In-vitro extraction and separation of copper ions from human blood samples based on amoxicillin/clavulanic acid by ultrasound assisted-dispersive centrifuge liquid-liquid micro extraction

The low concentration of copper (Cu2+) can be effected on the central nervous system (CNS) and caused to multiple sclerosis (MS). Although many antibiotics can treat the bacterial infections but some of antibiotics decrease essential metal concentrations in human body and must be controlled by determining. In this study, in-vitro extraction of copper (Cu2+) with amoxicillin/clavulanic acid (AMOXC) has been studied due to interacting with metals. By procedure, Cu2+ ions were separated from blood samples by ultrasound assisted-dispersive centrifuge liquid-liquid micro extraction (USA-DC-LLME). The mixture of AMOXC (0.01 g), ionic liquid ([BMIM][PF6]) and acetone injected to 10 ml of serum blood sample at human pH=7.2. After extraction, the concentration of Cu2+ ions was determined by flame atomic absorption spectrometry (F-AAS). The LOD, enrichment factor (EF), linear range (LR) and working range (WR) were obtained 6 μg L-1, 9.92, 0.018-0.5 mg L-1 and 0.02-2.58 mg L-1, respectively (RSD<1.1%).

human blood samples depended to kind of heavy metals and chorionic/acute exposure (cadmium was seen in blood up to 1 hour as acute exposure). In chorionic exposure the concentration of heavy metals can be followed in nail or hair. Copper used for many products as an essential element in nature and copper has concentration between 80-180 μg dL -1 in blood of human adults (Mean= 1.5 mg L -1 ). Copper exposure in industry occurs primarily through inhaled particulates and according to studies, copper can be a cause of the pathogenesis of MS [6]. The value of copper (Cu 2+ ) effect on the central nervous system (CNS) and reducing of their concentration caused to numerous CNS disorders such as multiple sclerosis (MS), Wilson disease, Alzheimer and Parkinson's diseases [7][8][9]. Also, some of antibiotics caused to decrease essential metals such as Zn 2+ and Cu 2+ in human body. So, in present of antibiotics, determination of Zn 2+ and Cu 2+ in human matrixes is very important. Some receptors such as β-ketoenolbipyridine and ketoenol-pyrazole were functioned on silica hybrid adsorbent and other sorbents [10,11]. For determination of copper the different analytical techniques such as, ET-AAS [12,13], ion selective electrode, Square wave-adsorptive anodic stripping voltammetric (SW-ASV) [14] and flame atomic absorption spectrometry (F-AAS) [15] were used in different samples such as blood, waters and tissues(fish) [16]. High interferences in human blood are main problems for determination copper by instruments. So, sample preparation must be used before analysis by instruments. The different procedures such as, cloud point extraction (CPE) [9], liquid-liquid extraction (LLE) [15],solidphase extraction (SPE) in human biological fluid (blood, serum and urine) [17,[18][19][20][21][22] and dispersive liquid-liquid microextraction (D-LLME) based on ionic liquids (IL) [23,24,15] were used before Cu analysis. Recently, a applied dispersive liquid-liquid microextraction was used for separation/extraction of ions from liquid samples. In this study, USA-DC-LLME based on AMOXC was used for efficient extraction of Cu ions in human biological samples before determination by F-AAS. The hydrophobic ionic liquid was used for simple separating of Cu-AMOXC from blood samples.

Instrumental
The spectra GBC 906 double beam atomic absorption spectrophotometer was used for copper determination (FAAS, GBC, Model; Plus 906, Australia). The air-acetylene as fuel gas based on background correction was selected for F-AAS. The light of HCL adjusted on burner by vertical and horizontal positions. The Avanta software of spectra was used for collecting data. The copper hollow cathode lamp (HCL, slit=0.7) with wavelength of 327.4 nm and current of 4.0 mA was adjusted in vertical and horizontal position. For copper analysis, the limit of detection, working range was obtained 0.06 mg L -1 and 0.2-24.0 mgL -1 , respectively by F-AAS. All samples as minimum volume were used by auto-sampler injector (0.5-5 mL). The pH values of the solutions were measured by a digital pH meter (Metrohm 744), especially in human samples. All samples were shaken with Thermo accessory as mixer (250 rpm, USA) and centrifuged with Thermo fisher (1000-4500 rpm, USA).

