Optimized HPLC-UV Method for Separation , Detection and Quantification of Endocrine Disrupting Estrogens in Low Quality Water

Endocrine disrupting estrogens are emerging contaminants in aquatic ecosystems and environment in general. There are no guidelines for routine monitoring of these chemicals, despite the existing evidences of their adverse health effect to living organisms at low concentrations. This study aimed at developing and validating an optimized HPLC-UV method for detection and quantification of estradiol and ethinylestradiol. Isocratic elution was used for separation and detection of ethinylestradiol and estradiol. The mobile phase was applied with A; water B; acetonitrile (50:50) at flow rate of 0.7mL/min and injection volume 10mL. The precision and accuracy of the method were within the acceptable range. Relative standard deviation of peak area for E2 ranged from 1.373 to 3.668%, and for EE2 ranged from 0.829 to 6.495 %. The percentage recovery for E2 ranged from 82.3 to 99.84 %, and for EE2 ranged from 84.6 to 103.52 %. Linearity of the method was realized at range of 2.5 to 50 ng/mL and 100 to 1000 ng/mL for both E2 and EE2. The linear regression coefficients were 0.9979 and 0.9973 for E2 whereas for EE2 were 0.9983 and 0.9976. Limit of detection were found to be 0.05 ng/mL and 0.08 ng/mL for E2 and EE2 respectively. The obtained limits of quantification were 0.18 and 0.28 ng/mL for E2 and EE2 respectively. In untreated sewage the concentrations of E2 and EE2 were 0.28 ng/ml and 0.18 ng/ml respectively. But in subsequent wastewater stabilization ponds the concentrations were below detection limit. Therefore, the optimized HPLC-UV method is suitable for detection and quantification of endocrine disrupting estrogens when a level of pollution is at least 0.15 ng/ml. At low extent of pollution would require use of the method in conjunction with ELISA technique.


Introduction
Estrogen hormones are emerging pollutants in water. There are natural and synthetic estrogens; the natural estrogens are not only produced by humans but also by other vertebrates as well as invertebrates particularly insects (Das, 2016). The synthetic estrogens are used for birth control, hormone replacement therapies, cancer drugs and other pharmaceuticals (Shook, 2011;Sood et al., 2014). Usually the estrogens are excreted from humans and animals mainly as conjugates of glucuronide or sulphate and a lesser amount as parent compounds in urine and faeces. The conjugates undergo deconjugation to form free estrogens by enzymatic action of bacteria or fungi (Kumar et al., 2012). The occurrence of these hormones in the environment is receiving considerable attention due to the fact that most of these compounds are endocrine disrupting chemicals (Christiansen et al., 2002;Cui et al., 2006;Zheng et al., 2013;Huang et al., 2016). Several studies have demonstrated that wastewater treatment systems are the main source of estrogens to aquatic environment (Ingrand et al., 2003;Swart & Pool, 2007;Yoon et al., 2012;Belhaj et al., 2014). A reliable analytical technique is necessary for separation, detection and quantification of the estrogens in the environmental matrix such as water.
sensitive for detection of estrogens (Faqehi et al., 2016), however their use is limited by high costs. Immunochemical methods are also sensitive but have poor selectivity compared to LCMS/MS and a single assay can only detect a single steroid (Ingerslev & Halling-Sørensen 2003;Faqehi et al., 2016).
High performance liquid chromatography is a dominant analytical technique due to its applicability to diverse analytes (Dong, 2013). The technique can quantify compounds in complex samples and can be coupled with different detectors such as UV, diode array detection (DAD), evaporation light scattering detection (ELSD), refractive index detector, fluorescence detection and mass spectrometry (MS) (Dong, 2013). The successful use of HPLC requires the right combination of variety of operating conditions such as the column length and diameter, column temperature, type of the column packing, type of the mobile phase, mobile phase flow rate, and injection volume (Settle, 2004). This implies that HPLC method development and validation are inevitable for successful application of HPLC.
Method validation ensures consistent capabilities of an analytical method by defining and confirming the analytical requirement for a specific application (Magnusson & Ornemark, 2014). The ISO/IEC 17025:2005, a guideline for competence testing and laboratory calibration, stresses that method validation is necessary requirement in analytical chemistry since it demonstrate that the method is fit for the purpose. Method validation include assessment of the parameters such as precision, linearity, accuracy, limit of detection, limit of quantification, specificity, range and robustness of the method (ICH, 2005;Magnusson & Ornemark, 2014). Schellinger & Carr (2006) found that gradient elution gave a shorter overall analysis with similar resolution compared to isocratic elution. On the other hand isocratic elution remain preferable when the sample contains less than ten weakly retained components (Schellinger & Carr, 2006).
This study aimed at establishing an optimized HPLC-UV method for separation, detection and quantification of estrogens hormones from low quality water particularly from wastewater stabilization ponds. The study focused on the two most potent estrogens namely ethinylestradiol and estradiol.

