Natural 2-Amino-3-Methylhexanoic Acid as Plant Elicitor Inducing Resistance against Temperature Stress and Pathogen Attack

2-Amino-3-methylhexanoic acid (AMHA) was synthetized as a non-natural amino acid more than 70 years ago; however, its possible function as an inducer of plant resistance has not been reported. Plant resistance inducers, also known as plant elicitors, are becoming a novel and important development direction in crop protection and pest management. We found that free AMHA accumulated in the mycelia but not in fermentation broths of four fungal species, Magnaporthe oryzae and three Alternaria spp. We unequivocally confirmed that AMHA is a naturally occurring endogenous (2S, 3S)-α-amino acid, based on isolation, purification and structural analyses. Further experiments demonstrated that AMHA has potent activity-enhancing resistance against extreme temperature stresses in several plant species. It is also highly active against fungal, bacterial and viral diseases by inducing plant resistance. AMHA pretreatment strongly protected wheat against powdery mildew, Arabidopsis against Pseudomonas syringae DC3000 and tobacco against Tomato spotted wilt virus. AMHA exhibits a great potential to become a unique natural elicitor protecting plants against biotic and abiotic stresses.


Introduction
Plants are constantly subjected to different abiotic and biotic stress conditions throughout their life cycle. For crops, the most characterized abiotic stresses include extreme temperature, drought, salinity and heavy metal toxicity [1]. Biotic stresses are imposed by bacteria, fungi, viruses, insects, parasites and weeds [2]. Severe and sustained stresses can lead to significant decreases in crop yield and even to plant death. Chemical pesticides have made a great contribution to overcoming these biotic stresses and ensuring food safety. According to the Food and Agriculture Organization of the United Nations (FAO), the use of pesticides was increased from 2 million tons in 1990 to 4 million tons in 2019, mainly in

AMHA is a Naturally Occurring α-Amino Acid
Although AMHA was successfully isolated in the 1980s from a mutant strain of S. marcescens cultivated in medium supplemented with norvaline [19,20], it has long been regarded as an unnatural amino acid. To probe whether fungal strains naturally produce AMHA, extracts from mycelia of A. alternata were separated using medium-pressure liquid chromatography (MPLC). The chromatogram of the AMHA standard had a clear peak at a retention time of 4.44 min (Figure 1a). Based on the chromatographic profile for mycelia extracts (Figure 1b), a peak observed at 4.43 min corresponded to that of the AMHA standard. Therefore, A. alternata has the ability to produce free AMHA in the absence of norvaline. To further confirm this finding, the fraction of the peak at 4.43 min was collected, purified and crystallized. After white crystals were obtained (Figure 1c), its melting point was determined to be 268 °C (decompose), which was the same in prior research [19]. Subsequently, its chemical structure was identified by the infrared (IR), mass (MS) and nuclear magnetic resonance (NMR) spectra. The IR spectrum of the fraction was determined for the identification of functional groups. Two peaks observed at about 3400 cm −1 and 3325 cm −1 were attributed to the asymmetric and symmetric stretching vibration of NH2, respectively (Figure 2a). An absorption peak at 2899 cm −1 indicated that the fraction had the symmetric stretching The IR spectrum of the fraction was determined for the identification of functional groups. Two peaks observed at about 3400 cm −1 and 3325 cm −1 were attributed to the asymmetric and symmetric stretching vibration of NH 2 , respectively ( Figure 2a). An absorption peak at 2899 cm −1 indicated that the fraction had the symmetric stretching vibration of an OH group. The characteristic peak at 1727 cm −1 corresponded to the stretching vibration of a C=O group. An additional peak at 1214 cm −1 was associated with the stretching vibration of a C-N group. Thus, the crystalized compound had an amino acid structure.
vibration of an OH group. The characteristic peak at 1727 cm −1 corresponded to the stretching vibration of a C=O group. An additional peak at 1214 cm −1 was associated with the stretching vibration of a C-N group. Thus, the crystalized compound had an amino acid structure. The electrospray ionization mass spectrum of the compound depicted a significant ion peak at a retention time of 1.66 min in the positive mode [M+H] + (Figure 2b). Subsequently, the dominant peak was analyzed with MS spectrum, indicating that the m/z value of the compound was 146.1172 ( Figure 2c). The molecular formula of the compound was deduced as C7H15NO2 (Calculated m/z: 146.1180) by elemental composition analysis. This was in agreement with the literature [19,21] and data for the AMHA chemical standard.
To further confirm the chemical structure, 1 H and 13 C NMR spectra of the compound were detected on the basis of the structure information of standard AMHA. The 1 H NMR data displayed typical proton signals that included a proton of hydroxyl (OH) at δH 13.74 The electrospray ionization mass spectrum of the compound depicted a significant ion peak at a retention time of 1.66 min in the positive mode [M + H] + (Figure 2b). Subsequently, the dominant peak was analyzed with MS spectrum, indicating that the m/z value of the compound was 146.1172 ( Figure 2c). The molecular formula of the compound was deduced as C 7 H 15 NO 2 (Calculated m/z: 146.1180) by elemental composition analysis. This was in agreement with the literature [19,21] and data for the AMHA chemical standard.
To further confirm the chemical structure, 1 H and 13 C NMR spectra of the compound were detected on the basis of the structure information of standard AMHA. The 1 H NMR data displayed typical proton signals that included a proton of hydroxyl (OH) at δ H 13.74 (1 H, br), two protons of amino (NH 2 ) at δ H 8.39 ( (Figure 2d). A signal at δ H 2.50 was attributed to the protons of the dimethyl sulfoxide (DMSO) solvent. The 1 H NMR results are in good accordance with an early characterization of AMHA [19]. The obtained 13 C NMR spectra of the compound exhibited seven absorption peaks attributable to a C=O carbon atom at δ C 170.87 (C-1), two CH carbon atoms at δ C 56.72 (C-2) and δ C 34.95 (C-3), two CH 2 carbon atoms at δ C 33.87 (C-4) and δ C 20.12 (C-5) and two CH 3 carbon atoms at δ C 15.38 (C-7) and δ C 14.30 (C-6). Two CH 3 carbon atoms of solvent DMSO revealed a strong signal at δ C 39.51 (Figure 2e). Together, the NMR data and the IR and MS spectra indicate that the purified compound was AMHA. This unequivocally confirmed that the fungal strain of A. alternata growing in conventional medium naturally produced free AMHA.
To confirm the conformation of AMHA, the optical rotation was firstly measured, which was [α] 25 D + 29.0 (c 0.5, 6 M HCl). Computational calculation of electrostatic circular dichroism (ECD) spectrum of the (2S, 3S)-AMHA was performed using time-dependent density functional theory (TDDFT). Comparison of the computational spectrum with experimental ECD spectrum of AMHA allowed us to determine its absolute configuration. The overall line shape and band positions of the experimental ECD spectrum were in exact agreement with the computational ECD spectrum (Figure 2f). Therefore, it is demonstrated that the absolute configuration of AMHA isolated from A. alternata is assigned to (2S, 3S)-2-amino-3-methylhexanoic acid. This result is compatible with a previous report [19].

