile phase entering into the mass spectrometer combined with the ultra-high performance liquid chromatography electrospray ionization-triple quadrupole mass spectrometer (UHPLC-ESI-MS/MS).

Materials and methods

Chemicals and materials

ACC (C₄H₇NO_2, purity > 97%), 1-aminocyclopropane-2,2,3,3,-d4-carbocylic acid ([²H₄]ACC, C₄H₃D₄NO₂, purity > 99.7%), and HPLC grade acetonitrile (CH₃CN) were bought from TRC (Toronto Research Chemicals, Canada), C/D/N Isotopes Inc (Canada), and Merck KGaA (Darmstadt, Germany), respectively. Analytically pure grade ethyl-acetate (CH₃COOCH₂CH₃), chloroform (C₂H₃Cl₃), n-hexane (C₆H₁₄), petroleum ether and acetic acid (CH₃COOH) were bought from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). Ultrapure water (resistivity ≥ 18.2 MΩ/cm) used throughout the study was obtained using a Simplicity-UV water purification system (Merk Millipore, USA) by reversed osmosis.

Fresh fruits as tested samples were bought from local supermarket, including lotus seed, grape, cherry tomato, plum, peach, pear, apple, navel orange, banana. These fruits were full ripe except lotus seed was at developmental stage.

Sample preparation

Samples were prepared using the reported methods with mild modification [55–57]. First, the fresh fruit tissue was homogenized and powdered with a mortar and pestle in liquid nitrogen. 10 mg of the powdered fruit tissues was put into a 1.5 mL centrifuge tube as a sample. After adding acetonitrile (0.5 mL) and isotope-labeled internal standards (IS) [²H₄]ACC (0.2 µg), the samples were shaken on a vortex (SI HYQ3110, USA) for 1 h at room temperature, and centrifuged at 15,000 g for 10 min (Eppendorf centrifuge 5418, Germany). The supernatant was collected into a new 1.5 mL centrifuge tube, the residue was mixed with 0.2 mL of acetonitrile and centrifuged again. The supernatant was also collected, and added into the previous supernatant. All supernatant were vacuum-dried in a Jouan RCT-60 concentrator (Jouan, France). Residues were reconstituted with 100 µL ultrapure water. After sonicating for 5 min, and adding 100 µL ethyl-acetate, the reconstituted residues were vortex oscillated (1 min) and centrifuged (15000 g, 1 min) at room temperature. The upper ethyl-acetate phase was removed. The aqueous phase was again purified by adding 100 µL ethyl-acetate again for LLME, then evaporated in vacuo. Dried residues were either stored at −20 °C (Haier BCD-216E, Qingdao, China) or were reconstituted with 100 µL 80% acetonitrile, sonicated for 5 min, centrifuged (15000 g, 10 min), transferred to a LC auto-sample vial, and analyzed by UHPLC-ESI-MS/MS. The overall process of sample pretreatment was shown in Fig. 1.

Fig. 1.

Schematic diagram of the high-sensitivity scheme operation for quantitative analysis of ACC. IS, internal standards; CPS, counts per second; LLME, liquid-liquid micro-extraction

Standard samples preparation

ACC and [²H₄]ACC standard products were accurately weighed and dissolved in ultrapure water to prepare reserve solutions of 1000 µg.mL⁻¹, respectively. The reserve solutions were further diluted with 80% acetonitrile to obtain the single standard solutions of 2 µg.mL⁻¹ ACC and [²H₄]ACC, for full-scan, product ion-scan and optimization of the multiple reaction monitoring (MRM)-MS method. The different concentrations of ACC standard solutions (1500, 750, 300, 150, 75, 50, 30, 10, 5, 2, and 0.5 ng.mL⁻¹ ACC/2 µg.mL⁻¹ [²H₄]ACC) were diluted with 2 µg.mL⁻¹ ACC and [²H₄]ACC single standard solutions for the calibration curve. The standard solutions were stored at −20 ℃ or analysed by UHPLC-ESI-MS/MS.

