Degradation of cyanobacterial neurotoxin β-N-methylamino-l-alanine (BMAA) using ozone process: influencing factors and mechanism

β-N-methylamino-l-alanine (BMAA), which has been considered as an environmental factor that caused amyotrophic lateral sclerosis/parkinsonism-dementia complex (ALS/PDC) or Alzheimer’s disease, could be produced by a variety of genera cyanobacteria. BMAA is widely present in water sources contaminated by cyanobacteria and may threaten human health through drinking water. Although oxidants commonly used in drinking water plants such as chlorine, ozone, hydrogen peroxide, and hydroxyl radicals have been shown to effectively degrade BMAA, there are limited studies on the mechanism of BMAA degradation by different oxidants, especially ozone. This work systematically explored the effectiveness of BMAA ozonation degradation, investigated the effect of the operating parameters on the effectiveness of degradation, and speculated on the pathways of BMAA decomposition. The results showed that BMAA could be quickly eliminated by ozone, and the removal rates of BMAA were nearly 100% in pure water, but the removal rates were reduced in actual water. BMAA was primarily degraded by direct oxidation of ozone molecules in acidic and near-neutral conditions, and indirect oxidation of •OH accounted for the main part under strong alkaline conditions. The pH value had a significant effect on the decomposition of BMAA, and the degradation rate of BMAA was fastest at near-neutral pH value. The degradation rates of TOC were significantly lower than that of BMAA, indicating that by-products were generated during the degradation process. Three by-products ([M-H]+  = 105, 90, and 88) were identified by UPLC-MS/MS, and the degradation pathways of BMAA were proposed. The production of by-products was attributed to the fracture of the C–N bonds. This work is helpful for the in-depth understanding on the mechanism and demonstration of the feasibility of the oxidation of BMAA by the ozone process. • The reaction of ozonation BMAA was easy to occur. • The degradation rate was fast under near-neutral conditions. • Direct oxidation under neural conditions was the main pathway for ozone degradation of BMAA. • Three products were detected, and the reaction path was inferred.


