Development and validation of quantitative methods for the analysis of clinical biomarkers of iron metabolism applying stable isotopes

Abstract

The continuous discovery of new biomarkers is reflected not only in the scientific literature but also in its application in current clinical practice. The urgent need to determine the values of appropriate clinical parameters (biomarkers) in biological fluids is driving the development of new analytical methodologies capable of providing quantitative, precise, accurate and validated information on the concentration of those biomarkers, both in control individuals as well as in patients affected by the considered pathologies. In this regard, diseases associated with disorders of the iron metabolism affect a significant proportion of the population, especially in the case of iron deficiencies (e.g. Fe-dependent anemia) as well as iron overload (e.g. hereditary hemochromatosis). In 2001, a liver-synthesized peptide hormone called hepcidin-25, present in biological fluids in the ng ml-1 range, was discovered. Despite its present importance, there is currently a lack of analytical methodologies validated for determination of this putative biomarker to the point of enabling its implementation in clinical routine analysis. While hepcidin-25 controls Fe absorption from food and its release into circulation, a significant part of this trace metal is stored intracellularly, incorporated in the protein ferritin. However, a small fraction of this protein in the ng ml-1 range can be detected in serum. Serum ferritin concentration forms part of a series of biomarkers routinely determined in clinical laboratories to assess the iron status in patients. However, the determination of ferritin-bound iron in serum could offer a higher potential to diagnose iron disorders than pure ferritin measurements since the concentration of the protein is altered in response to inflammation as well. In this context, the development and validation of new analytical methods based on elemental and molecular mass spectrometry for the reliable determination of putative clinical biomarkers is currently an urgent need. Therefore, the main objective of the present dissertation is the development and validation of MS-based quantitative methods, based on the use of stable isotopes, for the determination of clinical biomarkers in disorders of iron metabolism. This overall aim was pursued through the following specific objectives: Development of a quantification strategy of the peptide hepcidin-25 in human urine by HPLC-ICP-MS. Since the first publications dealing with hepcidin-25 and its biological function, more than one decade ago, many progresses have been made in the quantification of this peptide hormone. Thus, for the quantification of the hormone hepcidin-25, various methodologies have been developed throughout this decade which are mainly based on antibodies or those that employ molecular mass spectrometry (SELDI-MS or LC-MS). Of course, it should be taken into account that both strategies are accompanied by some disadvantages. Thus, this study has been focused towards the development of a new elemental method for the determination of hepcidin through the detection of sulfur, using ICP-MS as a detector, after capillary liquid chromatography for the separation of the species. Sulfur (which is present in nine of the 25 amino acids of the hepcidin-25) can be detected by ICP-MS after the elimination of spectral interferences. For this purpose, three different ICP-MS instruments were compared with each other, whereby each of them uses a different strategy to eliminate the polyatomic interference (e.g. collision cell, sector field (SF) and formation of oxides). Finally, ICP-SF-MS offered the best limits of detection, accuracy and precision. Quantitative recoveries of the peptide standard with the proposed configuration (LC-ICP-MS) were obtained maintaining the column temperature at 50°C. Then, various cleaning procedures of the sample were studied in order to apply that methodology in urine samples. The best results were obtained by a combination of dialysis and solid phase extraction. The treated extracts were analyzed by µHPLC-ICP-MS and also by LC-ESI-QTOF-MS (for confirmatory purposes). The results revealed that screening by ESI-QTOF-MS is more suitable for hepcidin in urine samples, regarding both the selectivity and sensitivity, mainly due to the relatively high limits of detection of sulfur by ICP-MS. Evaluation of a direct chemical labeling strategy with divalent metal ions (Cu2+) for the quantitative analysis of hepcidin in serum samples. Taking into account the limitations of quantification of hepcidin-25 through elemental detection of sulfur, the second