This study is focused on the development of a multiresidue analytical method for determination of steroid hormones (progestins, oestrogens, androgens and glucocorticoids) in vegetables and fruits. Pollutants’ presence -especially those with endocrine disrupting properties as steroid hormones- in edible plants is a topic of high concern, related to the use of reclaimed wastewaters for irrigation. Once in the soil, plant uptake, root accumulation or translocation to the leaves and fruits can occur depending on the physicochemical properties of the contaminants (pKa and logKow). The herein considered analytes present a wide logKow range, from 1.46 of Cortisone to 4.02 of 17β-estradiol, thus they tend to accumulate differently in edible parts (e.g. roots, leaves and fruits). Therefore, a root vegetable (carrot), a leafy vegetable (spinach), and a fruit (strawberry) were chosen as probes for this study. Ultrasound-assisted extraction (UAE) has been selected due to its short extraction times, reduced organic solvent consumption, energy and costs saved. The extraction procedure was studied in freeze-dried matrices, processing samples before (blank matrices) and after spiking (for each analyte 100 ng/g dry weight). Firstly, the role of different extraction solvents, namely Acetone (ACE), Methanol (MeOH), Acetonitrile (ACN), mixture MeOH: ACN (75:25, v/v) and mixture MeOH: ACN (50:50, v/v), was evaluated. A single extraction was performed (5 mL of solvent, 5 min, r.t.). The extract was evaporated to dryness under N2 stream, reconstituted in 1 mL of MeOH, and analysed by HPLC–MS/MS. The best overall performances in terms of recovery (%R) and matrix effect (%ME) were achieved with MeOH as demonstrated by ANOVA test. The extraction efficiency was then optimized by a chemometric approach to develop a multiresidue method. A 23 factorial design was constructed considering three variables at two levels: number of cycles (1 and 3); solvent volume (2 and 5 mL); sonication time (1 and 10 min). Each one of the eight experiments was replicated three times, once for each matrix. The multi-analyte recovery was selected as the experimental response for each experiment. The elaboration (R-based software CAT, Chemometric Agile Tool) showed that the number of cycles is significant with p <0.01 (**) like the interaction between number of cycles and solvent volume. Accordingly, the optimized extraction is comprised of 3 x 1 min ultrasonic cycles with a small volume of organic solvent (2 mL MeOH). To reduce matrix effects resulting from co-eluting residual matrix components, an additional clean-up step was studied entailing different sorbent materials and approaches: Activated Chemviron Carbon (d-SPE), Florisil (SPE), Primary-Secondary Amine (SPE), C18 (SPE), SUPERCLEANTM LC-NH2 (SPE). The latter was selected as SPE sorbent for the clean-up because it is able to retain most of matrix co-extracted interferences. The final analytical method was successfully applied for multiresidue extraction at lower concentrations (for each analyte 10, 24 and 50 ng/g dry weight) with good recoveries and repeatability. These appealing results together with relative low detection limits suggest the suitability of the method to monitor the presence of these emerging contaminants in vegetables and fruits and their potential uptake.
Questo studio è incentrato sullo sviluppo di un metodo analitico multiresiduo per la determinazione degli ormoni steroidei (progestinici, estrogeni, androgeni e glucocorticoidi) in frutta e verdura. La presenza di inquinanti - soprattutto quelli endocrini come gli ormoni steroidei – in frutta e verdura è un argomento di grande preoccupazione, legato all'uso di acque reflue bonificate per l'irrigazione. Una volta nel suolo, l'assorbimento da parte delle piante, l'accumulo radicale o la traslocazione alle foglie e ai frutti possono verificarsi a seconda delle proprietà fisico-chimiche dei contaminanti (pKa e logKow). Gli analiti qui considerati presentano un ampio intervallo di logKow (da 1,46 del cortisone a 4,02 del 17β-estradiolo) quindi tendono ad accumularsi in modo diverso nelle piante (radici, foglie e frutti). Pertanto, come matrici sono stati scelti un tubero (carota), un ortaggio a foglia (spinaci) e un frutto (fragola). L'estrazione assistita da ultrasuoni (UAE) è stata scelta per i suoi brevi tempi di estrazione, il ridotto consumo di solvente organico, l'energia e l’economicità. La procedura di estrazione è stata studiata in matrici liofilizzate, elaborando i campioni prima (matrici bianche) e dopo l'aggiunta di spiking (per ogni analita 100 ng/g di peso secco). In primo luogo, è stato valutato il ruolo di diversi solventi di estrazione, ovvero Acetone (ACE), Metanolo (MeOH), Acetonitrile (ACN), miscela MeOH: ACN (75:25, v/v) e miscela MeOH: ACN (50:50, v/v), eseguendo una singola estrazione (5 mL di solvente, 5 min, r.t.). L'estratto è stato evaporato a secco sotto flusso di N2, ricostituito in 1 mL di MeOH e analizzato mediante HPLC-MS/MS. Le migliori prestazioni complessive in termini di recupero (%R) ed effetto matrice (%ME) sono state ottenute con MeOH come dimostrato dal test ANOVA. L'efficienza dell'estrazione è stata quindi