A Newly Developed Approach for Analyzing the Degradation of Glyphosate and Aminomethylphosphonic Acid in Different Salinity Levels

Abstract

The report for the global presence of herbicide glyphosate and its metabolite aminomethylphosphonic acid (AMPA) is presently limited in the marine environment, presumably due to a lack of analytical methods capable of detecting these compounds at low concentrations in high-salinity matrices. In the present study, we aimed to develop a time-saving and reliable method for the analysis of glyphosate and AMPA in different salinity levels of seawater. This novel method integrates a derivatization process with a solid-phase extraction cleanup step to mitigate salt-matrix effects during high-performance liquid chromatography coupled with tandem mass spectrometry analysis. The present method was validated in environmental freshwater and seawater with the limit of quantitation of 2 and 0.5 ng/mL and coefficient of variation percentage of 0.63–3.15% and 0.59–3.07% for glyphosate and AMPA, respectively. The degradations of three concentrations of spiked glyphosate (10, 100, 1000 mg/L) were assessed under two treatment conditions: with and without sterilization and at three salinity levels (0, 17.5, 35‰) over a period of 112 days. The results show that glyphosate degradation is significantly higher in non-sterilized water compared to sterilized conditions, indicating that microbial activity is the primary driver of degradation. Furthermore, brackish water appears to provide a more favorable environment for the microbial biodegradation of glyphosate.

1. Introduction

The commercial introduction of glyphosate (N-(phosphonomethyl)glycine) in 1974 by Monsanto Company, USA, established it as the most widely used herbicide for weed control in agricultural areas worldwide [1]. Glyphosate acts by blocking the Shikimate pathway, which is present only in plants, bacteria, and fungi. It inhibits the 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) enzyme, thereby disrupting the synthesis of aromatic amino acids such as tryptophan, tyrosine, and phenylalanine, which are vital for protein synthesis and plant growth [1]. The persistence of glyphosate in the environment has become a topic of significant concern due to its widespread and extensive use, as well as its potential to harm ecosystems, diffuse into food chains [2], and produce resistant weeds [3]. In the past, the absence of the EPSPS enzyme in humans and animals was often cited as the reason why glyphosate is considered to have relatively low toxicity to mammals [1]. However, concerns persist due to glyphosate’s potential indirect effects on gut microbiota, which may affect the shikimate pathway and bioaccumulation in ecosystems and have broader environmental impacts [4]. Moreover, increasing evidence from various studies has revealed that glyphosate exposure can cause multiple toxic effects, including neurotoxicity, reproductive toxicity, and liver and kidney toxicity [5]. The oxidative stress induced by increased reactive oxygen species production and reduced antioxidant defenses has been proven to be a common mechanism behind the observed toxic effects of glyphosate [5]. These concerns highlight the need for ongoing studies into its safety and ecological consequences.

Glyphosate degradation occurs through a combination of abiotic and biotic processes, including adsorption, photolysis, and microbial degradation. Microbial degradation is the most significant mechanism for glyphosate breakdown in the environment [6]. In the main pathway, glyphosate is degraded by the enzyme glyphosate oxidoreductase, breaking the carbon–nitrogen (C–N) bond and forming two key metabolites of AMPA (aminomethylphosphonic acid) and glyoxylate, which are consumed and utilized as an energy source by most glyphosate-degrading microorganisms, effectively reducing its environmental presence [7]. Since AMPA is not naturally metabolized within plant cells or certain ecological conditions, its persistence raises concerns about toxicity and environmental contamination [8]. The extensive and widespread use of glyphosate-based products has resulted in their pervasive distribution in the environment over time. Various processes, including deposition, volatilization, spray drift, soil erosion, leaching, runoff, drainage, and percolation, contribute to their dispersal. These phenomena lead to a continuous influx of glyphosate and its byproducts into water resources, which often serve as the final receptors [9]. Upon entering aquatic ecosystems, there is increasing concern about its toxic effects on aquatic organisms [10,11]. The persistence and potential toxicity of glyphosate in water bodies make it a pressing issue for environmental monitoring and regulation [12]. As organic pollutants migrate from rivers to the sea, the gradual increase in water salinity adds complexity to the distribution of pollutants [13]. This process is further influenced by the variability of chemical–physical parameters such as salinity, temperature, pH, dissolved organic carbon, and clay mineral content. These parameters significantly affect the transport processes of contaminants, including their migration, sorption, and partitioning [13]. Moreover, they influence the composition and activity of microbial populations within the aquatic ecosystem, which play a critical role in the degradation and transformation of pollutants. This interplay between factors underscores the dynamic and multifaceted nature of contaminant behavior in aquatic environments [9].

