Aristotle University Of Thessaloniki (AUTH)
Ανάπτυξη και επικύρωση αναλυτικών μεθόδων για προσδιορισμό φαρμάκων και βιοδεικτών σε βιολογικά δείγματα
Abstract
dc:descriptionThe objective of this thesis is the development and validation of analytical methods for the identification of biomarkers and drugs in biological samples through metabolomics-based approaches. Metabolomics constitutes an interdisciplinary field that combines modern instrumental analytical techniques (mass spectrometry, NMR spectroscopy) with advanced statistical modeling, aiming in the holistic profiling of metabolites that are present in a biological system. Metabolomics-based studies are applied in various scientific fields, such as life sciences, biology, chemistry, requiring the use of bioinformatics in order to statistically analyze large amounts of analytical data. In life sciences metabolomics is used as a useful tool for the investigation of the potential of metabolomic profiling in early disease diagnosis, personalized therapy or drugs efficacy and toxicity. Metabolomics may provide new biomarkers at molecular level that can be used as a more efficient, less invasive means for these purposes. However, despite the great applicability of such studies, significant limitations still occur, not allowing the implementation on large-scale analysis of large number of samples. For example, existing biomarkers used so far for some diseases for diagnosis and monitoring, such as e.g. disorders of neonatal thyroid gland (Congenital Hypothyroidism), renal impairment in infants and neonates with congenital narrowing of the pyelonephritis contribution (UPJO) and cardiovascular disease (CAD) are characterized by low specificity. Also, many of the metabolites and enzymes are widely distributed in neonates, especially in those born prematurely, giving false-negative results. Other limitations are focused on low analytical reproducibility of measurement or in the difficulty of analyzing specific categories of compounds, such as highly polar compounds or non-polar ones (e.g. lipids). Moreover, in some cases analysis of endogenous compounds is difficult due to the absence of analyte-free matrices, or of Certified Reference Material (CRM). Another problem is system contamination of biological matrices, especially in liquid chromatography that results in reduced efficiency of analysis. Therefore, the revocation of the above restrictions is of great importance in order to use metabolomics into final practices. Nowadays, scientific interest focuses in the establishment of validated metabolomics – based studies, as well as in the establishment of retention index libraries in LC, that will allow the rapid, effective and valid identification of unknown compounds in untargeted analyses. Retention index libraries will help overcoming the problems occurring from analytical standards, isotope-labeled standards or CRM absence. On the other hand, life sciences focus in the development of sensitive diagnostic tools for early diagnosis and treatment of various diseases, improving patients’ life. Thus, both targeted and untargeted approaches were conscripted for the determination of endogenous compounds and a drug in biological matrices. The collected biological samples were analyzed with state-of-the-art analytical techniques, as gas and liquid chromatography coupled to mass spectrometry (GC-MS, UPLC-MS/MS, UHPLC-HRMS). Also, an attempt was made to calculate the retention index of a large number of endogenous compounds, in order to establish a universally applicable retention index library. Accurate determination of retention time is very important in confirming the detection of analytes, especially in complex matrices. Usually, the identification of a compound is based on the absolute match of the retention time and spectrum of the unknown compound to a standard reference. In liquid chromatography, variation in retention time is very common between different laboratories, instrumentation, chromatographic systems, etc., and the solution of using reference standards is not always feasible. Thus, an idea of overcoming this problem, is the establishment of universal retention index libraries. For this purpose, 95 endogenous metabolites, including amino acids, organic acids, carnitines, etc., and a reference material containing 20 homologous compounds (1-alkylpyrinium-3-sulfonate, APS) were analyzed using a RPLC-MS/MS method. Retention times for all analytes were recorded, and their respective retention indices were calculated based on the equation RI=RI0 + (RI1 – RI0) x [(RT-RT0)/ (RT1-RT0)]. A similar experimental procedure took place in 4 collaborating laboratories, using different instrumentation but exactly the same protocol of analysis regarding analytical column, mobile phases, column temperature and standards. Obtained data showed a linear correlation of R2> 0.8899 for all laboratories, proving results’ reproducibility and robustness despite different instrumentation but also demonstrating the applicability of a universal library among different laboratories. The next two studies were related to method development and validation of the analysis of drug in human urine and the analysis of four ceramides in human serum, respectively. Firstly, a NICI-GC-MS method was developed for the absolute quantification of acetazolamide and other sulfonamides in human urine. It was the first study, where sulfonamides derivatization with PFB-Br in mild conditions was reported. The method was based on an easy and quick sample preparation procedure including the evaporation of a small urine volume (50 μL), derivatization of sulfonamides and internal standard with 30% PFB-Br in ACN (60 min, 30 °C), evaporation and reconstitution with toluene. Quantification of acetazolamide and its respective internal standard was performed using the R-SO2-N (PFB)2 derivatives with m/z 581 and m/z 584, respectively. Method accuracy and precision were between 95.3% - 109% and 0.3% - 4.2%, respectively, while LOD and LOQ were 300 fmol and 1 μM, respectively. Regarding the other sulfonamides tested, only metazolamide and dorozolamide gave similar derivatives using PFB-Br, suggesting that the use of PFB-Br is limited to the size of the chain of this particular