Sample preparation
The glasses were washed with a HNO 3 solution (0.5M) for at least 24 h and thoroughly rinsed 10 times with ultra-pure water. For sampling, 20 ml of samples of blood were collected from subjects and control peoples (N=30, men, 30-55 age). 10 mL of blood sample mixed with heparin (pure, 20 micro liter) and storage in low temperature. Then, 10 mL of blood samples were stay at the room temperature for 30 minutes to coagulate and then centrifuged at 3000 rpm for 5 minutes. Next, serums were separated by a sampler. Plasma was also prepared by procedure by centrifuging process. The human blood/serum samples were maintained at -20 °C up to 72 h. In humans with low sample volume of blood, it diluted up to 20 mL and then was used. The Ethical Committee of Azad University (SN.IAU.ُ SRB.940522664) confirmed the human sample method.

Procedure
By USA-DC-LLME procedure, the 10 mg of pure AMOXC powder (15 micro molar conc.) mixed with ionic liquid ([BMIM][PF6], IL, 0.1 g) in present of 500 micro liter of acetone/ethanol, then, the mixture was injected to 10 ml of serum blood sample at human pH=7.2 (Fig.2). After 5 min shaking time, the hydrophobic ionic liquid BMIM] [PF6] was separated by centrifuging at 3500 rpm for 4 min in conical glass tube. The upper liquid phase separated by auto-sampler accessory (1-100 mL) and remained phase (IL-AMOXC/Cu 2+ ) were backextracted from IL phase by decreasing pH up to 1 (HNO 3 , 0.5 M, 0.5 mL). Finally the concentration of Cu 2+ ions was simply determined by F-AAS after dilution with DW up to 1 mL.

Results and discussion
For extraction copper from blood samples all parameters (pH, sample volume, amount of AMOXC and IL) must be optimized. The serum, blood, urine samples was prepared from patients with copper disease such as multiple sclerosis. After tuning pH with favorite buffers, the sample determines by ultrasound assisted-dispersive centrifuge liquidliquid micro extraction (USA-DC-LLME) coupled to F-AAS.

PH effect
The effect of sample pH on complexation of Cu(II) ions on AMOXC was studied for 0.2 mg L -1 and 2.0 mg L -1 of Cu(II) (lower limit of quantification; LLOQ and upper limit of quantification; ULOQ) from different pH 1-12. The complexation of AMOXC-Cu was depended on the pH of solutions and function group of ligand. The results showed, the maximum extraction of Cu(II) was achieved in human pH (7.0), and the recovery values for Cu(II) were below 5% in acidic or basic pH. The extraction mechanism of Cu(II) ions with AMOXC was highly referred to pH and complex formation between Cu(II) ions and dative sulfur bonding of AMOXC (Fig. 3). The sulfur groups can be deprotonated (:S-R) at wide range of pH from 6 to 9. The extraction efficiency of Cu(II) at pH values below 6 and upper of 8 cannot occurred due to similarity charge or precipitation of hydroxyl copper (Cu(OH) 2 ).

Sample volume effect
The sample volume in one of the important factors which must be optimized for Cu extraction from blood samples by AMOXC. As evaluation extraction efficiency, different volumes between 1.0-30 mL of blood and serum samples was examined and optimizes in pH=7. The results showed us that the maximum recovery for copper extraction based on AMOXC was achieved for less than 12 mL for human blood samples and less than 17 mL of standard solutions. Therefore, 10 mL and 15 mL of sample volume was selected as optimum point for copper extraction for blood and water samples, respectively by USA-DC-LLME procedure (Fig.4).

AMOXC effect
The amount of ligand is very important for copper extraction in blood samples, so, the concentration of AMOXC in human samples must be studied for high extraction. For this purpose, the vary concentration between 1-50 micro molar concentration of AMOXC was prepared and used for evaluation of copper extraction in blood/standard samples by USA-DC-LLME procedure. The experimental results showed, the high extraction of copper was obtained with more than 15 micro molar of AMOXC in 10 mL of samples (~ 10 mg pure powder) by USA-DC-LLME procedure (Fig. 5). Also, the extraction efficiency more than 97% was made by 12 micro molar of AMOXC concentration in 10 mL of water samples. First, 10 mg of pure AMOXC powder mixed with ethanol/acetone (0.5 mL) and then added to 0.1 g of IL in 2 mL syringe. After shacking, the mixture injected to 10 mL of blood samples. The extraction efficacy was decreased about 74.6% when the mixture added step by step without shacking.