Free AMHA Is Produced by Different Fungal Species
To further corroborate that AMHA is a naturally occurring product, mycelial extracts of four different fungi cultivated for 1 to 7 days on Czapek medium were used to monitor the dynamics of AMHA production by ultra-performance liquid chromatography mass coupled to a mass spectrometer (UPLC-MS). As before, a strong signal peak of AMHA standard with an [M + H] + ion at m/z 146.1180 was observed at the 1.66 min retention time (Figure 3a). A calibration curve of the peak area versus standard concentration was plotted to estimate the content of AMHA in mycelia ( Figure 3b). All mycelial extracts had peaks corresponding to AMHA (m/z 146.1180) with retention times of 1.61 min for A. alternata and M. oryzae, 1.65 min for A. alternata f. sp. lycopersici and 1.58 min for A. brassicicola (Figure 3c). The peak signal intensity of the extracts of A. alternata and M. oryzae was distinctly stronger than that of another two necrotrophic fungi, A. alternata f. sp. lycopersici and A. brassicicola (Figure 3c). Production of AMHA by mycelia was detected on the second day with a sharp increase up until the fifth day, beginning to plateau at the sixth day. The highest production of free AMHA by A. alternata mycelia was reached at 6.41 µg g −1 FW on the seventh day ( Figure 3d). The amount of AMHA produced by the other species was substantially lower after seven days of culture. M. oryzae, A. alternata f. sp. lycopersici and A. brassicicola produced 1.72, 0.94 and 0.51 µg AMHA g −1 FW, respectively ( Figure 3d). Production of free AMHA by fungal mycelia might be ubiquitous. Interestingly, typical signals of AMHA with an [M+H] + ion at m/z 146.1180 were observed at retention times of 1.70 to 1.74 min for the mycelial extracts when the four fungi cultivated for 7 days in Czapek liquid medium ( Figure 4a). Conversely, a distinct signal peak of a fragment with m/z 146.1180 could not be detected in the fermentation broths of these four fungi (Figure 4b). Mycelia of A. alternata grown in liquid medium produced substantially more AMHA than the other three fungi. At 7 days of culture, the content of AMHA in the mycelia of A. alternata, M. oryzae, A. alternata f. sp. lycopersici and A. brassicicola was 4.04, 1.12, 0.66 and 0.34 μg g −1 FW, respectively ( Figure 4c). It is likely that AMHA is an endogenous secondary metabolite naturally produced by fungal mycelia because free AMHA was not detected in their fermentation broths ( Figure 4c). Comparatively, AMHA was isolated from the fermentation broths of S. marcescens mutant strain Ar130-1 as an exocrine secretion product in the presence of norvaline [20]. After 72 h incubation, the yield of AMHA in the broths was over 2 mg mL −1 , which is much higher than that obtained with the fungi in our study. In addition, AMHA was not detected in the leaves of Arabidopsis, winter wheat or rice ( Figure S1).   Interestingly, typical signals of AMHA with an [M + H] + ion at m/z 146.1180 were observed at retention times of 1.70 to 1.74 min for the mycelial extracts when the four fungi cultivated for 7 days in Czapek liquid medium ( Figure 4a). Conversely, a distinct signal peak of a fragment with m/z 146.1180 could not be detected in the fermentation broths of these four fungi (Figure 4b). Mycelia of A. alternata grown in liquid medium produced substantially more AMHA than the other three fungi. At 7 days of culture, the content of AMHA in the mycelia of A. alternata, M. oryzae, A. alternata f. sp. lycopersici and A. brassicicola was 4.04, 1.12, 0.66 and 0.34 µg g −1 FW, respectively ( Figure 4c). It is likely that AMHA is an endogenous secondary metabolite naturally produced by fungal mycelia because free AMHA was not detected in their fermentation broths ( Figure 4c). Comparatively, AMHA was isolated from the fermentation broths of S. marcescens mutant strain Ar130-1 as an exocrine secretion product in the presence of norvaline [20]. After 72 h incubation, the yield of AMHA in the broths was over 2 mg mL −1 , which is much higher than that obtained with the fungi in our study. In addition, AMHA was not detected in the leaves of Arabidopsis, winter wheat or rice ( Figure S1).