ottimizzata con un approccio chemiometrico per sviluppare un metodo multiresiduo. È stato costruito un disegno fattoriale 23 considerando tre variabili a due livelli: numero di cicli (1 e 3); volume del solvente (2 e 5 mL); tempo di sonicazione (1 e 10 min). Ognuno degli otto esperimenti è stato replicato tre volte, una volta per ogni matrice. Il recupero multi-analita è stato selezionato come risposta sperimentale per ogni esperimento. L'elaborazione (software CAT basato su R, Chemometric Agile Tool) ha mostrato che il numero di cicli è significativo con p <0,01 (**) come l'interazione tra numero di cicli e volume del solvente. Di conseguenza, l'estrazione ottimizzata è composta da 3 x 1 min di cicli ad ultrasuoni con un piccolo volume di solvente organico (2 mL MeOH). Per ridurre gli effetti della matrice derivanti dai componenti residui della matrice, è stato studiato un ulteriore passaggio di pulizia che comporta diversi materiali sorbenti: Activated Chemviron Carbon (d-SPE), Florisil (SPE), Primary-Secondary Amine (SPE), C18 (SPE), SUPERCLEANTM LC-NH2 (SPE). Quest'ultima è stata selezionata come sorbente SPE per la pulizia, ridultando in grado di trattenere la maggior parte delle interferenze coestratte dalla matrice. Il metodo analitico finale è stato applicato con successo per l'estrazione multiresidua a concentrazioni inferiori (per ogni analita 10, 24 e 50 ng/g di peso secco) con buoni recuperi e ripetibilità. Questi risultati, insieme ai limiti di rilevabilità relativamente bassi, suggeriscono l'idoneità del metodo per monitorare la presenza di questi contaminanti emergenti nella verdura e nella frutta e il loro potenziale assorbimento.
SVILUPPO DI UN METODO ANALITICO PER L’ESTRAZIONE E L’ANALISI MULTIRESIDUALE DI ORMONI DA FRUTTA E VERDURA
SUPPINI, SARA
2020/2021
Abstract
This study is focused on the development of a multiresidue analytical method for determination of steroid hormones (progestins, oestrogens, androgens and glucocorticoids) in vegetables and fruits. Pollutants’ presence -especially those with endocrine disrupting properties as steroid hormones- in edible plants is a topic of high concern, related to the use of reclaimed wastewaters for irrigation. Once in the soil, plant uptake, root accumulation or translocation to the leaves and fruits can occur depending on the physicochemical properties of the contaminants (pKa and logKow). The herein considered analytes present a wide logKow range, from 1.46 of Cortisone to 4.02 of 17β-estradiol, thus they tend to accumulate differently in edible parts (e.g. roots, leaves and fruits). Therefore, a root vegetable (carrot), a leafy vegetable (spinach), and a fruit (strawberry) were chosen as probes for this study. Ultrasound-assisted extraction (UAE) has been selected due to its short extraction times, reduced organic solvent consumption, energy and costs saved. The extraction procedure was studied in freeze-dried matrices, processing samples before (blank matrices) and after spiking (for each analyte 100 ng/g dry weight). Firstly, the role of different extraction solvents, namely Acetone (ACE), Methanol (MeOH), Acetonitrile (ACN), mixture MeOH: ACN (75:25, v/v) and mixture MeOH: ACN (50:50, v/v), was evaluated. A single extraction was performed (5 mL of solvent, 5 min, r.t.). The extract was evaporated to dryness under N2 stream, reconstituted in 1 mL of MeOH, and analysed by HPLC–MS/MS. The best overall performances in terms of recovery (%R) and matrix effect (%ME) were achieved with MeOH as demonstrated by ANOVA test. The extraction efficiency was then optimized by a chemometric approach to develop a multiresidue method. A 23 factorial design was constructed considering three variables at two levels: number of cycles (1 and 3); solvent volume (2 and 5 mL); sonication time (1 and 10 min). Each one of the eight experiments was replicated three times, once for each matrix. The multi-analyte recovery was selected as the experimental response for each experiment. The elaboration (R-based software CAT, Chemometric Agile Tool) showed that the number of cycles is significant with p <0.01 (**) like the interaction between number of cycles and solvent volume. Accordingly, the optimized extraction is comprised of 3 x 1 min ultrasonic cycles with a small volume of organic solvent (2 mL MeOH). To reduce matrix effects resulting from co-eluting residual matrix components, an additional clean-up step was studied entailing different sorbent materials and approaches: Activated Chemviron Carbon (d-SPE), Florisil (SPE), Primary-Secondary Amine (SPE), C18 (SPE), SUPERCLEANTM LC-NH2 (SPE). The latter was selected as SPE sorbent for the clean-up because it is able to retain most of matrix co-extracted interferences. The final analytical method was successfully applied for multiresidue extraction at lower concentrations (for each analyte 10, 24 and 50 ng/g dry weight) with good recoveries and repeatability. These appealing results together with relative low detection limits suggest the suitability of the method to monitor the presence of these emerging contaminants in vegetables and fruits and their potential uptake.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/13245