Instruments and analytical conditions

ACC analysis was performed on a UHPLC-ESI-MS/MS described in our previous study [55]. A Universil AQ-C₁₈ column (2.1 mm×150 mm, 1.8 μm, Kromat Corporation, USA) was used for UHPLC separation. The mobile phases included 0.05% acetic acid in water (A) and acetonitrile (B). The separation conditions were as follows: flow rate 0.2 mL.min⁻¹, column temperature 30 °C, elution with 10% B for 3 min, 10–90% B at 3–4 min, and 90% B lasted for 1 min, followed by 3 min of re-equilibration with 10% B. The injection volume was 5 µL.

In addition, an FCV-20AH2 high-pressure six-port rotary valve (Shimadzu, Japan) was installed between Universil AQ-C₁₈ column and ESI source [55]. The valve was switched to the 1–2 position at 1.6–2.6 min, ACC was taken into the mass spectrometer by the mobile phase to be detected. At other times, the valve was switched to the 1–6 position, the sample matrix and other interfering substances entered into the waste liquid collector with the mobile phase (Fig. 1).

ACC was quantified by MRM in the positive mode. We used the single standard solution of ACC and [²H₄]ACC at a concentration of 2 µg.mL⁻¹ to scan and optimize the MRM parameters. ACC at 150 ng.mL⁻¹ was employed to optimize the ESI source parameters. The optimal MRM parameters for ACC were listed in Table 1. The scanning resolution for quadrupole1 (Q1) and quadrupole3 (Q3) were set at “Low”. The “Conversion Dynode” voltage value was set 9 kV. And the optimal conditions for ESI source parameters were as follows: desolvation (DL) temperature, 150 °C; heat block temperature, 400 °C; nebulizing gas, 3 L.min⁻¹; drying gas, 12 L.min⁻¹; capillary voltage, 4.5 kV.

Table 1.

Optimized MRM parameters for ACC (Q1 and Q3 pre bias [V]; CE [eV])

Analyte	Quantification				Confirmation
	Q1/Q3(m/z)	Q1 pre bias	CE	Q3 pre bias	Q1/Q3(m/z)	Q1 pre bias	CE	Q3 pre bias
ACC	102.15/56.05	-10	-14	-21	102.15/28.1	-10	-24	-30
[2H4]ACC	106.2/32.1	-11	-37	-28	106.2/60.1	-11	-14	-23

Method validation

As we reported previously [55], the linearity of the proposed method was evaluated by different concentration of ACC standards (0.5, 2, 5, 10, 30, 50, 75, 150, 300, 750, and 1500 ng.mL⁻¹) with a fixed concentration of internal standard (IS, [²H₄]ACC 2 µg.mL⁻¹). The precision and accuracy of the proposed method were evaluated by spiking 100 µL ACC standard solutions with authentic (ACC, 0, 15, 150, and 750 ng.mL⁻¹) and internal standard ([²H₄]ACC, 2 µg.mL⁻¹) into 10 mg fresh banana samples in triplicate and then treated with the proposed procedure shown in Fig. 1.

The ME of the proposed method was evaluated by spiking 100 µL ACC standard solution at a concentration of 150 ng.mL⁻¹ ACC/2 µg.mL⁻¹ [²H₄]ACC into 10 mg fresh banana samples, with each sample consisting of three replicates. The peak area ratio of ACC/[²H₄]ACC in these samples detected by UHPLC-ESI-MS/MS were compared with those of ACC/[²H₄]ACC in 80% acetonitrile, and the ME of ACC was calculated.

The proposed method was also used for the analysis of ACC in various fresh fruit samples. To ensure the accuracy of the experimental results, three technical replicates were set up for each sample.