AMHA Induces Plant Resistance to Temperature Stress
Amino acids are known for having positive effects on plant growth, de and yield because of their role in biosynthesis of proteins, pigments, vitamins, c purine and pyrimidine bases and other non-protein nitrogenous compound example, γ-aminobutyric acid, a non-protein amino acid, is involved in reg

AMHA Induces Plant Resistance to Temperature Stress
Amino acids are known for having positive effects on plant growth, development and yield because of their role in biosynthesis of proteins, pigments, vitamins, coenzymes, purine and pyrimidine bases and other non-protein nitrogenous compounds [24]. For example, γ-aminobutyric acid, a non-protein amino acid, is involved in regulation of pollen tube growth to the ovary, C:N balance, cytosolic pH, oxidative stress and stress response, acting as a signaling molecule in plants [25]. γ-Aminobutyric acid accumulates rapidly in tissues when plants are exposed to both biotic (insect herbivory, pathogen infection) or abiotic (heavy metal, salinity, heat, cold, drought, anoxia, wounding) stresses, showing specific defensive roles [25,26]. Besides γ-aminobutyric acid, there are other amino acids, such as β-aminobutyric acid, glutamate and proline, that have also been proposed to play relevant roles in plant defense [27,28]. As a natural α-amino acid, whether AMHA play any important role in plant response to stress is completely unknown.
Temperature is a key factor governing plant growth and development. Extreme temperature is a frequent stress in agriculture, severely delaying crop growth and development, reducing production and quality and even causing plant death [29]. To investigate the possible role of AMHA as an inducer of plant resistance against temperature stress, seedlings of bentgrass, wheat, tomato and strawberry were treated with this amino acid before being subjected to extreme temperatures. Temperature stress in these plants is one of the most common and major production problems.
The high temperature regime severely affected, almost to death, the three crop plants, bentgrass, wheat and tomato (mock treatment with only of 0.02% Tween-20), compared with the control not exposed to heat stress (Figure 5a). AMHA pretreatment significantly increased the heat resistance of these three plants relative to the mock (Figure 5a). The heat injury index of wheat and tomato seedlings treated with increased concentrations of AMHA dramatically decreased under high temperature conditions compared with the mock (Figure 5b). At 10,000 nM AMHA, most wheat and tomato plants remained alive (Figure 5a) with 60% and 40% lower heat injury index than the mock, respectively (Figure 5b). For bentgrass, high temperature caused severe chlorosis, shriveling and death of seedlings when treated only with 0.02% Tween-20 ( Figure 5a). AMHA enhanced plant resistance against heat damage even at 10 nM (Figure 5a,b). Furthermore, shoot growth of the three crop plants was totally inhibited by high temperature conditions. Exogenous pretreatment with 10,000 nM AMHA distinctly alleviated such growth inhibition, resulting in 88%, 78% and 438% increases in shoot fresh weight for bentgrass, wheat and tomato, respectively, compared to the corresponding mock after 7 d of recovery from heat treatment (Figure 5a,c). Based on the severity of heat damage and total aboveground biomass, the best AMHA concentration to alleviate heat damage and enhance heat resistance was 100 nM for bentgrass, 10,000 nM for wheat, 1000 to 10,000 nM for tomato, respectively. Evidently, AMHA has high capability of inducing plant resistance against heat stress.
Tomato plants treated with 0.02% Tween-20 and subjected to low temperature conditions (−2 • C) suffered freezing damage on some leaf margins and tips, developing slightly rolling and wilting of leaves at 5 d of recovery after cold stress (Figure 6a). Strawberry plants exposed for 12 h at −4 • C and recovered for 5 d showed their leaves to be severely wilted or necrotic (Figure 6a). AMHA pretreatment visibly reduced damage from cold stress in both tomato and strawberry plants (Figure 6a,b). Compared with the mock, the cold injury index was reduced by 34%, 66% and 66% in tomato, and 50%, 70% and 80% in strawberry by the pretreatment with 10, 100 and 1000 nM AMHA, respectively ( Figure 6b). Moreover, the pretreatment with AMHA significantly alleviated shoot growth inhibition caused by cold stress (Figure 6a,c). At 1000 nM AMHA pretreatment, the aboveground biomass of tomato and strawberry increased by 24% and 84%, respectively, compared to the mock, which reached over 70% of the control plants not exposed to cold stress (Figure 6c). This suggests that AMHA is also active against cold stress in several low-temperature sensitive plants. Tomato plants treated with 0.02% Tween-20 and subjected to low temperature conditions (−2 °C) suffered freezing damage on some leaf margins and tips, developing slightly rolling and wilting of leaves at 5 d of recovery after cold stress (Figure 6a). Strawberry plants exposed for 12 h at −4 °C and recovered for 5 d showed their leaves to be severely wilted or necrotic (Figure 6a). AMHA pretreatment visibly reduced damage from cold stress in both tomato and strawberry plants (Figure 6a,b). Compared with the mock, the cold injury index was reduced by 34%, 66% and 66% in tomato, and 50%, 70% and 80% in strawberry by the pretreatment with 10, 100 and 1000 nM AMHA, respectively ( Figure 6b). Moreover, the pretreatment with AMHA significantly alleviated shoot growth inhibition caused by cold stress (Figure 6a,c). At 1000 nM AMHA pretreatment, the aboveground biomass of tomato and strawberry increased by 24% and 84%, exposed to cold stress (Figure 6c). This suggests that AMHA is also active against cold stress in several low-temperature sensitive plants.