Introduction
As an ancient aquatic prokaryote, cyanobacteria are widely existing in all over the world. Cyanobacteria can produce some substances harmful to human health called cyanobacterial toxin. β-Methylamino-l-alanine (BMAA), which was a kind of cyanobacterial secondary metabolite, was first extracted from the seed of Cycas micronesica by Vega and Bell (1967). It has turned out to damage the human nervous system and cause neurodegenerative diseases (Chernoff et al. 2017;Murch et al. 2004;Rao et al. 2006). Some papers reported that BMAA could indirectly destroy the nervous system by enhancing the activity of other nerve inhibitors at low concentrations, while it could directly destroy nerve cells at high concentrations (Blaszczyk et al. 2021;Lobner et al. 2007). Since various species of cyanobacteria, dinoflagellates, and diatoms can produce BMAA (Cox et al. 2005;Jiang et al. 2014;Reveillon et al. 2016;Wang et al. 2021), the pollution of BMAA has become a worldwide environmental problem (Main et al. 2018;Metcalf et al. 2020;Wang et al. 2021;Wu et al. 2019). The life cycle of cyanobacteria mainly includes growth period, peak period, and decline period. Cyanobacteria reproduce rapidly during both growth period and peak period, while a large number of cyanobacteria cells will lose their vital activity at the decline period. Our previous studies have shown that BMAA can be released into the water through the lysis of dead cells in the decline periods (Yan et al. 2020a). When the alga-containing water was used as the water source, humans may ingest BMAA through drinking the water. From the perspective of prevention, the study of BMAA treatment technology is of great significance for guaranteeing the security of drinking water and solving the problem of BMAA pollution in drinking water treatment plants (DWTPs).
Coagulation, precipitation, and sand filtration processes are ineffective for the removal of soluble organic pollutants in DWTPs. Although the chlorination disinfection process can degrade some natural organic matter, there is a risk of producing chlorinated disinfection by-products with carcinogenic, teratogenic, and hereditary toxicity (Kali et al. 2021). It was confirmed that it would take more than 2 h for the degradation rate of BMAA to reach more than 90% in actual water in our previous work (Yan et al. 2020b). Moreover, it was reported by Cao et al. (2019) that the yields of disinfection by-products in the reaction of chlorination of BMAA were relatively higher. Meanwhile, advanced drinking water treatment technologies which target the removal of organic matters, such as ozone oxidation, activated carbon absorption, and membrane filtration, are increasingly widely used in DWTPs. Since BMAA is one of the hydrophilic polar small-molecule organic matters (Jiang et al. 2012), the removal efficiency of activated carbon adsorption and membrane filtration may be limited. In addition, the use of activated carbon adsorption and membrane filtration to remove pollution is a physical process, and subsequent regeneration processes and treatment of concentrated solutions are needed (Radjenovic et al. 2008;Suhnholz et al. 2018). And ozone is frequently used for sterilization, disinfection, removal of odorous substances, and decomposition of pollutants in DWTPs (Carbajo et al. 2015;Tay and Madehi 2015). It also showed excellent performance in the degradation of cyanobacterial toxins such as microcystins, anatoxin-a, and cylindrospermopsin (Chang et al. 2014;Rodriguez et al. 2007;Zhang et al. 2021). Chen et al. (2018) have studied the degradation rates of BMAA by permanganate, chlorine, ozone, hydrogen peroxide, and hydroxyl radical, and the results showed that hydroxyl radicals degrade BMAA at the fastest rate, followed by ozone and permanganate as the slowest. However, Chen et al. (2018) mainly focused on the comparison of the degradation efficiency of BMAA with different oxidants and did not conduct separate experiments on the effects of other operating parameters other than pH on the BMAA degradation efficiency of different oxidants. The factors affecting the degradation rate of BMAA in UV/H 2 O 2 (Yan et al. 2022) systems and during chlorination (Yan et al. 2020b) have been studied in our previous work, while the impact of the operating parameters on the efficiency of ozone oxidation of BMAA has not been reported so far.
Ozone, which is an extremely strong oxidant with an oxidation potential of 2.07 V, is able to effectively oxidize many inorganic and organic compounds (Jiao et al. 2019;Lim et al. 2022;Wang and Zhuan 2020), and its oxidation performance is higher than that of chlorine and potassium permanganate. Because ozone can decompose itself in a short time without causing secondary pollution (Joseph et al. 2019), it is an environmentally friendly oxidant. Ozone oxidation of organic pollutants in aqueous solutions included direct and indirect reactions (Chen et al. 2012). The oxygen atoms in the ozone molecule have strong electrophilic or nucleophilic properties, can directly react with the nucleophilic position of organic compounds to substitute the partial structure of compounds, and can also take place the nucleophilic addition reactions with compounds containing carbonyl groups or double and triple nitrogen bonds (Joseph et al. 2021). Ozone is prone to hydrolysis to produce the hydroxyl radical (•OH) and other active groups under alkaline conditions, and organic compounds can be indirectly degraded by these active groups in ozone aqueous solutions (Wang and Chen 2020). Compared with the direct reaction, the indirect reaction has a faster reaction rate and no selectivity (Malik et al. 2020). In order to explore the effectiveness of ozone process for the removal of BMAA in DWTPs, it is necessary to thoroughly investigate the mechanism and degradation pathways of ozone oxidation of BMAA.
To this end, this work focused on the degradation mechanism of BMAA by ozone. The objectives mainly include (1) evaluating the effectiveness of ozone oxidative degradation of BMAA, (2) studying the influence of different operating parameters on the degradation efficiency of BMAA, (3) exploring the efficiency of ozone process to remove BMAA in actual water, and (4) analyzing the conversion products and speculating potential reaction pathways.

Chemicals
L -BMAA hydrochlor ide (≥ 97%, powder) and nitrobenzene (≥ 99%, ACS reagent) were purchased from Sigma-Aldrich Co. LLC. (USA), D-2,4-diaminobutyric acid-2 D 5 dihydrochloride was provided by C/D/N Isotopes Inc. (Canada), and sodium thiosulfate, citric acid, and hydrochloric acid were obtained from Aladdin Industrial Corporation (Shanghai, China). The HPLC grade acetonitrile, methanol, and formic acid were provided by Sigma-Aldrich Co. LLC. (USA). Pure water (18 MΩ quality or better) was produced with a Milli-Q water system (Millipore Ltd., Bedford, MA, USA). The stock solution with the concentration of 100 mg/L was prepared with pure water, and it was stored at 0 °C to make sure that the solution is stable for more than six months (Combes et al. 2013).

