An LLE Based LC- ESI MS/MS Analytical Method Development to Detect Azithromycin Residue in Water to Monitor Contamination Level of River and Fish Farm of Bangladesh

The aim of this study to development and validate an analytical method to determine the Azithromycin (AZN) residue in water approach to monitor the contamination level of river and sh farm water of Bangladesh. Azithromycin (AZN) was analyzed using liquid chromatography mass spectrometry (LC-ESI MS/MS). The chromatographic separations of analyte were performed using ZORBAX RRHD Eclipse Plus C18 (2.1×100 mm, particle size 1.8 µm) column and mobile phase was 0.1% formic acid in water and acetonitrile with 50:50. The analyte were detected in positive electron spray ionization mode with multiple reaction monitoring mode (MRM) with mass transition from 749.5 m/z to 591 m/z and 158 m/z as quantier and qualier ion respectively. Liquid-liquid extraction method was used for the extraction of AZN residues. The developed method was validated in terms of accuracy, precision, linearity and specicity. The analyte showed a good linear in the range of 0.1-100 µg/L. Three spiking levels (0.25, 0.5 and 1.25 µg/L) was performed for determining accuracy and precision. Recoveries and RSD were in the range of 96.6-101.5% and 3.5-6.3 respectively. The estimated limit of detection and limit of quantication was 0.017 and 0.05 µg/L respectively. Using the developed method, we analyzed 5 different rivers and 5 different sh farm samples. We found no azithromycin residue in rivers water, but we found in one sh farm water azithromycin residue 0.35±0.06 µg/L. The results indicate that we should be concern to use antibiotics in sh farm water in different ways.

AZM is an antibiotic which is widely used in the treatment of bacterial infections such as streptococcal pharyngitis, pneumonia, diarrhea and intestinal in ammation (Talaiekhozani et al., 2020). It can also create an enhanced spectrum and potency against bacteria compared with other macrolides and superior stability in acidic media (Zeng et al., 2014). AZM not only can be used against atypical pathogen but also be effective against gram positive and gram-negative bacteria (Lalak and Morris, 1993;Retsema et al., 1990; Dunkin et al., 1988).
In this century, Major concerning issues are pharmaceuticals for their widespread use and frequent exposure in the water cycle. They can mix with water bodies from different ways, such as a human ("Occurrence and ecological risks of pharmaceuticals in a Mediterranean river in Eastern Spain -Science Direct," n.d.). After wreckage, these residue scan be excreted unchanged, or as metabolites, which might be more toxic substances than native compounds (Lalović et al., 2017). As a result, antimicrobial resistance is a great threat for human being which is caused by antibiotics residue in environment (Hendriksen et al., 2019). Due to the unregulated use of antibiotics in human and animals, lack of wastewater treatment, and poor sanitation, it is estimated that more than 40% death associated with resistant pathogens in 2050 will be occurred in low and medium-income countries in Asia (Singh et al., 2019;Coyne et al., 2019;Kakkar et al., 2018).
The use of antibiotics in Bangladesh is increasing day by day (Angeles et al., 2020), they are not only limited to human healthcare but also widely used in animal husbandry (Roess et al., 2013). Poultry and dairy farming are a common practice in the rural area of Bangladesh, for the lack of knowledge sometimes high dose of antibiotics are applied. The use of antibiotics in animal farming not limit to treat bacterial related disease, but also to enhance the growth and to prevent infections (He et al., 2015;Jechalke et al., 2014;Grijalva et al. 2009). These antibiotics excreted and directly leaching to the water, sometimes the byproducts produced from these farming used as sh feed and there become a great source of water contamination ("Global trends in antimicrobial resistance in animals in low-and middleincome countries | Science," n.d.). So, the continuous impute of antibiotics in the aquatic environment may lead to an adverse effect on aquatic organism (Hernando et al., 2006). Up to 95% administrated antibiotics to human and animal may be excreted to environment without metabolized (Hirsch et al., 1999). In Bangladesh waste water treatment plant not designed yet to remove completely the antibiotics from the industrially produced waste water ("Strategic Survey of Therapeutic Drugs in the Rivers Po and Lambro in Northern Italy | Environmental Science & Technology," n.d.)(Analysis, Removal, Effects and Risk of Pharmaceuticals in the Water Cycle, 2007). As Bangladesh is a poor country, its healthcare system also very poor, and over all these the regulatory authorities cannot control the purchase and use There are some reports already existing for AZM analysis in biological samples, including serum, tissues, and plasma. These methods followed bioassay, which are laborious and time consuming (Riedel et al., 1992). In some reports using High-performance liquid chromatography (HPLC) with ultraviolet detector (UVD) also described ( The aim of this study is to develop an LLE method to analyze AZM residue using LC-MS/MS, and to monitor the contamination level of river and sh farm water in Bangladesh. To the best of our knowledge there is no LLE method available to analyze AZM residue in water and also no study report yet has been reported about AZM contamination level of Bangladesh River and sh farm water.

Standard preparation
The stock solution of standard was prepared by weighing 10.01 mg Azithromycin standard, the standard was taken in a 100 mL volumetric ask (Duran group, Wertheim, Germany), and added MeOH up to the mark to make 100 ppm (mg/L) stock solution. The intermediate working solution of 10 ppm was prepared by diluting stock solution with the same MeOH solvent. To construct calibration curve serial dilution was done using same solvent to the concentration 100, 50, 10, 5, 1, 0.5, 0.1 ppb (µg/L). All the solution was kept at -4°C prior to analysis.