Taiwan, a small island with limited water resources, faces significant challenges due to its high-density agriculture and aquaculture operations. The overuse of pesticides and herbicides in agricultural fields has led to contaminants entering the soil and groundwater, eventually making their way into river systems via runoff. This contaminated water is then often utilized in aquaculture, as most aquaculture ponds are situated downstream or near agricultural areas. Among these contaminants, glyphosate is of particular concern due to its widespread use on limited cultivated land, which amplifies the risk of environmental water pollution. The increased application of glyphosate poses mounting challenges for maintaining water quality in aquaculture ponds and increases the risk of chemical residues accumulating in aquaculture organisms. Given that aquaculture represents a vital economic sector in Taiwan, it is especially vulnerable to the adverse effects of water pollution originating from agriculture. This situation underscores the urgent need for integrated water management policies and sustainable agricultural practices to protect the aquaculture industry and the broader environment.

Most methods for analyzing glyphosate and AMPA in various aqueous solutions rely on chromatography techniques coupled with different detection technologies, such as high-performance liquid chromatography coupled with tandem mass spectrometry (HPLC-MS/MS), or with fluorescence or UV detection, ion chromatography coupled with inductively coupled plasma mass, and gas chromatography coupled with mass spectrometry [14]. The technique most often employed for the analysis of glyphosate is the HPLC method because of the ionic characteristic of glyphosate [15]. This liquid chromatography method often requires derivatization to improve detection sensitivity and selectivity, particularly due to the polar and ionic nature of glyphosate and AMPA, which complicates direct analysis [16]. One previous study reported the detection of glyphosate and its main degradation product, AMPA, in German estuaries discharging into the Baltic Sea [17]. Their findings confirmed that both compounds are transported from freshwater environments into the marine ecosystem. However, their analytical method lacked the sensitivity to detect glyphosate and AMPA at low concentrations beyond the estuarine areas [17]. Therefore, a good method capable of detecting glyphosate and AMPA in seawater must include a pre-concentration step and an analytical procedure that is insensitive to the salt matrix [18,19]. In this study, we introduce a solid-phase extraction (SPE) cleanup and pre-concentration step after analyte derivatization with 9-fluorenyl methoxycarbonyl chloride (FMOC-Cl) following sensitive HPLC-MS/MS and investigate the effect of microbial and salinity on glyphosate degradation in Taiwan native seawater.