Results

Optimization of extraction solvent

In order to investigate the suitable extraction solvents, chloroform, n-hexane, ethyl-acetate and petroleum ether were tested as extraction solvents to extract 100 µL ACC standard solution at a concentration of 100 ng.mL⁻¹ (dissolved in ultrapure water) with different extraction conditions combined by vortex-shaken time (1, 2, or 5 min), extraction temperature (4, 15, or 25 °C), and solvent volume (100, 200, or 400 µL) (Supplementary Table S1). The contents of ACC in the aqueous phase and the organic phase were quantified respectively, The extraction efficiency was calculated. ACC was mainly distributed in the organic phase of petroleum ether or n-hexane extraction, and in the aqueous phase of ethyl-acetate or chloroform extraction. Among all extraction conditions tested, ethyl-acetate showed higher extraction efficiency, relative to other extraction solvents (Supplementary Table S1). Chloroform [58], n-hexane [59] and petroleum ether [60] are all toxic, while ethyl-acetate is an environmentally friendly organic solvent [61–63]. Thus ethyl-acetate with extraction conditions of vortex-shaken for 1 min, 100 µL volume and room temperature was used in later experiments.

Optimization of chromatographic separation conditions

Previous studies have shown that the interaction between octadecylsilyl ligands and adsorbate molecules in acetonitrile/water system is weaker than that in methanol/water system [64]. In addition, we did not obtain the characteristic ion 32 when methanol/water system was used as the mobile phase to do “product ion-scan” for [²H₄]ACC. It is possible that the pressure of methanol/water system is greater than that of acetonitrile/water system, resulting in lower signal-to-noise (S/N) ratio. Therefore, acetonitrile/water system was selected as the mobile phase for subsequent experiments.

In order to optimize chromatographic separation conditions, we tested Universil AQ-C₁₈ 2.1 mm×150 mm 1.8 μm (Kromat Corporation, USA), Shim-pack XR-ODSI 2.0 mm×75 mm 2.2 μm (Shimadzu, Japan), Acquity BEH C18 2.1 mm× 100 mm,1.7 μm (Waters, USA) column with 10% acetonitrile/90% water as the mobile phase. Compared to other columns, Universil AQ-C₁₈ column showed longer retention time, and symmetrical peak shape of ACC (data was not shown). Thus, Universil AQ-C₁₈ column was used in the later experiments.

To investigate the effect of acetonitrile ratio in the mobile phase on the retention time of ACC, different ratios of acetonitrile/water (5/95, 10/90, or 20/80) as the mobile phase were tested, respectively. We found that the retention time of ACC was almost unchanged, about 2 min. We selected 10% acetonitrile as the organic phase, maintained isobaric elution until ACC was eluted, and then increased the acetonitrile ratio to 90% to elute the residual interfering substance in the column to reduce the influence of co-elution interfering substances on ACC quantification, and protect the column.

In order to increase signal sensitivity and S/N ratio, different concentrations of formic acid (0.02%, 0.05%, 0.1%) and acetic acid (0.02%, 0.05%, or 0.1%) were tested as aqueous phase, respectively. Compared to other aqueous phases, 0.05% acetic acid showed better signal sensitivity and S/N ratio (data was not shown).

To optimize the sample solvent, we compared the effects of different acetonitrile/water ratios on signal intensity. We found that 80% acetonitrile/20% water showed the highest signal intensity of ACC, 52 times that of ultrapure water. Thus, 80% acetonitrile/20% water was chosen as the sample solvent.

Combined all the above information, the chromatographic separation conditions we used in the later experiment included: Universil AQ-C₁₈ 2.1 mm×150 mm 1.8 μm (Kromat Corporation, USA) column, 80% acetonitrile/20% water as sample solvent, acetonitrile and 0.05% acetic acid aqueous solution as the mobile phase gradient elution. In addition, in order to minimize contamination of the ESI ion source and the mass spectrometer, only the mobile phases close to the retention time of ACC within 1 min could pass through the high-pressure six-port rotary valve entering into the ESI ion source, and the rest was switched into the waste collector (Fig. 1).

