Hyphenated Mass Spectrometry Techniques And Their Applications To Elemental Speciation Analysis

The bioavailability and toxicity of an element are highly dependent on the chemical species of the element. It is important to characterize and determine the concentrations of individual chemical species, not only the total concentration of the element. Chemical species of the element vary on the basis of the valence state (redox), organometallic species, complex formation, and acid/base equilibrium. The concentrations of trace element species are often extremely low. They are present in complicate matrix in biological system. The sample volume is usually very limited. Therefore, determination of trace element species in biological systems is analytically challenging.Arsenic, a ubiquitous element, exists as more than50different chemical species in the environment and in biological systems. An organic arsenic species,3-nitro-4-hydroxyphenylarsonic acid (Roxarsone, ROX), has been widely used as a feed additive in poultry industry for decades. There is much concern and debate over the environmental and health risks associated with the use of ROX in animal feeds. However, most metabolites of ROX have not been identified, and it is not clear how ROX is metabolized. Without the knowledge of the arsenic species and their concentrations present in chickens fed the ROX-containing food, scientists cannot reliably assess the human exposure to arsenic species from consumption of chicken. One of the objectives of this thesis research is to fill this knowledge gap, by characterizing and determining arsenic species in the liver of chickens fed either the ROX-containing food or the control food.A35-day chicken feeding study was conducted, involving a total of1600chickens of two commercial strains (Ross308and Cobb500).800chickens of each strain were randomly divided to the control group and the treatment group. The control chickens were fed a basal diet. Chickens in the treatment group were first fed a ROX-supplemented diet for28days and then fed the basal diet (without supplementation of ROX) for another week. This experimental design allows for studies of uptake, metabolism, and elimination of arsenic. Liver samples were obtained from a subset of 256chickens, for the studies of arsenic speciation. The comprehensive arsenic speciation analyses required the development of a hyphenated HPLC-ICP-MS/ESI-MS/MS method that incorporated HPLC separation with complementary atomic (ICP-MS) and molecular (ESI-MS/MS) mass spectrometry detection.Three new metabolites of ROX were identified using the hyphenated technique. The toxicities of phenylarsenicals were tested on T24bladder cancer cells. The trivalent arsenicals were found to bind to proteins. Labeling of proteins with elemental tags, can exploit the exceptional capabilities of ICP-MS for virtually all proteins. Elemental tagged proteomic is expected to provide new functional insights into biological processes, increasing interest in quantitative determination of proteins. A number of strategies have been employed for the comparative investigation of different proteomes on the basis of absolute protein quantification. Furthermore, new techniques are necessary to allow for the investigation of a large set of proteins or peptides at the same time without the need for highly specific standards. Generally, the selection of metals was based on three aspects:(1) the metals could be chelated strongly to chelating reagent;(2) the metals should not be naturally present in the samples (which means a low blank) and (3) the metals should have high sensitivity with less interference in ICP-MS detection. Cd, Hg, Pb were used to label various proteins. Two novel nonisotopic immunoassay for quantitative determination of carcinoembryonic antigen and multiple glycoproteins was developed using affinity magnetic separation, elemental labelling, and ICP-MS detection.The overall objectives of this research are to develop hyphenated mass spectrometry techniques and to demonstrate their applications to elemental speciation analyses. The major contents of this dissertation are described as follows:1) The primary objective of the first component of this thesis is to establish the method for arsenic speciation in chicken liver. Chickens were fed with either a Roxarsone-supplemented diet (treated group) or a diet free of Roxarsone (control group). Each chicken liver sample was homogenized. Arsenic species in the homogenized liver samples were extracted with the assistance of enzyme digestion and sonication. The use of pepsin digestion improved the extraction efficiency. The extracts were subjected to analysis by high performance liquid chromatography (HPLC) separation with both inductively coupled plasma mass spectrometry (ICP-MS) and electrospray ionization tandem mass spectrometry (ESI-MS/MS) detection. Anion exchange HPLC enabled the separation of Roxarsone and other potentially occurring arsenic species within10min. Detection with both ICP-MS and ESI-MS/MS allowed for identification and quantification of arsenic species in the samples. The methods are used to investigate arsenic speciation in the control and Roxarsone-fed chicken.2) It is necessary to selectively capture and pre-concentrate trace arsenic species for their subsequent characterization. For this purpose, a hybrid organic-silica monolith was synthesized using a simple "one-pot" reaction process. This hybrid monolith is functionalized with phosphate and titanium(IV). This unique affinity monolith is able to selectively capture diverse range of arsenic compounds, including inorganic, methylated, and phenyl arsenicals. Concomitant ions, such as phosphate, present at concentrations as high as10,000fold excess over the arsenic concentration, do not interfere with the retention of arsenic