Ionic liquid effect
Separation process for AMOXC-Cu from liquid samples (blood, standard solution) is very difficult. The different techniques with filter, centrifuging and organic solvents was used for separation ligands from liquid phase. Recently, the benign solvents as green solvents (IL) was introduced in many papers. In this study we used different hydrophobic ionic liquids for separation processes in blood samples. The different volumes or gram (0.05-0.3 g) of 1-Ethyl-3-methylimidazolium acetate, 1-Ethyl-3methylimidazolium tetrafluoroborate, 1-Ethyl-3methylimidazolium hexafluorophosphate, and 1-butyl-2,3-dimethylimidazolium hexafluorophosphate [BDMIM][PF 6 ] as a hydrophobic ionic liquids were was selected for separation AMOXC from blood sample. Based on results, 0.1 g of [BDMIM][PF 6 ] has more extraction efficiency as compared to others (Fig.   6). The 1-Ethyl-3-methylimidazolium acetate and 1-Ethyl-3-methylimidazolium tetrafluoroborate had lower recovery because of solubility in liquid phase and missed of IL. The1-Ethyl-3-methylimidazolium hexafluorophosphate as hydrophobic IL can be separated copper-AMOXC from blood samples but has lower recovery (85%) as compared to [BDMIM] [PF 6 ] with (97.8%).

Interference ions study
Many ions such as Zn 2+ , Mn 2+ , Na + , K + , HCO 3 -, SO 3 -, Fe 2+ , Co 2+ , exist in water or blood samples and can be effected on complexations of copper with AMOXC in presence of interference ions. The complexation copper depended on power of ligand and competition  of ions in different pH. So, in optimized pH many ions had no effect on extraction process, but some of them may be done. Therefore the effect of important ions which was bonding with amino acids or proteins was studied. The recovery of copper extraction based on AMOXC in presence of different concentration of 1-4 mg L -1 of cations and anions was evaluated by USA-DC-LLME procedure. As selectivity of determination of F-AAS, the concentration of other ions can't reported. So, the recovery was checked as effect of interference ions. The results showed, the interference ions had no effect on complexation processes of copper -AMOXV by USA-DC-LLME procedure (Table 1).

Validation procedure
The methodology of copper extraction must be validated by different technologies. By spiking of blood, serum, urine, plasma samples, the favorite recovery more than 95% was achieved for LLOQ and ULOQ range. So, the USA-DC-LLME procedure was validated for copper complexation based on AMOXC (Table 2). In addition, ET-AAS coupled with microwave digestion was used for evaluation of purposed procedure after dilution samples up to 100 mL(1:10) ( Table 3). As intra-day and interday studies, the mean copper concentration based on AMOXC in subjects and control groups (N=30, men, 30-55 age) was checked by USA-DC-LLME procedure. The results showed no significance difference between subjects and control groups with favorite p-value ( Table 4). The mean concentration of copper in control groups were a little higher than Cu blood subjects. The regression analysis and t-test were achieved between Cu in subject and control groups. There were a correlation (0.45< r <0.5) between blood of subject and control groups. (p-value < 0.001).

Conclusions
A simple Pharmacology and biology method based on USA-DC-LLME was used for evaluating of AMOXC for copper extraction in human blood samples. The results demonstrated that AMOXC can decreased copper concentration in blood samples in efficient time. On the other hands amino acids and proteins was also complexed with copper and caused to decrease copper extraction by AMOXC. So, the AMOXC competed with amino acids and proteins for extraction copper and other metals at pH 7.2. In addition, the antibiotics such as AMOXC can treat the bacterial infections but they can decrease essential metals (Cu) in biological matrixes and cause disease in humans. By purposed procedure, the copper extracted with AMOXC at pH=7 and determined by F-AAS. The 1-butyl-2,3-dimethylimidazolium hexafluorophosphate ([BDMIM][PF 6 ] as hydrophobic IL was used for separation of Cu from liquid phase. In optimized conditions, the low LOD and RSD% values as well as good working ranges (0.02-2.58 mg L -1 ) and high recoveries caused to consider as a recent innovative procedure.

Acknowledgements
The authors thank Azad University of Medical Sciences for supporting this work. The Ethical Committee of Azad University confirmed the human sample method (SN.IAU.ُ SRB.940522664).