2. Materials and Methods

2.1. Materials and Analytical Chemicals

The standard chemicals used in the presently analyzed study included Glyphosate (C₃H₈NO₅P, MW: 169.07 g/mol, Toronto Research Chemicals, Toronto, ON, Canada), AMPA (CH₆NO₃P, MW: 111.04 g/mol, Toronto Research Chemicals, Toronto, ON, Canada), and the isotope-labeled reference standard Glyphosate-¹³C₂¹⁵N (¹³C₂CH₈¹⁵NO₅P, MW: 172.05 g/mol, Toronto Research Chemicals, Canada). Glyphosate, AMPA, and isotope-labeled glyphosate were, respectively, dissolved in 100% methanol (MeOH, DAEJUNG, Siheung, Republic of Korea), resulting in a stock solution with a final concentration of 1 mg/mL. The working solution for analysis (100 μg/mL) was prepared from each stock solution by 10 times dilution with 100% MeOH into 2 mL amber plastic vials. The borate buffer used in the derivatization process was prepared by dissolving 1.9 g of sodium tetraborate (Merck, Darmstadt, Germany) in 40 mL of deionized water (18.2MΩ. cm) at 40 °C. The pH was then adjusted to 10 using sodium hydroxide (J.T. Baker, Phillipsburg, NJ, USA), and the final volume was brought to 50 mL with deionized water. The FMOC-Cl solution (4 mg/mL) was prepared freshly by dissolving the deriverization reagent FMOC-Cl (Alfa Aesar, Haverhill, MA, USA) in 100% acetonitrile (J.T. Backer, Phillipsburg, NJ, USA). For the degradation test, the glyphosate solution (Glyphosate isopropylamine, 41% SL) was purchased from SINON Corporation, Taichung, Taiwan.

2.2. Analytical Methods

The derivatization process was conducted by combining 100 μL of a working standard solution or analyzed water sample and 225 μL of borate buffer in a microcentrifuge tube, followed by adding 225 μL of the freshly prepared FMOC-Cl solution and 390 μL of deionized water. The mixture was incubated in a water bath at 50 °C for 30 min. Subsequently, 60 μL of formic acid (98% wt., Sigma, St. Louis, MO, USA) was added to the derivative mixture to stop the reaction. The derivative solution was transferred into a new plastic tube, and 3 mL of pentane (Scharlau, Barcelona, Spain) was added. The mixture was vortexed for 15 s, and the pentane layer was removed. This process was repeated three times before drying with a nitrogen evaporator. The dry pellet was then resuspended in 1 mL of 5% acetonitrile and loaded onto an SPE cartridge (ProElut, PLS, 1 mL, 60 mg, Dikma, Lake Forest, CA, USA), pre-activated with 1 mL of 100% MeOH followed by 1 mL of deionized water. The cartridge was washed with 3 mL of deionized water and dried via a vacuum pump. Finally, the target analytes were eluted from the cartridge with 1 mL of 100% MeOH. The eluent was then evaporated by nitrogen under 45 °C and reconstituted with 1 mL of 15% MeOH. The extracted analytes were filtered through 0.22 μm PTFE syringe filters (Dikma, Lake Forest, CA, USA) into 1.5 mL amber plastic vials prior to analysis.

The analysis was accomplished using the HPLC-MS/MS system. The injection volume was 5 μL, and the total run time for HPLC was 11 min. The HPLC system was contained with an Agilent 1100 series degasser (G1379A, Agilent Technology, Santa Clara, CA, USA), a binary pump (G1312A, Agilent Technology, USA), and a CTC PAL autosampler (CTC Analytics AG, Switzerland). The binary pump delivered mobile phase A (aqueous phase) with 5 mM ammonium acetate (J.T. BACKER, USA) (pH = 9.0) and mobile phase B (organic phase) with 100% MeOH to an Agilent Poroshell 120 EC-C18, 2.7 μm particle size, and 4.6 × 50 mm column (Agilent Technology, USA). The elute gradient with a flow rate of 300 μL/min was as follows: 15% mobile phase B for 2 min, ramped from 15% to 100% of mobile phase B over 1 min, held at 100% mobile phase B for 4 min, and then rapidly returned to 15% mobile phase B and maintained at 15% for an additional 4 min (Table S1). After each sample injection, the autosampler performed washing steps for the needle and the injection valve using 100% acetonitrile and 50% MeOH, respectively.

The API4000 LC-MS/MS triple quadrupole mass spectrometry system (SCIEX, Framingham, MA, USA) was used as a detector. The API4000 system was equipped with a Turbo V^® ionization source with an electrospray ionization probe and operated in the negative multiple reaction monitoring (MRM) mode. The MRM transition pairs of FMOC-Cl-derived target compounds and compound-dependent parameters such as de-clustering potential (DP), entrance potential (EP), collision energy (CE), and collisional exit potential (CXP) are listed in

Table 1. MRM transition pairs for derivation analytes and optimized voltage parameters.