AMHA Induces Plant Resistance to Pathogen Infection
Powdery mildew, caused by the fungal pathogen B. graminis, is one of the most destructive wheat diseases worldwide [30]. At 8 dpi, abundant grayish white patches of B. graminis had grown on the leaves of susceptible wheat plants pretreated with 0.02% Tween-20 (mock), but much less fungal growth was observed on those wheat leaves pretreated with AMHA at increasing concentrations (Figure 7a,b). Wheat plants were much stronger and straighter in the presence of AMHA compared with the mock as a result of decreased fungal growth on the surfaces of leaves and stems (Figure 7a,b). This suggests that AMHA pretreatment conferred wheat plants resistance to powdery mildew. Consistent with this observation, extensive chlorotic lesions were formed on the leaves of mock-treated wheat plants, and only a few chlorotic spots were found on the leaves of AMHA-treated plants (Figure 7c). These results were further supported by data of chlorophyll fluorescence imaging (Figure 7d). Images of maximum PSII quantum yield (FV/FM) can be used as a sensitivity indicator to visualize early tissue damage during pathogen infection [31]. B. graminis resulted in fading images of FV/FM in mocked-treated wheat leaves, indicating serious foliar damage caused by the pathogen. Pretreatment with AMHA prevented such damage in a concentration-dependent manner. At the highest concentration of 100 μM AMHA, only slight damage was found on the wheat leaves