Analytical methods
The concentration of BMAA was measured by an ultraperformance liquid chromatography system equipped with a tandem mass spectrometry (UPLC-MS/MS, Waters Corporation, Milford, USA). The separation of BMAA from other organic compounds was carried out with a hydrophilic interaction chromatography column (ACQUITY UPLC BEH, 150 mm, 2.1 mm, 1.7 mm particle size), purchased at Waters Corporation. Acetonitrile containing 0.1% (v/v) formic acid and water containing 0.1% (v/v) formic acid were used as mobile phase for isocratic elution in a 70:30 ratio. The flow rate of a mobile phase was set to 0.4 mL·min −1 , and the column temperature was controlled to 30 °C. Because BMAA is positively charged in acidic conditions, the positive mode was utilized in mass spectrometry. The voltages of cone and capillary were set at 4 kV and 15 V, respectively, and the temperature and gas flow rate of desolvation were set at 650 °C and 1000 L·Hr −1 , respectively. Multiple reaction monitoring (MRM) containing four transitions (119 > 102, 119 > 88, 119 > 76, and 125 > 107) was applied. Additional detailed information about BMAA detection method could be found in our previous paper (Yan et al. 2017).
The concentrations of ozone in aqueous solutions were measured according to the indigo method. Three milliliters of indigo solutions was added into 5 mL of ozone solutions at pH 2, and then, the mixture was diluted to 25 mL. The absorbance was measured at 612 nm with an ultraviolet spectrophotometer (DR6000, Hach Company, Colorado, USA). The concentrations of nitrobenzene (NB) were analyzed with ultra-performance liquid chromatography with a photodiode array detector (UPLC-PDA, Waters Corporation, Milford, USA) using a C18 column (50 mm × 4.6 m), and the detection wavelength was 262 nm. The pH values of the solutions were determined with a PHS-25 precision meter (Yueping Scientific Instrument Co., Ltd., Shanghai, China), and total organic carbon (TOC) was determined by a TOC-VCPN analyzer (Shimadzu, Japan).

Experimental procedures
All experiments were carried out in a cylindrical batch reactor with a lid (Fig. S1), and the reactor was placed on a magnetic stirrer to mix the solutions enough. Two hundred milliliters of pure water was added to the reactor; 10 mM phosphate buffer was added to control the pH values during the reaction and adjust the initial pH of the water with H 2 SO 4 and NaOH. Ozone was generated by an ozone generator (CH-ZTW6G, Guangzhou Chuanghuan Ozone Electric Appliance Co., Ltd., China) and was introduced into the reactor through a sand core aeration head, and the aeration was stopped when the concentration of ozone reached the set values. A certain volume of BMAA stock solution was quickly poured into the ozone aqueous solution and timing was started. Onemilliliter samples were collected at 0, 5, 10, 30, 60, 120, 180, and 300 s, and 0.1 M sodium thiosulfate was quickly added into the samples to stop the reactions.
In the experiments to explore the effect of the operating parameters on the degradation of BMAA, the ranges of change of the initial ozone concentration, the initial BMAA concentration, the pH value of solution, and the temperature of reactions were 1 ~ 2.4 mg·L −1 , 0.6 ~ 2 mg·L −1 , 4 ~ 2, and 10 ~ 30 °C, respectively. In the experiment to determine whether the ozone degradation of BMAA was through direct or indirect reaction under different pH values, 10 mM tert-butanol was applied to capture •OH. In the experiments of TOC analysis and by-product identification, the concentrations of initial ozone and BMAA increased to 7 mg·L −1 and 5 mg·L −1 to facilitate instrumental analysis. All experiments were replicated at least three times.

Results and discussion
Degradation efficiency of BMAA with ozone As shown in Fig. 1 (the red line), the decomposition of BMAA was very fast with ozone, and the degradation rate of BMAA cloud was close to 100% within 60 s, which implied that the ozone process can effectively solve the pollution of BMAA. As mentioned above, the degradation of organic matters by ozone oxidation mainly includes two ways: direct reaction and indirect reaction. Then, which reaction is the ozone degradation of BMAA mainly carried out?
In order to identify the main reaction of BMAA degradation in ozone aqueous solutions, tert-butanol was added into the reaction solutions. The reaction rate constant of •OH and tert-butanol is 6.0 × 10 8 M −1 ·s −1 , while the reaction rate constant of ozone and tert-butanol is only 3.0 × 10 −3 M −1 ·s −1 , so tert-butanol can effectively inhibit •OH and has no effect on the direct oxidation reaction of ozone. It can also be seen in Fig. 1 that the degradation rates of BMAA with and without tert-butanol were similar. The degradation rates of BMAA with tert-butanol were only about 2% lower than that without tert-butanol at 120 s. Therefore, it can be inferred that ozone degraded BMAA mainly through the direct oxidation of ozone molecules at pH 7. In the study of Chen et al. (2018), it was also considered that ozone molecules can directly degrade BMAA. It has been reported that ozone molecules can degrade a variety of organic pollutants such as flumequine (Feng et al. 2016), ciprofloxacin (CIP), and oxytetracycline (OTC) (Oncu and Balcioğlu 2013).