Analyte	Q1 Mass (Amu) Precursor Ion	Q3 Mass (Amu) Product Ion	DP (V)	EP (V)	CE (V)	CXP (V)
FMOC-Cl- Glyphoste	390.0	149.9 (Qual.)	−70	−10	−35	−10
		167.8 (Quan.)	−70	−10	−20	−13
FMOC-Cl-AMPA	332.1	135.8 (Qual.)	−29	−10	−22	−8
		109.9 (Quan.)	−29	−10	−14	−8
FMOC-Cl- Glyphoste 13C215N	393.2	170.8	−30	−10	−16	−14

Note(s): Abbreviations: FMOC-Cl: 9-fluorenyl methoxycarbonyl chloride; AMPA: aminomethylphosphonic acid; DP: de-clustering potential; EP: entrance potential; CE: collision energy; CXP: collisional exit potential.

. For example, for the detection of FMOC-Cl-Glyphosate, the MRM transition pairs were m/z 390.0 → m/z 149.9 for qualification (Qual.) and m/z 390.0 → m/z 167.8 for quantitation (Quan.). Moreover, DP and EP were set at −70 V and −10 V, respectively, while CE and CXP were set at −35 V and −10 V for qualification (Qual.) and −20 V and −13 V for quantitation (Quan.). The parameters of the ion source were −4500 V for ion source voltage. The nitrogen gas from the liquid nitrogen tank was applied for mass spectrometry, and the parameters were 20 psi for curtain gas, 45 psi for nebulization gas, 50 psi for heating gas, and level 4 (collision cell pressure 3.2 × 10⁻⁵ torr) for collision gas. The resolution of both Q1 and Q3 was set as unit/unit resolution. The operation voltages were optimized and determined before the chromatography condition determination for each target compound and-isotope labeled reference standard was performed. The data acquisition and process were accomplished by Analyst 1.4.2 software (SCIEX, Framingham, MA, USA).

2.3. Experimental Procedure

The freshwater (salinity 0‰) was obtained from the Keelung River (Keelung, Taiwan), and the seawater (salinity 35‰) was obtained from the Aquatic Animal Centre, NTOU (National Taiwan Ocean University, Keelung, Taiwan). The freshwater and seawater were confirmed as blank matrices with zero concentrations of glyphosate and AMPA. The brackish water (salinity 17.5‰) was prepared by mixing equal parts of seawater and freshwater. These three waters were divided into two treatments: sterile water and non-sterile water. The sterile water group was treated in an autoclave at 121 °C, 15 psi, for 30 min. Glyphosate solutions were spiked into 50 mL of the aforementioned six treatment waters with three final concentrations of 10, 100, and 1000 mg/L. Samples were collected after 0, 1, 2, 4, 7, 14, 28, 56, and 112 days for the detection of glyphosate and AMPA. All experiments were performed independently in triplicate at room temperature in a dark environment. The level of glyphosate degradation was calculated as the glyphosate removal rate (GRR) according to a previously published report [20]. GRR = (Ci − Cf)/Ci) × 100%, where Ci and Cf are the concentrations of glyphosate at the initial time (Day 0) and final sampling time, respectively. The statistics analysis (factorial experiment) was performed with SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA) to reveal the effect of salinity and sterility on glyphosate degradation.