Optimization of mass spectrometer conditions

In order to optimize the MRM parameters, the concentration of 2 µg.mL⁻¹ of ACC and [²H₄]ACC single standard solutions were used for Full-scan and Product ion-scan, to obtain their precursor ions and product ions. [²H₄]ACC is an isotopic compound of ACC, and has similar structure of ACC except that [²H₄]ACC has four deuterium atoms instead of hydrogen atoms. Thus, theoretically, they should be broken at the same location under the action of CE of the mass spectrometer (MS). However, we obtained the precursor ion 102 (Fig. 2A) and the characteristic ions 56 (removed the carboxyl group) and 28 (removed the carboxyl group and amino group) of ACC (Fig. 2B). In contrast, in [²H₄]ACC, the precursor ion 106 (Fig. 3A) and the characteristic ion 60 (removed the carboxyl group) were obtained but not the characteristic ion 32 (removed the carboxyl group and amino group) (Fig. 3B). In order to get characteristic ion 32, acetonitrile was used as solvent and mobile phase instead of methanol for product ion-scan of [²H₄]ACC standard (2 µg.mL⁻¹) at collision energy of −10, −15, −20, −25, and − 30 eV, respectively. As expected, the characteristic ion 32 was found at CE of −30 eV (Fig. 3C). Finally, we obtained the parameters of automatically optimized MRM parameters using the “Optimization for Method” data acquisition software, and established the MRM parameters of the UHPLC-ESI-MS/MS method shown in Table 1.

Fig. 2.

MS/MS spectra and proposed fragment pathways of ACC for Full-scan (A) and Product ion scan (B). Product ion scan at collision voltage of −12 eV

Fig. 3.

MS/MS spectra and proposed fragment pathways of [²H₄]ACC for Full-scan (A) and Product ion scan (B and C). Product ion scan at collision voltage of −15 eV (B) and − 30 eV (C)

In order to optimize the ion source parameters, we adjusted DL temperature (120, 140, 150, 160, or 180 °C), heat block temperature (300, 350, 400, or 450 °C), nebulizing gas flow (2, 2.5, 3 L.min⁻¹), and drying gas flow (10, 12, 15 L.min⁻¹) to test ACC standard sample with a concentration of 150 ng.mL⁻¹, respectively. Compared to these ion source parameters, DL temperature, 150 °C; heat block temperature, 400 °C; nebulizing gas, 3 L.min⁻¹; drying gas, 12 L.min⁻¹ showed better signal sensitivity (data was not shown).

In addition, the instrument specification has shown that the mass spectrometer detector comprises a “Conversion Dynode” and an “Electron Multiplier”, that detects positive and negative ions passing through the quadruple rods. The ion signal can be amplified when the “Conversion Dynode” is applied a voltage value, and Shimadzu 8030 Plus mass spectrometer provides a range of voltages at 1–10 kV [65]. We tested the different voltage values of “Conversion Dynode”(6, 7, 8, or 9 kV), and found that the best ion signal intensity was detected at 9 kV.

Moreover, previous studies have shown that reducing the scanning resolution of the mass spectrometer can improve the detection sensitivity by increasing the number of ions passing through the quadrupole filter and the scanning speed [66, 67]. Shimadzu 8030 Plus mass spectrometer provides three resolution options, Low (~ 1.0 Da), Unit (~ 0.7 Da), and High (~ 0.5 Da). Therefore, we tested the scanning resolution of Q1 and Q3 (High, Unit, or Low), the signal intensity of ACC was increased about 1.3 and 1.8 times, indicating that the sensitivity of ACC was improved at low scanning resolution of the quadrupole. Hence, the scanning resolution for Q1 and Q3 was set at “Low” in the MRM method for analysis ACC.