AMHA Induces Plant Resistance to Pathogen Infection
Powdery mildew, caused by the fungal pathogen B. graminis, is one of the most destructive wheat diseases worldwide [30]. At 8 dpi, abundant grayish white patches of B. graminis had grown on the leaves of susceptible wheat plants pretreated with 0.02% Tween-20 (mock), but much less fungal growth was observed on those wheat leaves pretreated with AMHA at increasing concentrations (Figure 7a,b). Wheat plants were much stronger and straighter in the presence of AMHA compared with the mock as a result of decreased fungal growth on the surfaces of leaves and stems (Figure 7a,b). This suggests that AMHA pretreatment conferred wheat plants resistance to powdery mildew. Consistent with this observation, extensive chlorotic lesions were formed on the leaves of mock-treated wheat plants, and only a few chlorotic spots were found on the leaves of AMHA-treated plants (Figure 7c). These results were further supported by data of chlorophyll fluorescence imaging (Figure 7d). Images of maximum PSII quantum yield (F V /F M ) can be used as a sensitivity indicator to visualize early tissue damage during pathogen infection [31]. B. graminis resulted in fading images of F V /F M in mocked-treated wheat leaves, indicating serious foliar damage caused by the pathogen. Pretreatment with AMHA prevented such damage in a concentration-dependent manner. At the highest concentration of 100 µM AMHA, only slight damage was found on the wheat leaves (Figure 7d). This was in good agreement with the status of B. graminis grown on wheat leaves (Figure 7b). Our early experiments demonstrated that AMHA could not inhibit the growth of fungal pathogens on PDA medium, even at 1 mM concentration, including M. oryzae, Rhizoctonia solani, Fusarium moniliforme, F. graminearum, R. cerealis, Botrytis cinerea and Sclerotinia sclerotiorum (Table S1). This suggested that AMHA has no antifungal activity. Thus, we conclude that AMHA can effectively elicit wheat resistance against powdery mildew by suppressing the fungal growth and infection on plants. ( Figure 7d). This was in good agreement with the status of B. graminis grown on wheat leaves (Figure 7b). Our early experiments demonstrated that AMHA could not inhibit the growth of fungal pathogens on PDA medium, even at 1 mM concentration, including M. oryzae, Rhizoctonia solani, Fusarium moniliforme, F. graminearum, R. cerealis, Botrytis cinerea and Sclerotinia sclerotiorum (Table S1). This suggested that AMHA has no antifungal activity. Thus, we conclude that AMHA can effectively elicit wheat resistance against powdery mildew by suppressing the fungal growth and infection on plants. To test the activity of AMHA against bacterial disease, the model pathogen P. syringae DC3000 was used to infect Arabidopsis plants. P. syringae is the one of the most common plant pathogens because it infects almost all economically important crops [32,33]. P. syringae DC3000 also causes bacterial speck disease in its natural host, tomato [34]. P. syringae DC3000 caused severe chlorotic lesions in most leaves of mock-treated Arabidopsis plants at 7 dpi (Figure 7e). AMHA pretreatment greatly reduced chlorosis severity by inducing strong resistance in Arabidopsis against this bacterial disease. The level of plant resistance against P. syringae DC3000 gradually increased with AMHA concentration (Figure 7e). To estimate the effect of AMHA on bacterial growth, the number of bacteria in Arabidopsis plants was also measured after 4 days of P. syringae DC3000 inoculation. A concentrationdependent decrease in the bacterial population was observed in the presence of AMHA. Total bacterial number concentration of fresh Arabidopsis plants pretreated with 10, 100, 1000 and 10,000 nM AMHA, respectively, reduced by 30%, 51%, 65% and 95% compared to the mock-pretreated plants (Figure 7f). Previous studies showed that 1 mM concentration of AMHA were only slightly antagonistic to P. aeruginosa and were not obviously antagonistic to P. fluorescens [19]. Clearly, bacterial growth was strongly suppressed in AMHA-pretreated plants, demonstrating that AMHA-induced resistance response of Arabidopsis plants triggered protection against P. syringae DC3000. These results confirm that AMHA can activate plant immunity against the virulent P. syringae DC3000 due to the inhibition of bacterial growth in plant tissues.
Besides fungi and bacteria, virus are also important plant pathogens. On the basis of scientific and economic importance, Tomato spotted wilt virus (TSWV) is among one of most destructive plant viruses in the world [35]. TSWV has global distribution and a very wide host range of over 1000 plant species from more than 80 families, causing crop losses at above 1$ billion annually [35,36]. To investigate whether AMHA can protect plants against TSWV infection, tobacco leaves were sprayed with AMHA followed with the infiltration of the Agrobacterium carrying TSWV SR (+)eGFP + M (−)opt + L (+)opt constructs. At 7 days post TSWV inoculation, eGFP fluorescence expression of mock-pretreated tobacco plants was highly abundant in a group of cells; however, eGFP fluorescence of AMHA-pretreated plants was sporadically expressed in leaf tissues (Figure 7g). The average intensity of eGFP fluorescence signal was significantly decreased by 37%, 45% and 46% in leaves pretreated with 0.1, 1 and 10 nM AMHA compared to the mock-treated leaves, respectively (Figure 7h). These results suggest that AMHA pretreatment could also effectively induce plant defense against TSWV infection even at a very low concentration.

Fungus Strains and Chemicals
The A. alternata (Fr.) Keissler wild-type strain NEW001 was isolated from the invasive plant Ageratina adenophora [37]. The stock solution of AMHA isolated from A. alternata at 1 mM was prepared with distilled water.

Cultivation of Fungi
Fungal strains were maintained on potato dextrose agar (PDA) medium. Cultivation experiments were performed on Czapek medium as previously described [38]. The Czapek medium (pH 5.5) contains 40 g D-glucose, 1 g NaNO 3 , 0.25 g NH 4 Cl, 1 g KH 2 PO 4 , 0.25 g KCl, 0.25 g NaCl, 0.5 g MgSO 4 ·7H 2 O, 0.01 g FeSO 4 ·7H 2 O, 0.01 g ZnSO 4 ·7H 2 O, 1 g yeast extract, 15 g agar (for solid medium) and enough distilled water to make one liter. A 5-mm-diameter agar disc was taken from the margin of mycelia of A. alternata, A. alternata f. sp. lycopersici, A. brassicicola and M. oryzae grown on PDA medium, and transferred to a new 100-mm-diameter Petri dish (Corning, NY, USA) containing 20 mL of Czapek solid medium. Fungi were allowed to grow at 25 • C for 7 days in darkness. For liquid cultivation, five 5-mm-diameter agar discs taken from the margin of four strains grown on PDA medium were transferred to a new 250 mL flask containing 100 mL of Czapek liquid medium and grown at 25 • C, 140 rpm for 7 days in darkness. Fresh mycelia and Czapek fermentation broths were separated for further experiments by centrifugation at 1000× g for 10 min (Allegra TM 64R centrifuge, Beckman, Coulter, CA, USA).