3. Results and Discussion

3.1. Method Validation

The analysis of glyphosate and AMPA in seawater appears to be primarily hindered by methodological challenges. Both glyphosate and AMPA contain phosphoric acid groups, which result in poor retention in reverse-phase liquid chromatography systems and low separation efficiency in normal-phase liquid chromatography systems. Our laboratory experiences demonstrated that a derivatization procedure must be included in the analysis of a seawater matrix to improve the detection and quantification of glyphosate and AMPA. Martins-Júnior et al. reported the direct analysis of glyphosate and AMPA by liquid chromatography–tandem mass spectrometry [21], while Mol et al. described the analysis of glyphosate and AMPA using flow injection–tandem mass spectrometry [22]. However, our laboratory was unable to replicate the results obtained with these two methods. While replicating Martin’s method, we encountered a relatively weak signal along with shifted retention times, resulting in a poor limit of detection (LOD) and a lower limit of quantitation (LOQ). On the other hand, while replicating Mol’s method, very poor and split peak shapes were produced, accompanied by a serious matrix effect due to the lack of any chromatographic separation. A better peak shape could be observed using the Agilent Poroshell column for chromatographic separation after derivatization with FMOC-Cl. However, noticeable peak tailing was still present for FMOC-Cl AMPA. This tailing could be attributed to the presence of the free phosphate group from glyphosate in the derivatized analyte [23]. The phosphate groups may interact with free silanol groups and bind to active metal sites within the separation column. Glyphosate has been reported to form metal complexes with divalent matrix cations, which can cause significant retention time shifts during analysis or even completely hinder glyphosate detection [19]. Subsequently, an SPE cleanup step was incorporated into our method to address the issues of peak tailing and unstable signals caused by excess derivatizing agents.

The present method validation was carried out in accordance with the Analytical Procedures and Methods Validation for Drugs and Biologics Guidance for Industry issued by the US FDA (Food and Drug Administration, USA). The assessment of method validation included linearity, specificity, LOD, LOQ, accuracy, and precision. The calibration curve of the FMOC-Cl derivation of glyphosate and AMPA was set individually. Based on our pre-test of the signal intensity for the two target analytes, the range of analytic concentrations were 2, 10, 50, 200, 1000, 3000 ng/mL, and 0.5, 2, 10, 50, 200, 1000 ng/mL for FMOC-Cl glyphosate and FMOC-Cl AMPA, respectively. The isotope-labeled reference standard (IS) concentration was 500 ng/mL. The calibration curve was performed in triplicate to check the repeatability. The response peak area of the analyte divided by the response peak area of IS was plotted on the Y-axis, while the analyte concentration divided by the concentration of IS was plotted on the X-axis to generate the equation of the calibration curve for the detection of FMOC-Cl glyphosate (y = 0.000794 x + 0.9998, r² = 0.9998) and FMOC-Cl AMPA (y = 0.00594 x − 0.0018, r² = 0.9975), respectively, (Figure 1). The results represented both good repeatability and linearity. As shown in Figure 2, the specificity of glyphosate and AMPA was characterized by the identified peaks observed at 6.85 and 7.33 min, respectively.

The LOD is determined to have the lowest detection level of the analytical method with a signal-to-noise ratio (S/N) ≥ 3. The LOQ is two times the concentration of LOD established in the analytical method, and the lowest limit of quantitation (LLOQ) is two times the concentration of LOQ. In the analysis of FMOC-Cl glyphosate, LOD, LOQ, and LLOQ were determined to be 0.5, 1, and 2 ng/mL, respectively, with the S/N ratios at 13.7, 26.1, and 73.1, respectively. In the analysis of FMOC-Cl AMPA, the LOD, LOQ, and LLOQ were determined to be 0.05, 0.1, and 0.5 ng/mL, respectively, with S/N ratios of 15.2, 21.9, and 45.8, respectively (

Table 2. Linearity study, LOD, LOQ, and LLOQ (n = 3).

	Concentration (ng/mL)	S/N	CV (%)	Remark
FMOC-Cl-Glyphosate	0.5	13.7		LOD
	1	26.1		LOQ
	2	73.1	10.2	LLOQ
	10		2.3
	50		1.6
	200		1.8
	1000		0.9
	3000		1.3
FMOC-Cl-AMPA	0.05	15.2		LOD
	0.1	21.9		LOQ
	0.5	45.8	4.4	LLOQ
	2		1.7
	10		2.1
	50		4.6
	200		5.3
	1000		3.7

Note(s): Abbreviations: S/N: signal-to-noise ratio; CV: coefficient of variation; LOD: limit of detection; LOQ: limit of quantitation; LLOQ: lowest limit of quantitation.