Effect of pH value
The particle morphology of BMAA and oxide species in ozone aqueous solutions are affected by pH value, so it is vital to study the effect of the pH values of the solution on the degradation of BMAA in ozone aqueous solutions. The pH values of the solutions were adjusted to 4, 6, 7, 8, 10, and 12. It can be viewed in Fig. 2A that the pH value had a significant effect on the removal of BMAA by ozone oxidation. At 60 s of reactions, the degradation rates of BMAA at pH 4, 6, 7, 8, 10, and 12 were 23.54%, 97.35%, 97.93%, 98.96%, 83.35%, and 42.08%, respectively. Degradation rates of BMAA first accelerated and then slowed down with the increase of pH values. And strong acid and strong alkali environments were not conducive to the degradation of BMAA in ozone aqueous solutions.
As described in the "Degradation efficiency of BMAA with ozone" section, the decomposition of BMAA was mainly attributed to the direct oxidation of ozone molecules. It is well known that ozone molecules can be decomposed to generate hydroxyl radicals, of which oxidation potential is higher than that of ozone, under alkaline conditions. And the steady-state concentration of hydroxyl radicals increased with increasing pH values. In order to explore the ratio of direct oxidation to indirect oxidation at different pH values, tert-butanol, which could quench hydroxyl radicals, was added to the reaction systems at pH 4, 6, 8, 10, and 12. It can be seen in Table 1 that the contribution of hydroxyl radicals enhanced with the increase of pH value. The direct oxidation was the main pathway under acidic conditions, while because of the increase of the steadystate concentrations of hydroxyl radicals, the indirect reaction accounted for more than 50% in 120 s at pH 12, and indirect oxidation became the main reaction of BMAA degradation. Since the reaction rate constant between BMAA and hydroxyl radical is higher than that between BMAA and ozone, the rate of ozonation degradation of BMAA should be accelerated with increasing pH values, whereas degradation rates of BMAA were decreased in the range of pH 8 to 12. Therefore, it can be inferred that the difference in the composition of oxidative active species at different pH values may not be the main reason for the discrepancy in the degradation rate of BMAA at different pH values.
There are two amino groups and one carboxyl group in the BMAA molecule, and the pK a values of BMAA have been determined to be 2.1, 6.63, and 9.76 (Diaz-Parga et al. 2018), so there may be four forms of BMAA, namely BMAA 2+ , BMAA + , BMAA 0 , and BMAA − in the aqueous solutions. It can be observed in Fig. 2B that the concentrations of BMAA 2+ in the solutions could be negligible in the pH range used in this work, and the proportion of the remaining three species changed with the pH values of the solutions. The degree of deprotonation of the amino group of BMAA increased as the pH value rose. It can be seen in Fig. S3 that there are zero or one, one, and two deprotonated amino groups in the structure of BMAA + , BMAA 0 , and BMAA − , respectively. Some researchers have proposed that the deprotonated amino groups could reduce the steric hindrance of BMAA (Wei et al. 2013), increase the electron-donating ability of BMAA (Ao and Liu 2017), and accelerate the reaction with ozone and radicals . Therefore, the reason why the degradation rates of BMAA by ozone increased with the increase of pH value in the range of 4 ~ 8 was the enhancement of the degree of deprotonation of BMAA.
If the changes in the particle morphology of BMAA were only considered at different pH values, the degree of deprotonation of BMAA increased with the increase of pH value, and the degradation rate of BMAA in ozone solution should be accelerated, but the experimental results are contrary to it. The same phenomenon was observed in our previous work using UV/PS system to degrade BMAA, and the degradation rate of BMAA was the highest at pH 8 . The decrease of degradation rate of BMAA at pH above 8 was attributed to the electrostatic repulsion between SO 4 •− and BMAA − hindering particle collision in the UV/PS system, but •OH and ozone molecules are not charged. Nevertheless, •OH could react with OH − to form O •− 2 , which has a weaker oxidation ability than •OH (Nosaka and Nosaka 2017). And the amount of •OH converted to O •− 2 increased with the increase of pH values. This may be the reason why the rate of •OH oxidative of BMAA decreased when the pH values were higher than 9 (Yan et al. 2022). The proportion of BMAA degraded by free radicals increased with the increase of pH values, while the rate of free radical degradation of BMAA decreased when the pH values were above 9. Therefore, the degradation rate of BMAA in ozone aqueous solution showed a first upward and then downward trend.
To sum up, the reasons leading to the degradation rates of BMAA to be first increased and then decreased in ozone aqueous solution may include the following three aspects: (1) the ratio of direct reaction to indirect reaction was different under different pH values; (2) the pH value of solutions can affect the composition of oxidative active species in ozone aqueous solutions; (3) and the pH value of a solution can affect the charged state of the amino and carboxyl groups in the BMAA molecule.