Extraction, Purification and Structure Analysis
After 7 days of growth in Czapek solid medium, 1 g A. alternata mycelia were harvested and homogenized in liquid nitrogen. The mycelial powder was put into a 10 mL tube containing 5 mL distilled water and completely mixed by shaking for 60 min to extract AMHA at room temperature.
The supernatant was transferred to a new 10 mL tube after centrifugation at 5500× g for 10 min. The crude extracts were dried for 24 h under a vacuum in an LGJ-10 lyophilizer (Xinyi Inc., Beijing, China) and redissolved in 1 mL distilled water. The extract was filtered through a GF/B 0.22-µm glass microfiber filter (Whatman, Kent, UK) and stocked in a clean 2 mL tube. The isolation and purification of AMHA was performed on a MPLC (NGC TM Quest 10 Plus Chromatography System, Bio-Rad Laboratories, Inc., Hercules, CA, USA) equipped with a SunFire TM Prep C 18 column (10 × 250 mm, 10 µm) (Waters, MA, USA). The detection wavelength was set at 210 nm. The mobile phase was a water solution containing 0.1% formic acid (v/v) and acetonitrile (60: 40, v/v). For each separation, 50 µL samples were injected and passed at a flow rate of 2.0 mL min −1 at ambient temperature. The major components with a retention time of 4.4 min as the standard were collected and freeze-dried for 24 h under vacuum to obtain the purified white crystals. The melting point of white crystals was measured with a MPA100 automated melting point system (Stanford Research Systems, California, CA, USA). Optical rotation was determined using an Autopol IV automatic polarimeter (Rudolph Research Analytical, Newburgh Road Hackettstown, Hackettstown, NJ, USA) with 6 M HCl as solvent. ECD spectra were measured by a JASCO J-1500 CD Spectrophotometer (JASCO Co., Tokyo, Japan).
The IR spectra of the crystals were recorded according to Bagheri et al. (2018) [39]. The IR data were acquired on an FTIR system (Nicolet TM 460 FT-IR spectrophotometer, Thermo Fisher Scientific, Waltham, MA, USA). IR data were analyzed on OMNIC software (Thermo Fisher Scientific).
The crystals were analyzed using an acquity UPLC system (Waters, Milford, MA, USA) coupled to a Waters G2-XS Q-TOF mass spectrometer, as previously described with minor modifications [40]. UPLC separations were performed with a binary solvent delivery system and an auto-sampler on a Waters Acquity UPLC BEH C 18 column (2.1 mm × 100 mm, 1.7 µm particles). The flow rate was 0.4 mL min −1 . The injection volume of samples was 2 µL. The detection wavelength was 280 nm. Two consecutive isocratic mobile phase mixtures consisted of water containing 0.1% formic acid and 2 mM NH 4 Ac (solvent A) and acetonitrile containing 0.1% formic acid and 2 mM NH 4 Ac (solvent B). The gradient program was as follows: within 14 min, 5% solvent B was held for 1 min, then 5% solvent B was increased within 9 min to 95% and held for an additional 2 min and then 95% solvent B was decreased within 1 min to 5%, which was held for 1 min.
Mass detection was performed using electrospray source in positive ion mode with MS acquisition mode at a selected mass range of 50-1200 m/z. The lock mass option was enabled using leucine-enkephalin (m/z 556.2771) for recalibration. The ionization parameters were set as follows: capillary voltage of 3.0 kV, cone voltage of 30 V, source temperature at 120 • C, and desolvation gas temperature at 400 • C. The collision energy was 20-40 eV. Data acquisition and processing were performed using Masslynx 4.1 (Waters, MA, USA). Extraction of centroid spectra peaks with a width of 0.01 Da was used to determine the extracted ion chromatograms (EICs) from the total ion chromatogram (TIC).

Conformational Analysis, Geometrical Optimization and ECD Calculation of (2S, 3S)-AMHA
The absolute configuration analysis of AMHA was carried out on the basis of the systematic conformation search using Molecular Operating Environment (MOE) software version 2016 (Chemical Computing Group Inc., Montreal, QC, Canada) under the MMFF94 force field. An energy cut of 2.5 kcal·mol −1 was chosen in order to select a wide distribution of conformers. The stable conformers subjected to ECD calculation were re-optimized using density functional theory (DFT) at the B3LYP/6-31G (d) level in Gaussian 09 software (Gaussian Inc., Wallingford, CT, USA) [41]. Seven lowest conformers were selected for theoretical calculation of ECD, which was performed using TDDFT method at the B3LYP/6-311+G (d, 2p) level with CPCM solvation model in water. The ECD spectra were simulated using the overlapping Gaussian function by applying a half-bandwidth at 1/e peak height, σ = 0.20 for all. The final computational ECD spectra were obtained according to the Boltzmann-calculated contribution of each conformer.