). The present analytical method demonstrated a relatively lower LOQ compared to the LOQ of 50 ng/mL established by the European Food Safety Authority in 2018 [24].

The quality control (QC) samples were prepared by adding glyphosate and AMPA in freshwater, which had been confirmed to be free of both glyphosate and AMPA. The intra-day and inter-day validation consisted of three replicates at each of the three different QC concentration levels (high (H), medium (M), and low (L)). The concentrations of the QC sample were 2000 ng/mL and 500 ng/mL (QC-H), 500 ng/mL and 100 ng/mL (QC-M), and 5 ng/mL and 1 ng/mL (QC-L) for FMOC-Cl glyphosate and FMOC-Cl AMPA, respectively. As shown in

Table 3. Precision and accuracy of glyphosate and AMPA. (A). Intra-day validation (B). Inter-day validation.

(A) Intra-Day	n = 3	Conc. (ng/mL)	Recovery (%)	Accuracy (%)	CV (%)
			99.0
	QC-L	1	99.6	100.41	1.94
			102.6
			108.5
AMPA	QC-M	100	102.5	105.92	2.91
			106.8
			101.3
	QC-H	500	102.3	102.10	0.72
			102.8
			97.8
	QC-L	5	99.7	98.83	0.97
			99.0
			101.2
Glyphosate	QC-M	500	103.0	102.98	1.75
			104.8
			103.9
	QC-H	2000	103.4	105.54	3.15
			109.4
(B) Inter-day	n = 3	Conc. (ng/mL)	Recovery (%)	Accuracy (%)	CV (%)
	QC-L day1		99.6
	QC-L day2	1	99.9	100.51	1.34
	QC-L day3		102.1
	QC-M day1		108.6
AMPA	QC-M day2	100	102.4	106.07	3.07
	QC-M day3		107.2
	QC-H day1		101.6
	QC-H day2	500	102.8	102.13	0.59
	QC-H day3		102.0
	QC-L day1		99.3
	QC-L day2	5	98.8	99.39	0.63
	QC-L day3		100.0
	QC-M day1		101.0
Glyphosate	QC-M day2	500	103.8	103.04	1.69
	QC-M day3		104.2
	QC-H day1		104.7
	QC-H day2	2000	104.3	106.03	2.51
	QC-H day3		109.1

, the ranges of recovery were 97.8–109.4% and 99.0–108.6% for FMOC-Cl glyphosate and FMOC-Cl AMPA, respectively. The accuracy (average recovery) was 98.83–106.03% for FMOC-Cl glyphosate and 100.41–106.07% for FMOC-Cl AMPA. The precision was presented as the coefficient of variation percentage (CV% = Standard Deviation/Mean × 100). The CV% were 0.63–3.15% and 0.59–3.07% for FMOC-Cl glyphosate and FMOC-Cl AMPA, respectively. The present method’s validation met the criteria proposed by the US FDA validation guidelines, which required quality control accuracy between 85% and 115% and precision with CV ≤ 15%.