Sample collection
For method development Milli Q water was used as control sample,the river water was collected from The Padma, The Meghna, The Shitalakkha, The Gomati and The Buriganga river of Bangladesh, the sh farm water was collected from 5 different sh farms located at Chandpur a district of Bangladesh. All the samples are collected in February 2020. The samples were coolected using 1 L glass bottle. The samples were carried to laboratory in the same day, using the ice box. Samples are stored in laboratory -4°C refrigarator until analysis.
Sample extraction 100 mL water sample was taken in a 500 mL separatory funnel (Duran group, Wertheim, Germany), 10 g NaCl was added to the water. 70 mL DCM was added to it. The separatory funnel was shaken strongly for 5 minutes with hand to mix the sample properly. Then the separatory funnel was kept untouched for 30 minutes to complete the partition of two aqueous and organic layer. The organic layer was collected in 250 mL round bottomed ask using a funnel where cotton was used to control the leakage of anhydrous Magnesium sulphate. Anhydrous Magnesium sulphate was used to dry/absorb aqueous moisture from organic part. The whole process was repeated using 30 mL DCM to complete the extraction. The collected organic part was dried using vacuum evaporator (Buchi) maintained temperature below 40°C.Then the sample was reconstructed using 10 mL MeOH. Around 1.5 mL was taken through lter using 40 µm syringe lter and taken in 2 mL vial for LC-MS/MS analysis.

Method validation
The Speci city linearity, limit of detection and quanti cation (LOD and LOQ), precision, and accuracy was taken as parameter of analytical method validation. The speci city of the method was established by comparing the control sample, spiked sample, and real sample chromatogram. The linearity of the method was determined by constructing a calibration curve of concentration against peak area of different concentration. The LOD was explained as the minimum amount that can be determined using the instrument with signal-to-noise ratio (S/N=3) and LOQ explained as the minimum amount that can be quanti ed using the instrument with signal-to-noise ratio (S/N=10). Relative standard deviation (RSD) was assessed to determine the precision of the analytical method. Accuracy of the method was determined using the recovery study, at two different spiking level recovery study was done. During the recovery study optimized sample extraction method was followed.

Method optimization
To obtain good extraction n-Hexane and dichloromethane (DCM) both the solvent was tested. DCM found better extraction performance over n-hexane, so DCM was taken as the extraction solvent. Using DCM as solvent salting out extraction was also performed. To optimize the method sodium chloride (NaCl) salt was used and we perform extraction using 5, 10, and 15% NaCl salt in the sample. The result of these experiment showed 10% NaCl and DCM solvent is the best optimized method (Table. 1). In extraction DCM solvent amount also optimized. We tried extraction using three different solvent amounts, the amount of DCM solvent was experiment (50 +30) = 80 mL, experiment (70+30) =100mL and experiment (100+30) =130 mL. Among the three experiment, extraction with (50+30=80mL) solvent, cannot extract the analyte properly, but extraction with (70+30=100mL) and (100+30=130mL) solvent showed similar performance of extraction. Considering the solvent amount extraction with (70+30=100 mL) was taken as the optimized method (Table 1).

Speci city
The speci city of the method was determined by comparing the control sample chromatogram, real sample chromatogram with standard chromatogram. No unwanted peak was observed at the retention time of analyte peak, which determine the speci city of the method. Linearity Linearity was checked by constructing a linear curve range of concen6tration level (0.1-200 ppb). The linear graph was obtained with the regression equation Y=45423.1299x+64.2074 and coe cient R 2 =0.9994 (Table 2) Limit of detection and quanti cation (LOD and LOQ) The estimated LOD and LOQ was 0.017 and 0.05 µg/kg respectively, this amount is very low considering the others method. Xiang et.al. reported that the LOD and LOQ of AZM in feather sample were 0.5 and 2.0 µg/kg respectively (Xiang et.al. 2020).

Accuracy and precision
To obtain accuracy and precision recoveries studies was done in triplicate by spiking standard in three concentration level. The spiking level was 0.5, 0.25, and 1.25 µg/kg and the recovery were in the range (96.6-101.5) % and RSD (3.5-6.3) ( Table 2). These recoveries and RSD were in the acceptable range (70-120%) which are recommended by SANTE/11945/2015 (2015) Residue amount in real sample AZM residue determined in the river and farm water sample. As the representative of the total Bangladesh river ve important and big river was selected. To represent the farm water condition ve different farm water was collected. Following the developed and validated method. We found no residue in river water but we found in one farm water residue as 0.35 (µg/L) ( Table 3

Conclusion
An LLE based analytical method development to anlyze azithromycine antibiotics residue in river and sh farm water of Bangladesh. The new developed and validated method is used to analysis the AZM residue in water to nd out the contamination level of river and sh farm water. Though the river water water of Bangladesh have the chance to be contaminated with various antibiotics but we did not nd any AZM residue in any river water of Bangladesh. In sh farm water we found 0.35 µg/L residue only in one sh farm water among the ve different sh farm water. The residue may be come from sh feed or from the nearby dairy farm as that the farm located closely. We can suggest from the ndings of our experiment, the river water may still safe from pharmaceuticals contaminants and need to monitor the sh feed used in sh farming need to monitor the contamination level, the dairy farm wastage also need to control to minimized the contamination nearby sh farm water.  Figure 1 Chemical structure of Azithromycin