Test of AMHA in Different Fungus Strains and Plants
To quantitatively analyze the yield of AMHA in A. alternata, A. alternata f. sp. lycopersici, A. brassicicola and M. oryzae, 1 g fresh mycelia of each of the four strains grown in Czapek solid or liquid medium were collected. Extraction of AMHA in mycelia and UPLC-MS analysis were carried out in the same way described before. The content of AMHA in Czapek fermentation broths of four trains was also measured. After centrifugation, the separated fermentation broths of each strain were transferred to a new 250 mL flask. Then, 1 mL of fermentation supernatant was taken and filtered through a 0.22-µm glass microfiber filter (Whatman, Kent, UK). The filtrate was collected in a new 2 mL tube. All the samples were transferred to new 2 mL short thread vials with pre-slit blue short screw (Beijing Labgic Technology Co., Ltd., Beijing, China) before UPLC-MS analysis. Quantification of natural AMHA was performed by employing the external calibration curve. The standard AMHA was dissolved in distilled water to 100 mg L −1 as stock solution. The working standard solutions were prepared at 0.25-8 mg L −1 from the stock solution by gradient dilution with distilled water.
To conform whether AMHA could be produced in plants, 1 g fresh leaves of 21-day-old Arabidopsis thaliana ecotype Col-0 plants, two-leaf stage seedlings of winter wheat (Triticum aestivum) and rice (Oryza sativa) grown in greenhouse were collected. Extraction in plants and UPLC-MS analysis were carried out in the same way described for fungi.

High/Low-Temperature Resistance Assays
Seeds of winter wheat (T. aestivum), bentgrass (Agrostis capillaris) and tomato (Lycopersicon esculentum) cultivar "Zhongsu 4" were purchased from Jiangsu Mingtian Seed Science and Technology Co. (Nanjing, China). Thirty wheat seeds, 0.2 g of bentgrass seeds and three tomato seeds per pot were planted separately in peat-vermiculite mixtures (1:1, v/v) in 12-cm-diameter plastic pots and grown at 25 • C under 200 µmol m −2 s −1 illumination with 12 h photoperiod. Strawberry (Fragaria ananassa) cultivar "Hongyan" seedlings at the 5-leaf stage were obtained from T. Gu (College of Horticulture, NAU, China) and then transplanted to 12-cm-diameter pots (one plant per pot) containing peat-vermiculite substrate (1:1, v/v) and cultivated for 7 days at 25 • C under 200 µmol m −2 s −1 light for 12 h. Fourteen-day-old wheat seedlings, twenty-five-day-old bentgrass seedlings and eighteen-day-old tomato seedlings were pretreated with AMHA before high temperature treatment. The adapted strawberry seedlings and 45-day-old tomato seedlings were treated with AMHA prior to subjecting them to low temperature stress. Plants were sprayed twice using distilled water plus 0.02% (v/v) Tween-20 (mock) or AMHA (10, 100, 1000 and 10,000 nM) plus 0.02% (v/v) Tween-20 with a 24-h interval by a mist sprayer (SKS Bottle & Packaging Inc., Saratoga Springs, NY, USA) until the leaf surfaces were completely wet.
After 24 h of the second spraying, the pretreated bentgrass and wheat plants were exposed to 45 • C for 9 h under 200 µmol m −2 s −1 illumination and transferred to their normal growth conditions for 7 days for recovery. The pretreated tomato seedlings were subjected to 42 • C for 9 h under light with 200 µmol m −2 s −1 and recovered for 7 days. For low temperature treatment, the pretreated tomato and strawberry plants, respectively, were subjected to −2 • C for 24 h and −4 • C for 12 h under light of 200 µmol m −2 s −1 and recovered for 5 days under normal growth conditions. After recovery, plants were photographed using a digital camera (Canon model SX11S, Tokyo, Japan). Heat injury level per plant was rated on a 0-5 scale: 0, no damage; 1, less than 25% of leaf area wilted; 2, 25%-50% of leaf area wilted or yellow; 3, 50%-75% of leaf area wilted or yellow; 4, more than 75% of leaf area wilted or yellow; 5, whole plant died. Cold injury degree per plant was recorded with the following scale: 0, no damage; 1, leaf edge slightly wilted; 2, wilting less than 25% of total leaf area; 3, 25%-50% of leaf area wilted and several spots; 4, more than 50% of leaf area wilted or spots; 5, leaf completely wilted. All leaves of each plant were assessed for injury. The injury index was calculated according to the formula: injury index = 100 × ∑(v × n)/(5 × N), here n is the number of plants in a scale, v is the scale value, and N is the total number of plants tested. Finally, the seedlings of each pot were collected by cutting the stems just above soil surface for fresh weight determination. Plants of four pots were used in each experiment as four independent biological replicates.