Lastly, we detected the contents of ACC in banana samples with the above established MRM-MS method for ACC. As shown in Fig. 4A-B, the same characteristic ions (m/z:102.15/56.05 and 102.15/28.1) could be detected as ACC standard solution in fresh banana extract. However, we found that the peak area of [²H₄]ACC extracted ion chromatogram (EIC) (m/z: 106.2/60.1) in banana extract was 7.5 times that of the ACC standard solution (Fig. 4C-D), indicating that the extracted ion (m/z: 106.2/60.1) of [²H₄]ACC contained interfering ion in the sample, which would interfere with the quantitative result of ACC. Thus, m/z (106.2/32.1) was chosen as the quantitative ion for [²H₄]ACC.

Fig. 4.

The extraction ion chromatograms of ACC and [²H₄]ACC in standard sample (A, C) and banana extracts (B, D) by UHPLC-ESI-MS/MS. The standard sample and banana extracts dissolved in ultrapure water, respectively

Method validation

To evaluate the quantitative capability of the established method, linearity, LOD, limit of quantitation (LOQ) were investigated. The levels of eleven ACC standard solutions were prepared in 80% acetonitrile with a concentration of 0.5 to 1500 ng.mL⁻¹, and the concentration of IS ([²H₄]ACC) was 2 µg.mL⁻¹. The calibration curve was set up by plotting the peak area ratio (analyte/IS) versus ACC concentrations. ACC had the same retention time with [²H₄]ACC (Supplementary Fig. S1A-B). The correlation coefficient (R²) was 0.9998 (Supplementary Fig. S1C). The LOD (S/N = 3) and LOQ (S/N = 10) were 2.5 and 8.3 pg, respectively. In addition, the accuracy and precision of this method were evaluated by the recovery and the intra- and inter-day relative standard deviation (RSD). The recovery, intra- and inter-day precisions were calculated with ACC standard spiked in 10 mg fresh banana at four different concentrations (0, 15, 150, and 750 ng.mL⁻¹ ACC/2 µg.mL⁻¹ [²H₄]ACC) in triplicate. The average recovery ratio was 95.82%, and the RSD of intra- and inter-day precision was 3.54% (Table 2). The results indicated that the developed method had good reproducibility and accuracy, and was highly sensitive for profiling ACC in plant samples.

Table 2.

Accuracy and precision (intra- and inter-day) for the determination of ACC in banana (10 mg fresh mass)

spiked conc. (ng.mL-1)	Repeatability (%, n=3)			Average recovery rate (%)	Average RSD (%)
	Recovery rate	intra-day precision (RSD)	inter-day precision (RSD)
750	96.31	2.75	3.74	95.82	3.54
150	95.36	2.69	3.27
15	95.79	3.84	4.21
0		3.62	4.17

Evaluation of matrix effect

Complex sample matrix can limit or enhance the ionization of the analytes in ESI, and affect the MS signal [52, 68]. In order to evaluate the ME of the developed method, 100 µL ACC standard solution at a concentration of 150 ng.mL⁻¹ ACC/2 µg.mL⁻¹ [²H₄]ACC was added into fresh banana (10 mg) and 80% acetonitrile, respectively. The ME was calculated by comparing the peak area ratio of ACC/[²H₄]ACC in fresh banana extract and 80% acetonitrile. We found that the ME value of ACC was 92.6%, much better than the 23.74% of the previous study [49].

Determination of ACC in fresh fruit samples

To investigate the universality of the established method, the level of ACC in different fresh fruits were detected by this method. As shown in Fig. 5 and Supplementary Fig. S2, the same characteristic ions (m/z:102.15/56.05 and 102.15/28.1) could be detected in these fresh fruit extracts as in ACC standard samples, and the retention times were almost the same as standard and internal standard, validating the high selectivity and the high sensitivity of our method. In addition, Fig. 6 shows the content of ACC in different fruit samples. The ACC levels change with the type of fruits. The highest level is in apples, almost reaching 2000 ng.g⁻¹, followed by cherry tomatoes, about 700 ng.g⁻¹, and the least content in pears, only about 60 ng.g⁻¹.

Fig. 5.