objective was focused on the use of a labeling strategy, using heteroelements, by taking advantage of hepcidin's affinity to form stable complexes with bivalent metal ions such as Cu2+. This type of selective union would allow the quantification of the peptide by Cu detection via ICP-MS. Therefore, an incubation of the peptide with Cu2+ is required, followed by chromatographic separation (anion exchange chromatography) of the complex from the excess of metal ions applied for incubation, but maintaining the integrity of the complex through the separation. For this purpose, conditions of incubation (between the hepcidin-25 and Cu2+) were optimized by using ESI-QTOF-MS. It was found that the complex "Cu:hepcidin-25" is stable under physiological conditions and shows an equimolar stoichiometry (1:1). Once the stability of the complex through chromatographic separation was demonstrated, two IDMS strategies employing ICP-MS for detection were developed for the quantification of hepcidin-25. The first one was based on the post-column addition of a 65Cu tracer (species-unspecific or post-column isotope dilution analysis approach), and secondly, by the synthesis of the complex "65Cu:hepcidin-25" used as spike for IDA (species-specific isotope dilution analysis). Analytical performance of each of both methods was critically compared in serum samples. The determination of hepcidin-25 in different serum samples of healthy individuals showed a concentration range of about 7.7-19.3 ng ml-1 (95% confidence level), a value in agreement with previously published data. Development of a quantification method for serum ferritin by using a "sandwich"-type immunoassay linked to ICP-MS. In this section, the use of an antibody-labeling strategy (using Ru-labeled antibodies) for the absolute detection of ferritin without the need for chromatographic separation was evaluated. The developed method uses two monoclonal anti-ferritin antibodies: one of them biotinylated and the other marked with a chelate of ruthenium [Ru(bpy)3]2+. In aqueous solutions, a complex (bioconjugate) between the ferritin and two antibodies is formed. The use of streptavidin-coated magnetic microparticles allows to retain the analyte as well as to remove the excess of antibodies and unbound species. Subsequently, the resuspended particles can be directly introduced into the ICP-MS using flow injection (FI). Since the Ru complex also allows the quantification by electrochemiluminescence (ECL), the combination of both data (ICP-MS and ECL) enabled the establishment of the stoichiometry between the ferritin protein and the element Ru ~1:23). This provided the basis for ferritin quantification by post-FI using isotopically enriched 99Ru (species-unspecific isotope dilution analysis) without the need for ferritin calibration standards. The developed strategy allowed the absolute quantification of ferritin at the femtomolar level with good precision ~8%) and recoveries between 85-109% (determined using a recombinant ferritin reference material NIBSC 94/572). Development of a quantitative strategy to address the ferritin-bound iron concentration in human serum. For this aim, two different analytical methods were evaluated: in the first approach a multi-dimensional purification strategy was developed in order to separate the ferritin-bound iron from other possible iron-containing species. After the purity of the analyte was successfully evaluated (by using size exclusion chromatography coupled to ICP-MS), quantitative analysis of the ferritin-bound iron was carried out by species-unspecific IDA-ICP-MS. The achieved analyte recovery rates were found to be relatively low (31 ± 10%). Therefore, individual recoveries have to be taken into account. In order to overcome this limitation, the second strategy employs species-specific IDA-ICP-MS for the determination of ferritin-bound iron. In this vein, the synthesis and characterization of isotopically enriched 57Fe-ferritin as tracer was accomplished by a) biosynthesis of recombinant plant ferritin by heterologous expression in E. coli (scientific research stay in the National University of Singapore) and b) by incorporation of 57Fe into apo-ferritin. Here, the most promising result were obtained with the second strategy which in turn uses ammonium iron(II) sulfate as 57Fe(II) source. The fully characterized spike enabled the quantification of ferritin-bound iron by HPLC coupled to ICP-MS using species-specific isotope dilution analysis. With the concentration of iron determined by ICP-MS and the concentration of ferritin (determined by ECLIA) the Fe:ferritin ratios can be calculated. This ratio is a putative clinical biomarker which may prove invaluable to assess disorders of iron metabolism.