Preparation of Standard Solution and Calibration Standard Solution
Stock solution of 10,000 ng/ml in methanol was prepared for a mixture standard E2 and EE2, and then stored at 4 o C in a fridge. Serial dilution of the stock solution was carried out to obtain 1000, 750, 500, 250, 100, 50, 25, 10, 5, 2.5, 1, 0.5, and 0.1 ng/ml for the mixture of the two hormones. In addition, separate standard solutions of the two hormones were prepared at 100 ng/mL for determination of their retention time.

Spiking Standard Estrogens in Distilled Water
For determination of recoveries the estrogens standard solutions were spiked in 1 L of distilled water to make different concentrations for both E2 and EE2. The established concentrations were 1000, 750, 500, 250, 100, 50, 25, 10, 5, 2.5, 1, 0.5, and 0.1 ng/ml. Each solution was filtered twice using by GFCs, a procedure which could be applied for removal of debris from low quality water.

HPLC Conditions
Analyses were performed at Ecotoxocology and Natural Product Laboratory at the College of Veterinary and Medical Sciences, Sokoine University of Agriculture. The HPLC (Shimadzu 20AD) fitted with an auto sampler and a SPA-UV detector at 230 nm was used for analysis. A reversed phase phenomenex C-18 column (150 x 4.6 mm and particle size of 5μm) was used for separation with the oven temperature set at 35 o C. The sample injection volume was 10 μL and flow rate 0.7 mL/min. The mobile phases employed were A; water B; acetonitrile (50:50).

Method Development
With the theoretical background, detection, separation and quantification of the two estrogens hormones were achieved by manipulating factors such as wavelength, mobile phase composition, flow rate and temperature. Several trials were carried out to separate, identify and quantify a mixture of estradiol, and ethinylestradiol.
When the mobile phases were; A-water and B-methanol, the hormones were detected at shorter retention time but the resolution was very poor even after several trials upon changing column temperature, flow rate and gradient. When the http://ijc.ccsene mobile phases even when gr run for longer 11.89 ± 0.03 m

Method Va
Validation of m method such a The procedure

Collection
Wastewater sa sewage waste drawn from ou sampled waste estrogens ( H packs to the Sciences at So were conducte

Extraction
Extraction of with some mo using GFC filt acetone, and cartridges (Bo extraction the was achieved reconstitution

Statistical
Microsoft exc deviation, mea  Abbreviations: RT = retention time; PA= peak area; Std = standard deviation; RS = relative standard deviation Table 3 and 4 display the method performance accuracy for quantification of E2 and EE2 respectively. For E2 recoveries ranged from 82.3 to 99.84 %, on the other hand recoveries for EE2 ranged from 84.6 to 103.52 %. The recoveries are within the recommended range for analytical method between 80 % to 120 % (Shabir, 2004).

Method Sensitivity
Limit of detection (LOD) and limit of quantification (LOQ) were determined using equation 1 and 2 respectively. The LOD and LOQ for E2 were found to be 0.054 ng/mL and 0.18 ng/mL respectively. The corresponding LOD and LOQ for EE2 were 0.084 ng/mL and 0.28 ng/mL respectively. This result implies that the method is suitable for moderate and higher extent of pollution of low quality water by estrogens. At low extent of pollution the method could be used in conjunction with ELISA technique. Huang & Sedlak (2001)

Levels of Estrogens in LQW Detected through HPLC-UV Method
Estrogens were detected in untreated sewage sample but in the outlets of anaerobic and facultative ponds were below the detection limit. The concentration of E2 and EE2 in untreated sewage was found to be 0.28 ng/ml and 0.18 ng/ml respectively. When ELISA competitive technique was employed revealed that the levels of estrogens in the study area were below the detection limit of the developed method. The concentrations EE2 and E2 at Mafisa WSPs ranged from 0.036 ng/ml to 0.0015 ng/ml and 0.086 ng/ml to 0.0044 ng/ml respectively (Msigala et al., 2017).

Conclusion
This paper presents a fast, linear, precise, accurate and robust HPLC-UV method for separation, detection and quantification of estrogens hormones from low quality water. The method can give precise and accurate results for moderate and higher extent of pollution of low quality water by estrogens. At lower extent of pollution (below 0.15 ng/mL) can be used in conjunction with ELISA technique, in such combination the HPLC method offset the shortcomings of the results which could be obtained through ELISA alone, on the other hand utilize the high sensitivity of ELISA technique.

Conflict of Interest
The authors declare that there are no conflicts of interest which could potentially influence this work