The extraction ion chromatograms of ACC and [²H₄]ACC in standard sample (A) and fresh fruit extracts (B) by UHPLC-ESI-MS/MS. The standard and fresh fruit extracts dissolved in 80% acetonitrile, respectively

Fig. 6.

The contents of ACC measured in various fruits. These data are the means ± SDs (n = 3)

Method comparison

Compared to the previously reported methods, our method has some advantages. First, [²H₄]ACC was used as the isotope labelled internal standard, and was employed for analyzing ACC in plant samples without derivatization, which reduced the pretreatment steps and the amount of toxic chemicals, and accurately quantified the content of ACC in various fresh fruits. Second, no other solvent except ethyl-acetate and the mobile phase was used in the pretreatment, which significantly reduced the ion inhibition effect of interfering substances on low abundance target substances. Third, a lot of interfering substances was switched into the liquid waste collector via a high-pressure six-port rotary valve, which reduced contamination of the ESI source and mass spectrometer, and increased the stability of the results. Compared with representative methods reported in the last decades [40–45, 49–51], our method has few pretreatment steps, less chemical reagents, less amount of plant tissues (10 mg) and low LOD (2.5 pg, Table 3).

Table 3.

Comparison of our developed ACC analytical method with previously established methods

Pretreatment method	Chemical reagents	Amount of plant tissues	Internal standard	LODa	References
Derivatization, SPE coupled with HPLC purification	Isobutanol, benzoic-anhydride, hexane, hydrochloric-acid, alcohol, pyridine, n-propanol, ethyl-acetate, acetic-acid, acetyl-chlorid, ammonia, methanol, methylene-chloride	100–200 mg fresh maize roots, cucumber leaves, rice seedlings	N-benzoyl isobutyl- ACC	2.5 ng	[40]
Derivatization, SPE coupled with HPLC purification	Isobutanol, n-propanol, alcohol, ammonia, benzoicanhydride, acetyl-chlorid, methanol, butylated-hydroxytoluene	3 g tomato leaves and xylem sap	N-benzoyl isobutyl- ACC	Not mentioned	[41]
Tandem SPE purification, derivatization coupled with LLE purification	Trifluoroacetic-acid, methanol, triethylamine, hydrochloric-acid, acetone, ammonia, ethyl-acetate, pentafluorobenzyl-bromide	200 mg Brassica and Arabidopsis seedlings	[2H4]ACC	10 fmol	[42]
Triple layers SPE coupled with HPLC purification, derivatization	Hydrochloric-acid, methanol, triethylamine, ethanol, trifluoroacetic-acid, ammonium-acetate, acetic acid, ammonia, phenyl-isothiocyanate	1 g Nicotiana tabacum cv. leaves	[2H4]ACC	10 pmol	[43]
Triple layers SPE purification, derivatization	Methanol, hydrochloric-acid, trifluoroacetic-acid, triethylamine, ammonium-acetate, ethanol, phenyl-isothiocyanate	0.1 and 3 g Nicotiana tabacum cv. leaves, stem and roots	[2H4]ACC	2.5 ng	[44]
No purification, no derivatization	Acetonitrile, methanol, tetrahydrofuran, sulfosalicylic-acid, nonafluoropentanoic-acid	100 g apples	Not used	20 pmol	[45]
No purification, derivatization	AccQ-Tag dltra derivatization kit, formic acid, acetonitrile	10 mg Arabidopsis seedling	[2H4]ACC	Not mentioned	[49]
SPE purification, no derivatization	Oasis MCX column, methanol, ammonium hydroxide, isopropanol, acetonitrile, formic acid, acetic acid	80 mg Arabidopsis seedling	[2H4]ACC	3 pmol	[50]
No purification except filtration, no derivatization	Amicon centrifugal filter, methanol, acetonitrile, formic acid	100 mg Arabidopsis shoot and root	[2H4]ACC	0.006 nmol.g-1	[51]
LLME purification, no derivatization	Acetonitrile, ethyl-acetate, acetic-acid	10 mg fresh fruits	[2H4]ACC	2.5 pg	This work