Wheat Resistance to Fungus Powdery Mildew
Wheat powdery mildew (Blumeria graminis f. sp. tritici) (Bgt) strains were maintained on seedlings of susceptible wheat variety NAU0686 in a spore-proof glasshouse. Sixty wheat seeds were planted in a soil-peat (3:1, v/v) substrate mixture in 18-cm-diameter plastic pots at 22 • C under 100 µmol m −2 s −1 light with 16: 8 h photoperiod and 70% relative humidity in a controlled culture room. Sixty wheat seeds were sown per pot. The seedlings at the two-leaf stage were sprayed three times with AMHA solutions (15 mL per pot) at different concentrations (1, 10 and 100 µM) containing 0.02% Tween-20 or with 0.02% Tween-20 alone (mock) with a 24-h interval using a SKS mist sprayer. After 24 h following the third spraying, the pretreated seedlings were inoculated with powdery mildew conidia (about 15 to 20 conidia on leaf surface per cm 2 ) according to Wang et al. (2014) [30], and then incubated at 22 • C under 100 µmol m −2 s −1 with 16-h photoperiod.
Eight days after inoculation, whole plants and disease symptoms of representative leaves were photographed as before. Simultaneously, chlorophyll fluorescence images of F V /F M of these representative leaves were also measured under a pulse-modulated Imaging-PAM Mseries fluorometer (MAXI-version, Heinz Walz GmbH, Effeltrich, Germany), as previously described [31]. Plants of three pots were used in each experiment as three independent biological replicates.

Arabidopsis Resistance to Bacterial Infection
A. thaliana ecotype Col-0 was grown in a peat-vermiculite substrate mixture (1:3, v/v) at 22 • C under 16 h photoperiod with a light-intensity of 100 µmol m −2 s −1 . Three-weekold healthy plants were sprayed twice to runoff with 0.02% Tween-20 (mock) or four concentrations (10, 100, 1000 and 10,000 nM) of AMHA solution containing 0.02% Tween-20 with a 24-h interval using a SKS mist sprayer, and then incubated for 24 h after a second spraying under the same growth conditions before inoculation with bacteria. P. syringae DC3000 was cultured on Luria-Bertani (LB) media with 50 µg mL −1 rifampicin at 28 • C for 48 h. Bacteria were collected from LB media and suspended in 10 mM MgCl 2 solution. Bacterial cell densities (OD 600 ) were adjusted to 0.2 (equal to 5 × 10 8 cells mL −1 ) and then further diluted with 10 mM MgCl 2 solution with 0.02% Tween-20 to 5 × 10 6 cells mL −1 for inoculation. The pretreated Arabidopsis plants were inoculated with P. syringae DC3000 suspension and incubated under continuous light (about 100 µmol m −2 s −1 ) at 22 • C. Four days post-inoculation (dpi), leaves were collected from at least three different plants (four leaves per plant) and weighed. Subsequently, leaves were homogenized to measure bacterial population, as previously described [34]. The colony forming units (CFU) were normalized as CFU mg −1 using the fresh weight of leaves. Seven days after inoculation, disease phenotypes of whole plants were also observed and photographed. Each experiment has three pots with four plants in each.

Tobacco Resistance to Virus Infection
Tobacco (Nicotiana benthamiana) plants were grown at 25 • C under 100 µmol m −2 s −1 light with 16 h photoperiod. Prior to virus infection, 8-week-old plants were pretreated twice by spraying with 0.02% Tween-20 (mock) or 0.1, 1 and 10 nM AMHA solution containing 0.02% Tween-20 with a 24-h interval using a SKS mist sprayer, and then incubated 24 h after second spraying. Tomato spotted wilt virus isolate from asparagus lettuce (Lactuca sativa) (TSWV-LE) was used in this work [36]. Leaves of the pretreated tobacco plants were infiltrated with a 1:1:1 mixture of Agrobacterium, respectively, carrying infectious clones of TSWV SR (+)eGFP , M (−)opt and L (+)opt . The infectious clones of TSWV were described by Feng et al. (2020) [36]. Seven days post infiltration, the accumulation of TSWV in the infected leaves was visualized by monitoring eGFP fluorescence expression from the SR (+)eGFP replicon under an Axio Imager M2 fluorescence microscope (Carl Zeiss Microscopy GmbH, Oberkochen, Germany). The fluorescence images were quantified using Image J v1.8.0 software (National Institute of Health, Bethesda, Rockville, MD, USA). Plants of three pots were used in each experiment as three independent biological replicates.

Statistical Analysis
One-way ANOVA was carried out and means were separated by Duncan LSD at 95% using SPSS Statistics 26.0 (IBM ® , Armonk, NY, USA).

Conclusions
Firstly, it has been demonstrated that AMHA is a naturally occurring endogenous α-amino acid since free AMHA can be produced by fungi including M. oryzae and three Alternaria species in a conventional medium. Furthermore, AMHA indeed exhibits excellent activity against extreme temperature stress as well as against fungal, bacterial and viral diseases by enhancing plant resistance at the concentration range of 0.1-100,000 nM. Undoubtedly, AMHA is a unique plant elicitor with great potential.
Before AMHA can be commercialized as a crop protection product, it is advisable to determine its physio-biochemical and molecular mechanisms for inducing plant resistance against abiotic and biotic stresses. Additionally, greenhouse and field experiments are also essential for assessing the activity spectrum of AMHA as a potent plant elicitor and developing practical technical frameworks for its use in crop protection.

Patent
The authors have obtained a Chinese patent (ZL202011549486.6) based on the results reported in this paper.