Effect of initial BMAA concentration
The degradation of BMAA with different initial BMAA concentrations (0.6, 1.0, 1.4, and 2.0 μM) was carried out with 1.4 mg/L ozone, pH 7, and 20 °C. It can be viewed in Fig. 3A that ozone can rapidly degrade BMAA, and the concentrations of BMAA were constant after 120 s. Degradation rates of BMAA were decreased with the increase of the concentrations of BMAA. The degradation rate could be as high as 98.18% at 120 s with the initial concentration of 1 mg·L −1 BMAA. However, when the initial concentration of BMAA was 2 mg·L −1 , the degradation rate of BMAA was only 64.36% at 120 s. The probability of effective collisions between BMAA and oxidant enhanced as the initial concentrations of BMAA increased, and the absolute removal of BMAA should increase. However, since the concentration of ozone was constant, the increase of the initial concentrations of BMAA led the molar ratio of ozone to BMAA to reduce, and the relative degradation rate of BMAA decreased.
As seen in Fig. 3B, the concentrations of residual ozone in the solutions and the degradation rates of BMAA at 120 s both declined with the increase of the initial BMAA concentrations. The ozone in solution was depleted when the initial concentration of BMAA was 2 mg·L −1 . The same phenomenon was observed in the experiments using ozone to degrade microcystin. Ozone could rapidly degrade microcystin, and the microcystin could be completely removed when the remaining concentration of ozone was higher than 0.075 mg·L −1 . Besides, Rositano et al. (1998) reported that the remaining concentration of ozone was at least 0.05 mg·L −1 to ensure the complete removal of algal toxins. And when the initial concentrations of BMAA were lower than 1.4 mg·L −1 , namely the molar ratio of ozone to BMAA was greater than 1, the remaining concentrations of ozone were higher than 0.05 mg·L −1 . Therefore, the main reason that the degradation rates of BMAA decreased with the increase of its initial concentrations may be due to the insufficient oxidant concentrations in the solutions.

Effect of the initial concentrations of ozone
The degradation rates of BMAA with different initial concentrations of ozone (1, 1.4, 2, and 2.4 mg·L −1 ) at 1 mg·L −1 BMAA, pH 7, and 20 °C are shown in Fig. 4A. When the initial concentration of ozone was increased from 1 mg·L −1 to 2.4 mg·L −1 , the time required for the complete removal of BMAA was shortened from 120 to 10 s. The molar ratios of ozone to BMAA were all greater than 1 at different initial concentrations of ozone, which could make sure that the remaining concentrations of ozone are higher than 0.05 mg·L −1 , so BMAA can be completely degraded. Since the concentrations of reactive oxygen species per unit volume increased with the increase in ozone concentrations, which resulted in the enhancement of the efficiency of effective collision, the degradation rates of BMAA increased.