species. The selective capture and the subsequent release of arsenic species substantially improve the detection of trace arsenic species and reduce matrix interference. The removal of sample matrix enables direct HPLC electrospray ionization mass spectrometry analysis of trace arsenic species in complex sample matrix, e.g. chicken liver, muscles, kidney, and skin.3) The combination of pre-concentration, chromatography separation, and complementary mass spectrometry detection allowed for the identification of new arsenic metabolites in chicken liver. Three new metabolites of ROX in ROX-fed chicken liver were identified as methylated3-amino-4-hydroxyphenylarsonic acid (3-AHPAA), methylated N-acetyl-4-hydroxyphenylarsonic acid (N-AHAA) and methylated ROX. Their molecular weight of231,273,261and the characteristic fragments at m/z of91(AsO-),107(AsO2-),123(AsO3-) were consistent with the identity of these arsenic species. These three methylated phenylarsenicals were synthesized, as a part of this thesis. The synthesized arsenic compounds were further used to confirm the identity of the new metabolites. In vitro enzyme assays were performed using As3MT/wt enzyme and trivalent phenylarsenicals as the substrate. The time-dependent formation of methylated arsenicals suggested the bio-transformation of various metabolites of ROX. The arsenic speciation results supported the possible biotransformation pathway of ROX in chicken liver.4) The arsenic speciation techniques described above were further used to study the temporal profiles of arsenic species in the liver samples from a subset of256chickens in the35-day feeding experiments. The concentration of ROX in the ROX-fed chicken liver increased during the feeding period to the maximum of275.9μg kg-1(Cobb500) and318.9μg kg-1(Ross308) on day28, as compared to6.5μg kg-1(Cobb500) and10.7μg/kg (Ross308) in the control chickens. The difference between the treated group and the control group was statistically significant (p<0.01). Total arsenic concentrations also increased using the first28days, and reached the maximum of1023.8μg kg-1(Cobb500) and1309.6μg kg-1(Ross308) on day28. After feeding of ROX stopped, arsenic was eliminated quickly from the chickens. However, at the end of the7-day clearance period, the concentrations of arsenic remaining in the ROX-fed group were207.0μg kg-1(Cobb500) and185.9μg kg-1(Ross308). These were still significantly higher than the background concentrations of arsenic in the control group:54.4μg kg-1(Cobb500) and46.5μg kg-1(Ross308). In the ROX-fed group, the residual concentration of total inorganic arsenic was19.9μg kg-1and the sum of MMA and DMA was5.5μg kg-1.5) The cytotoxicity of trivalent phenylarsenicals were tested on T24bladder cancer cells. T24cells were exposed to varying concentrations of the synthesized trivalent phenylarsenic compounds and monitored for80hours post-treatment. At the highest concentration tested,50μM for3-AHPAAⅢ, and0.75μM for ROXⅢ, there were no detectable viable cells. The ICso values for3-AHPAAⅢ and ROXⅢ on T24after24h were21μM and0.19μM. Thus, the ICso for the trivalent ROXⅢ was30,000times lower than that for the pentavalent ROX (5.7mM). The cytotoxicities of the five arsenic compounds were in the order of ROXⅢ>3-AHPAAⅢ>3-AHPAA> ROX>N-AHAA. The interactions between ROXⅢ and glutathione (GSH), hemoglobin (Hb) and metallothionein (MT2) were further studied. Consistent with the fact that MT contains20cysteines, each MT was able to bind to10ROXⅢ molecules, and the binding kinetic was faster than binding to inorganic AsⅢ.6) A sensitive and selective method based on mercury labeled magnetic immunoassay and inductively coupled plasma mass spectrometry (ICP-MS) detection was developed for the quantification of carcinoembryonic antigen (CEA) in real human serum samples. Size exclusion chromatography (SEC) combined with ICP-MS was applied for the separation and detection of secondary antibody labeled with mercury and excess mercury. After a complete sandwich immunoreaction, with consumption of only50μL sample solution, the established method presented a limit of detection of0.041ng mL-1for CEA, with the relative standard deviation (RSD) of5.2%(c=8ng mL-1, n=7). The response of magnetic solid phase extraction (MSPE)-ICP-MS method for CEA was linear over a dynamic range from0.1to50ng mL-1. The developed method evaluated by three different human serum samples was versatile and could be easily extended to other proteins quantitative analysis, especially for specific biomarker determination.7) A sensitive and selective method using magnetic separation coupled with inductively coupled plasma mass spectrometry (ICP-MS) detection was proposed for simultaneous determination of multiple glycoproteins. Haptoglobin (HP), hemopexin (HPX) and ovalbumin (OVA) were selected as target glycoproteins. And they were firstly selected by lectin conjugated magnetic particles due to the glycan-structure, and then immunoreacted with antibodies labeled with Cd, Hg, Pb through poly(acrylic acid), respectively. The metal ions, corresponding to the concentration of glycoproteins, were released from antibody with an acid-dissolution step and subjected to subsequent ICP-MS detection. With a consumption of only50μL sample solution, the established method presented a limit of detection of0.032,0.027and0.13ng mL-1for HP, HPX and OVA, with the relative standard deviation of7.4%,6.8%and7.6%(c=5ng mL-1, n=7), respectively. The response of magnetic immuno-ICP-MS assay for HP, HPX and OVA was linear over a dynamic range of0.1-100ng mL-1,0.1-100ng mL-1and0.5-100ng mL-1, respectively. The recoveries for HP, HPX and OVA in spiked human serum samples were in the range of97.6-105%. The developed method was demonstrated to be versatile to multiple glycoproteins quantification in complicate samples.