3.2. Effect of Sterilization and Salinity on Glyphosate Degradation

The present glyphosate degradation experiments were assessed based on two major factors (water sterilization and salinity) using three testing concentrations over a period of 112 days. As shown in Figure 3, under varying test concentrations and water salinity conditions, a similar trend was observed, indicating that sterilization conditions impeded the degradation of glyphosate. For example, in 10 μg/mL of glyphosate concentration, the GRR of the non-sterile freshwater group reached 50% (53.33% ± 1.53) on day 28 and was 76.50% ± 3.82 on day 112, while the GRR of the sterile freshwater group was 19.10% ± 2.59 on day 112. The GRRs were 0.24–78.90% (n = 216) and 0.10–21.40% (n = 216) for non-sterile and sterile conditions, respectively. There was a significant difference between the non-sterile and sterile groups (Type III SS, F = 384.08, p ˂ 0.0001). Moreover, it was found that the salinity factor affects the degradation of glyphosate. For example, for analysis of 1000 μg/mL of glyphosate on day 112, the GRR in non-sterile freshwater was 30.36% ± 0.39, whereas in non-sterile brackish water and seawater, the GRRs increased to 53.77% ± 0.39 and 49.34% ± 0.69, respectively. Conversely, the GRR in sterile freshwater was 5.60% ± 0.41, whereas in sterile brackish water and seawater, the GRRs decreased to 1.83% ± 0.03 and 0.73% ± 0.01, respectively. The GRRs were 0.12–78.90% (n = 144), 0.09–76.40% (n = 144), and 0.10–62.08% (n = 144) in freshwater, brackish water, and seawater, respectively. There was a significant difference among different salinity groups (Type III SS, F = 15.46, p ˂ 0.0001). An interaction was observed in sterilization cross-salinity factors (Type III SS, F = 9.39, p ˂ 0.0001).

The present study supports previous research [25] demonstrating that glyphosate is persistent in marine water and its degradation is primarily mediated by the native microbial community. The earlier study reported a glyphosate half-life of 267 days under dark conditions at 25 °C [25]. Glyphosate is necessary to include in marine monitoring programs due to several forms of evidence suggesting that its toxicity toward aquatic animals may increase when combined with elevated salinity levels. For example, the combined exposure to glyphosate and increased salinity significantly reduced longevity, as well as the number of offspring and clutches in Daphnia spinulata [26]. Glyphosate and elevated salinity exhibited a synergistic effect, leading to notably poor growth, reduced activity, and feeding, as well as sluggish escape swimming behavior in Anaxyrus terrestris [27].

The impact of sterilization and salinity on the degradation dynamics of glyphosate was further demonstrated by measuring the formation of its major metabolite, AMPA (Figure 4). The increase in GRR was accompanied by an increase in AMPA production. An overall analysis of different glyphosate concentrations revealed that under non-sterile conditions, average glyphosate degradation was higher in seawater environments compared to freshwater environments. Conversely, under sterile conditions, average glyphosate degradation was higher in freshwater environments compared to seawater environments. Another detailed analysis of different salinity levels revealed that under non-sterile conditions, average glyphosate degradation was highest in brackish water environments, while under sterile conditions, average glyphosate degradation was highest in freshwater environments (Figure 5).

The first report on the measurement of native seawater glyphosate and AMPA concentrations in the Warnow Estuary (16.3‰) revealed a linear correlation with salinity [19]. The observed overall decrease in concentrations was attributed to the mixing of water from the Warnow River with saline water from the Baltic Sea, leading to the dilution of glyphosate and AMPA. The results of the present study indicated that glyphosate degradation is more pronounced in brackish water environments, suggesting that brackish water provides a more favorable condition for the microbial degradation of glyphosate. A previous report demonstrated that mild salinization (10‰) stimulated glyphosate degradation and enhanced microbial activities in riparian soil [28]. Moreover, several native halophilic or semi-halophilic bacteria have been isolated and used for the biodegradation of glyphosate herbicide [29]. One previous study revealed that temporal increases in total cell counts, bacterial diversity, and the abundance of distinct bacterial operational taxonomic units could be observed in brackish water containing 82.45 µM of glyphosate. The degradation efficiency of glyphosate could reach 99% after 140 days of incubation with native bacteria in the southern Baltic Sea [30]. Efficient glyphosate degradation requires the careful consideration of microbial communities and environmental factors [31]. Future research could focus on the application of suitable microorganisms to enhance glyphosate degradation in environments with varying salinity levels.

4. Conclusions

The present study provides a time-saving and reliable method for the sensitive determination of glyphosate in pesticides under a high-salinity environment. Moreover, it was found that brackish water provides a more favorable environment for the microbial biodegradation of glyphosate. Our work can assist authorized government departments in taking action and formulating policies to reduce or prevent glyphosate residues in agriculture and aquaculture.