^aLOD, Limit of detection

Discussion

Ethylene is synthesized from S-adenosyl-L-methionine (SAM) by two dedicated enzymes (ACC synthase (ACS) and ACC oxidase (ACO)), via ACC as an intermediate [4]. Metabolic regulation of ethylene is achieved by ACC homeostasis, including ACC biosynthesis, transport and conjugation, and three derivatives of malonyl-ACC (MACC), glutaminyl-ACC (GACC) and jasmonyl-ACC (JA-ACC) have been identified (Fig. S3) [69, 70]. The ACC-to-MACC conversion is catalyzed by the enzyme ACC-N-malonyl transferase (AMT), and has been shown to exist in planta [51]. The formation of GACC is catalyzed by the enzyme g-glutamyl-transferase (GGT). GACC is only detected in vitro assays by incubating plant enzyme extracts with relatively high amounts of exogenous ACC. It is still unclear whether it exists in plants [51]. JA-ACC is formed by jasmonic acid resistance 1 (JAR1) synthetase from JA and ACC, and can be formed in vitro under the catalysis of the recombination JAR1 enzyme. It has also been shown to be present in plant tissues [71]. However, due to its complex purification methods and low content, JA-ACC is not consistently found in plants as an endogenous compound [51].

Recent insights pinpoint, ACC is not only considered as the unique precursor of ethylene biosynthesis, but also acts as a signaling molecule independent of ethylene [30–37] in plant growth, development, cell wall signaling, guard mother cell division, stomatal development, pollen tube attraction, stress responses and pathogen virulence [31–35, 69, 70, 72]. ACC can also be metabolized by bacteria using ACC-deaminase, favoring plant growth and lowering stress susceptibility [17, 73]. Therefore, accurate quantification of ACC will facilitate the study of the dual function of ACC in signaling and ethylene biosynthesis.

In this paper, we explored the accurate quantitative method of ACC in various fresh fruit samples. We found that the extraction efficiency could reach 99.42% with ethyl-acetate as the extraction solvent for LLME. Ethyl-acetate extraction is a rapid and effective purification method and can remove esters and other organic compounds from samples [62]. The time of SPE purification, derivative reaction, and the solvent drying was saved, and the cost of chemical reagents was reduced for using LLME purification in this experiment. The experiment had a LOD of 2.5 pg, and could sensitively detect the content of ACC in trace plant samples as low as 10 mg fresh mass, showing the LOD and sensitivity are better than the reported previously [40, 42–45, 50, 51]. In addition, the developed method has a low ME of 92.6%, which is better than 23.74% [49] of the previous reports. The possible reasons are that in our method, the sample pretreatment process is simple, the amount of reagents added is less, and the mobile phase entering into the mass spectrometer is precisely controlled, resulting in cleaner samples. The real reason remains unclear.

Isotope labelled compounds are usually chosen as internal standards of the target because they have similar physical and chemical properties [52]. They are used to correct the loss during sample preparation, compensate the change of matrix signal during mass spectrometry analysis, obtain accurate recovery rate, and ensure more reliable results [46, 52]. In this study, we found that the characteristic ion m/z 106.2/60.1 of isotope labelled internal standard [²H₄]ACC had interfering ions in the sample as previously reported [50]. However, we did not further confirm the interfering ions due to the limitation of the instrument and no other standards. As an alternative, we found that characteristic ion (106.2/32.1) had no cross-contribution signal to serine or other interfering ions, consistent with previous reports [50]. In addition, [²H₄]ACC is the internal standard. ACC was quantified by correlating the concentration of ACC in the standard solution with the area ratio of ACC/[²H₄]ACC. Thus, the intensity of the 106.2/32.1 ion doesn’t affect the quantification of ACC, although it was lower than the 106.2/60.1 ion. Serine and [²H₄]ACC are both polar small molecules with same molecular weight and similar chemical properties, which makes them difficult to separate under the same chromatography conditions. AQ-C₁₈ column with polar end-capped silanol groups is designed for hydrophilic and polar compounds, and has good retention and selectivity [74]. Thus, we used it in our method. Recent studies reported that they developed and validated a HILIC-MS/MS method for the determination of ACC with satisfactory results [50, 51]. Maybe, it was an alternative choice.

