The constant need for biomedical research to stay updated is essential for advancing the diagnosis, treatment, and prevention of diseases. This thesis proposes two different biochemical approaches to solving biomedical problems. In Part A, the evolutionary algorithm BINDesignER/PARCE is used to generate VHHs optimized for recognizing new therapeutic targets. This section includes a case study in which the specificity of a nanobody is redirected to a completely different target compared to the original. Specifically, the antigen-recognition regions are mutated to create VHHs optimized to recognize the active site of Hen Egg-White Lysozyme. The two VHHs with the highest affinity for the new target, with binding constants of 60-80 nM, are able to bind their epitope only at pH conditions that ensure the deprotonation of a histidine residue at the antigen-binding site. To confirm that the enzyme active site is involved in the interaction with the nanobodies, successful inhibition assays of the target enzyme were conducted using the optimized VHHs. Subsequently, the PARCE algorithm was applied to generate VHHs capable of distinguishing between two very similar forms of HER2, a well-known tumor marker for breast cancer, but with a significant functional difference: P105, used as a breast cancer biomarker, and P100, which appears to inhibit tumor growth. This work presents the preliminary results necessary for characterizing the in silico-designed VHHs, specifically focusing on the production of functional proteins. To perform binding assays, a peptide library was synthesized to mimic the C-terminal sequence of P105. Part B of this work focuses on characterizing a library of potential inhibitors of the protease of HIV-1, the virus responsible for AIDS. Despite forty years of research, the high mutation rate of the virus and the development of drug resistance drive the need for new compounds as potential drugs. Inhibition tests against the "wild-type" form of HIV-1 protease identified a lead compound in the ALES library with an IC50 comparable to that of Darunavir. Using a structural approach, crystals of this inhibitor were produced in complex with various drug-resistant HIV-1 protease mutants. Structural analysis revealed the presence of the inhibitor in the active site pocket of all these proteins, with full occupancy and in the same orientation. Crystals of one mutant were also obtained in complex with another inhibitor from the same library.
La costante necessità della ricerca biomedica di aggiornarsi è fondamentale per garantire il progresso nella diagnosi, nel trattamento e nella prevenzione di malattie. In questa tesi sono proposti due differenti approcci biochimici per la risoluzione di problemi di ambito biomedico. Nella parte A l’algoritmo evolutivo PARCE viene utilizzato per generare VHH ottimizzati per il riconoscimento di nuovi target terapeutici. Questa sezione si articola nell’analisi di un caso studio, in cui la specificità di un VHH viene dirottata verso un target differente rispetto a quello originale. In particolare, le regioni di riconoscimento dell’antigene vengono mutate fino a generare VHH ottimizzati per il riconoscimento del sito attivo del lisozima del bianco d’uovo di gallina. I due VHH con più alta affinità per il nuovo bersaglio, con costanti di legame 60-80 nM, sono in grado di legare il proprio epitopo solo in condizioni di pH che garantiscono la deprotonazione di un residuo istidinico sul sito di legame per l’antigene. Per confermare che il sito attivo dell’enzima è coinvolto nell’interazione con i nanobody, sono stati condotti con successo saggi di inibizione dell’enzima bersaglio utilizzando i VHH ottimizzati. Successivamente, l’algoritmo PARCE è stato applicato alla generazione di VHH in grado di distinguere due fome estremamente simili di HER2, un noto marker tumorale per il cancro al seno, ma con una significativa differenza funzionale: P105, utilizzata come biomacatore del cancro al seno, e P100, che invece sembra coinvolto nell’inibizione la crescita tumorale. In questo lavoro, vengono mostrati i risultati delle fasi preliminari necessarie alla caratterizzazione dei VHH disegnati in silico, ovvero quelle relative alla produzione di proteine funzionali. Al fine di effettuare saggi di binding, è stata sintetizzata una libreria di peptidi in grado di mimare la sequenza dell’estremita C-terminale di P105. La parte B di questo lavoro si concentra sulla caratterizzazione di una libreria di potenziali inibitori della proteasi di HIV-1, virus responsabile dell’AIDS. Nonostante quarant’anni di ricerca sul tema, l’alta frequenza di mutazione del virus e lo sviluppo di resistenze portano alla ricerca di composti sempre nuovi da impiegare come potenziali farmaci. Test condotti contro la forma “wild-type” dell’HIV-1 proteasi hanno permesso di individuare nella libreria ALES un lead compoud, che mostra un IC50 paragonabile a quello del Darunavir. Tramite approccio strutturale, è stato possibile produrre cristalli di questo inibitore in complesso con diversi mutanti farmacoresistenti di HIV-1 proteasi. L’analisi delle strutture ha permesso di individuare l’inibitore nella tasca del sito attivo di tutte queste proteine, a piena occupazione e sempre nello stesso orientamento. Cristalli di uno dei mutanti sono stati ottenuti anche in complesso con un altro inibitore della stessa libreria.
Biochemical Approaches in Medical Research: Computationally Designed Nanobodies and Novel Inhibitors for HIV-1 Protease
NOLASCO, FEDERICO
2025
Abstract
The constant need for biomedical research to stay updated is essential for advancing the diagnosis, treatment, and prevention of diseases. This thesis proposes two different biochemical approaches to solving biomedical problems. In Part A, the evolutionary algorithm BINDesignER/PARCE is used to generate VHHs optimized for recognizing new therapeutic targets. This section includes a case study in which the specificity of a nanobody is redirected to a completely different target compared to the original. Specifically, the antigen-recognition regions are mutated to create VHHs optimized to recognize the active site of Hen Egg-White Lysozyme. The two VHHs with the highest affinity for the new target, with binding constants of 60-80 nM, are able to bind their epitope only at pH conditions that ensure the deprotonation of a histidine residue at the antigen-binding site. To confirm that the enzyme active site is involved in the interaction with the nanobodies, successful inhibition assays of the target enzyme were conducted using the optimized VHHs. Subsequently, the PARCE algorithm was applied to generate VHHs capable of distinguishing between two very similar forms of HER2, a well-known tumor marker for breast cancer, but with a significant functional difference: P105, used as a breast cancer biomarker, and P100, which appears to inhibit tumor growth. This work presents the preliminary results necessary for characterizing the in silico-designed VHHs, specifically focusing on the production of functional proteins. To perform binding assays, a peptide library was synthesized to mimic the C-terminal sequence of P105. Part B of this work focuses on characterizing a library of potential inhibitors of the protease of HIV-1, the virus responsible for AIDS. Despite forty years of research, the high mutation rate of the virus and the development of drug resistance drive the need for new compounds as potential drugs. Inhibition tests against the "wild-type" form of HIV-1 protease identified a lead compound in the ALES library with an IC50 comparable to that of Darunavir. Using a structural approach, crystals of this inhibitor were produced in complex with various drug-resistant HIV-1 protease mutants. Structural analysis revealed the presence of the inhibitor in the active site pocket of all these proteins, with full occupancy and in the same orientation. Crystals of one mutant were also obtained in complex with another inhibitor from the same library.File | Dimensione | Formato | |
---|---|---|---|
TESI DEFINITIVA_FN_PDFA.pdf
embargo fino al 25/02/2026
Dimensione
9.35 MB
Formato
Adobe PDF
|
9.35 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/193381
URN:NBN:IT:UNITS-193381