Effect of temperature
The experiments to study the effect of temperature were conducted at the condition of 1 mg·L −1 BMAA, 1.4 mg·L −1 ozone, and pH 7, and the temperature was changed by a constant temperature water bath (schematic devices (6) and (8) in Fig. S1). It can be seen in Fig. 4B that with 30 s of reaction, the degradation rates of BMAA with 10 °C, 20 °C, and 30 °C were 99.54%, 96.32%, and 92.24%, respectively. As the temperature increased, degradation rates of BMAA slowed down. It is well known that ozone is unstable in aqueous solutions and easy to decompose. High temperature is beneficial to the decomposition of ozone. And the direct oxidation of ozone molecules was the main reaction at pH 7, so the concentrations of ozone molecules in aqueous solution at low temperature were higher, which was more advantageous for the removal of BMAA.

Degradation mechanism analyses
To investigate the mineralization of BMAA by ozone, the TOC was analyzed at 10, 30, 60, 120, and 180 s. It can be viewed in Fig. 5 that ozone could quickly degrade BMAA with high degradation rates, but the degradation rates of TOC were relatively low. The degradation rates of BMAA and TOC at 10 s were 89.04% and 46.63%, respectively. BMAA was almost completely removed for 180 s, while the rate of TOC was only 48.31%. It can be inferred that ozone cannot completely mineralize BMAA, and by-products, which are more difficult to degrade, were produced during the oxidation of BMAA by ozone. The identification of the by-products of BMAA degradation by ozone was performed by UPLC-MS/MS with positive ion mode. Three by-products of BMAA with [M-H] + values of 105, 90, and 88 were identified (Fig. S5-S7). The degradation pathways of BMAA were proposed using the information of by-products as shown in Fig. 6. It can be seen that the main approach for ozone to oxidize BMAA was the cleavage of the C-N bond. The different positions of the cleavage lead to different products. Because the degradation of BMAA was quick, some intermediates were difficult to detect. Identification of intermediates and analysis of product toxicity will be carried out in our future experiments.

Degradation of BMAA in actual water
The degradation of BMAA in actual water was carried out with water from the Songhua River (E 126° 35′, N 45° 41′) in Harbin. Concentrations of BMAA and ozone were 1 and 1.4, and the pH and temperature were set to 7 and 20 °C. It can be seen in Fig. 7 that ozone could effectively degrade BMAA in actual water, while oxidation efficiency was weaker than in pure water. The degradation rates of BMAA at 120 s in actual water and pure water were 55.46% and 99.94%, respectively. And the degradation rate of BMAA tended to be stable after 10 s in actual water. There is a large amount of natural organic matter in actual water, and natural organic matter could compete with BMAA to consume oxygen molecules in the solution leading to the decrease of the degradation rate of BMAA. Therefore, the effect of water matrix on the degradation of BMAA by ozone will be investigated in depth in subsequent work. The concentration of ozone in the solutions kept decreasing over time, which may be the main reason why the increase of degradation rates of BMAA was small in the subsequent reaction time.

Conclusion
In this work, we investigated the degradation efficiency of BMAA by ozone. The effects of reaction parameters were systematically explored, and the reaction path of BMAA degradation was inferred. The results provided a theoretical basis for the use of ozone process to remove BMAA in actual water. The main conclusions obtained were as follows: (1) Ozone can quickly degrade BMAA, and the degradation rate of BMAA in pure water was close to 100%, but it had declined in actual water. The principal way for ozone aqueous solution to oxidize BMAA was changed as pH values. The direct oxidation by ozone molecules was the main reaction under acidic and near-neutral conditions, and the indirect oxidation by radicals was the main reaction under strong alkaline condition.
(2) The pH values had a significant effect on oxidative degradation of BMAA by ozone. Near-neutral conditions were conducive to the degradation of BMAA. The difference in particle morphology of BMAA, the difference in the composition of oxidative species in the solutions, and the difference in the proportion of direct and indirect oxidative degradation at different pH values were the main reasons for this phenomenon. (3) The concentration of ozone was positively correlated with the degradation rate of BMAA, and the concentration of BMAA and temperature were negatively correlated with the degradation rate of BMAA. (4) The mineralization of ozone-oxidized BMAA was lower than the degradation rate of BMAA. Three byproducts ([M-H] + = 105, 90, and 88) were identified. Ozone decomposed BMAA mainly by attacking the C-N bond in BMAA molecules.

Author contribution
Boyin Yan and Chunyu Han designed the experiment and collected the experimental data. Boyin Yan, Zhiquan Liu, and Guizhi Wu were involved in writing, reviewing, and editing. Jincheng Li and Wenxiang Xia helped with the analysis of data. Songxue Wang and Fuyi Cui were involved in the conceptualization and funding acquisition.