compound class. The developed and validated method was applied to real urine samples of an adult volunteer for the quantification of acetazolamide at different time points, after drug Acemit consumption. The second method was on the quantification of four specific ceramides, namely Cer(d18:1/16:0), Cer(d18:1/18:0), Cer(d18:1/24:0), and Cer(d18:1/24:1), in human serum with a RPLC-MS/MS method. These specific ceramides were found to play an important key-role to various disease states, such as diabetes, cardiovascular diseases, cancer, etc. Different sample preparation methodologies were tested for accurate and efficient extraction of ceramides from human plasma, with liquid-liquid extraction using dichloromethane: methanol, 2:1 v/v being selected. Supported liquid extraction also presented satisfactory results, regarding accuracy, precision, extraction recovery and matrix effect. Final sample preparation protocol included extraction of the compounds of interest into organic phase, evaporation to dryness and reconstitution in the mobile phase. For the quantification of the endogenous compounds the use of surrogate matrix was also tested. As such five % of bovine serum albumin was compared to both an external and a standard addition curves, providing satisfactory results. The proposed method showed good linearity for all compounds (R2>0.99), as well as satisfactory accuracy and precision. Extraction recovery ranged between 82.2% to 120.2% for all ceramides. The developed method was then applied to real serum samples of CAD patients, in order to confirm method’s suitability into clinical practice. The followed three chapters were related to the application of metabolomics-based studies for biomarkers discovery into two clinical studies in human cohorts and one nutritional intervention in animals. The first study focused in the metabolic profiling of dried blood spots (DBS) of infants with congenital hypothyroidism. In total, 36 sampled of infants with CH and 24 samples of healthy matched controls, were analysed with both an untargeted gas chromatography-mass spectrometry method and a targeted liquid chromatography tandem mass spectrometry method. From the target analysis only 14 metabolites were detected in DBS, including creatinine, benzoic acid, nicotinamide and others, probably due to low sensitivity. Statistical analysis failed to highlight any statistically significant difference between the tested cohorts. Thus, untargeted GC-MS analysis was performed. Three hundred and forty-seven features were detected and identified using AMDIS and GaVIn software, with 96 of them passing all analytical criteria set. Univariate statistical analysis performed highlighted four amino acids, namely glutamate, glutamine, lysine and methionine to significant up-regulated or down-regulated in hypothyroidism group. The results are in agreement with the literature, confirming the fact that amino acids are key-compounds directly related to the thyroid gland state. The next study targeted in the analysis of a specific class of compounds, in order to distinguish the metabolic profile of serum and urine of infants with severe or mild ureteropelvic junction obstruction (UPJO) compared to healthy matched controls. Two targeted methodologies were applied using HILIC-MS/MS for the determination of 100 polar endogenous compounds and GC-MS for the analysis of amino acids and derivatives, as well as for nitrite, nitrate and malondialdehyde. Both univariate and multivariate statistical analyses highlighted a panel of metabolites significantly altered between the groups tested. From the differentiated metabolites great importance is given in homocysteine, choline, isoleucine, homoarginine, malondialdehyde and asymmetric dimethylarginine in serum and in glutamic acid, xanthine and hypoxanthine in urine. The results indicate that a panel of validated metabolites can be used in clinical practice for early and accurate diagnosis and prognosis of the disease. In the final chapter two metabolomics-based methodologies were applied for the evaluation of the nutritional effects of carob in everyday diet. Carob is considered to be of highly nutritional value, as it is rich in vitamins, tannins and nutrients. The beneficial effects of carob against cancer, metabolic syndrome, diabetes, diarrhea, hyperlipidemia and gastro esophageal reflux disease are only a few of its therapeutic actions. For this reason, 8 male Wistar rats were treated with carob powder for a 15-day period and compared to 8 non-treated rats. Fecal and urine samples were collected at 5 time points (0, 1, 5, 10 and 15 days) and were analysed with both HILIC-MS/MS method and RP LC-HRMS method. Based on the targeted approach a clear group separation was observed after day one and day fifteen on fecal metabolome, while a mild one was observed on day one on urine metabolome. Twenty-one fecal metabolites were differentiated, including amino acids and their derivatives, vitamins and organic acids, and 7 metabolites were found to be altered in rat urine samples. On the other hand, untargeted approach was able to separate the two groups based on fecal metabolome at day fifteen. Twenty-one features were altered in positive ionization mode, among which phenylalanine and dodecanedioic acid were identified. Metabolic alterations in fecal samples could be attributed to physiological and biochemical adaptations derived from the nutritional intervention. Fecal targeted metabolomics were proven to be suitable for highlighting such alterations.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Begou, Olga-Angeliki
- Μπέγου, Όλγα-Αγγελική
Subjects
dc:subject × 22- Ανάπτυξη μεθόδου
- Μεταβολομική
- Βιοδείκτες
- Υγρή χρωματογραφία συζευγμένη με φασματομετρία μαζών
- Αέρια χρωματογραφία - φασματομετρία μάζας
- Method development
- Metabolomics
- Biomarkers
- Gas chromatography-mass spectrometry
- Liquid chromatography – tandem mass spectrometry
- Χημεία
- Φυσικές Επιστήμες
- Αναλυτική χημεία
- Επιστήμες Υγείας
- Ιατρική και Επιστήμες Υγείας
- Επιστήμες υγείας, άλλοι τομείς
- Chemical Sciences
- Natural Sciences
- Analytical Chemistry
- Health Sciences
- Medical and Health Sciences
- Health sciences, miscellaneous
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/49190