In this study, we set up an easy, fast, safe and accurate method for the detection of ACC, and it can be used to detect the level of ACC in micro samples. It may better help elucidate the roles of ACC in plant growth and development if the levels of ACC are measured mutant plant species such as ACOs and ACSs in this way [35, 36, 69]. In addition, three ACC derivatives identified all are formed by combining a different group, such as a malonyl or glutaminyl, on the amino group of the ACC molecule (Fig. S3). Although these substances have similar physical and chemical properties, they have different molecular weights, and different precursor ions ([M + H]⁺or [M-H]⁻) will be obtained by the quadrupole mass spectrometry. They will be separated from ACC in the first stage of the quadrupole mass spectrometer, and not be detected as ACC. Moreover, the extracts were first separated by liquid chromatography, and only the sample matrix and mobile phase close to the ACC retention time of 1 min could pass through the high-pressure six-port rotary valve into the mass spectrometer for detection eliminating a large number of interfering substances. On the contrary, the extracts obtained by this pretreatment method can simultaneously be used to detect the ACC derivatives. Additionally, full-scan analysis using high-resolution mass spectrometry may reveal new derivatives.

In summary, our method will promote the study of signaling and biological role of ACC in plants. It can also aid to shed more light on ACC formation, conjugation, degradation and localization.

Conclusions

We have developed and validated an easy, sensitive, and cost-saving method for the quantification of ACC in plant tissues based on LLME purification, precise control of the mobile phase entering into the mass spectrometer coupled with the UHPLC-ESI-MS/MS. Moreover, [²H₄]ACC was used as internal standard without derivatization in this method.

Abbreviations

ACC

1-Aminocyclopropane-1-carboxylic acid

LLME

Liquid-liquid micro-extraction

UHPLC-ESI-MS/MS

Ultra-high performance liquid chromatography electrospray ionization-triple quadrupole mass spectrometer

LOD

Limit of detection

LOQ

Limit of quantitation

GC-MS

Gas chromatography-mass spectrometry

FID

Flame ionization detector

PID

Photoionization detector

SPE

Solid phase extraction

PITC

Phenyl isothiocyanate

ESI

Electrospray ionization

S/N

Signal-to-noise

DL

Desolvation

CE

Collision energy

Q1

Quadrupole1

Q3

Quadrupole3

CID

Collision-induced dissociation

MRM

Multiple reaction monitoring

IS

Internal standard

ME

Matrix effect

SAM

S-adenosyl-L-methionine

ACS

ACC synthase

ACO

ACC oxidase

MACC

Malonyl-ACC

GACC

Glutaminyl-ACC

JA-ACC

Jasmonyl-ACC

AMT

ACC-N-malonyl transferase

GGT

G-glutamyl-transferase

JAR1

Jasmonic acid resistance 1

Authors’ contributions

Conceptualization, J.T., W.Z. and L.X. Methodology, J.T. and W.Z. Software, J.T., K.W., F.Z. and D.D. Formal Analysis, J.T., W.Z., and K.W. Investigation, J.T. and L.X. Writing—Original Draft Preparation, J.T. and W.Z. Writing—Review and Editing, J.T., F.Z. and L.X. Visualization, J.T. Supervision, L.X. All authors read and approved the final version of the manuscript.

Funding

This research was supported by the National Natural Science Foundation of China (32261143733 and 90817101).

Data availability

All data presented in this study are included in this published article and its supplementary materials.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

This manuscript has been reviewed and approved by all authors.

Competing interests

The authors declare no competing interests.