Recombinant adeno-associated viruses (rAAVs) are among the most promising vectors in the field of gene therapy, due to their high transfection efficiency and low immunogenicity. During their manufacturing, however, only a portion of the expressed viral capsids are efficiently filled with the gene of interest, leading to full/empty rAAVs mixtures. Since empty capsids do not contribute to the therapeutic action, while competing with the full ones for binding to the cell receptors, they need to be removed from the product formulation. Chromatography is the technique of choice for this purpose, as it is able to reach the purity levels required for a pharmaceutical product. Notably, anion exchange chromatography is selected, since the presence of genetic material in full capsids enhances their negative surface charge, with an isoelectric point of 5.9 compared to 6.3 in the case of empty ones. Nevertheless, the difference between the two species is very subtle and the design of a process which is able to guarantee their separation and recovery of full AAVs at high purity is challenging and cost-intensive, if based on traditional laboratory experiments. For this reason, the aim of this thesis is the development of a detailed model able to reliably describe the transport of AAVs in an anion exchange column and to exploit it for an in silico optimization of the process, ultimately reducing the laboratory burden, the costs and the time needed. The model relies on a general rate model, describing the radial distribution of AAVs in the resin particles as a result of the competition between adsorption, convection to the surface and intraparticle diffusion. The model was calibrated through dedicated experiments and its reliability was demonstrated by simulating experimental chromatography runs. Finally, an optimization procedure was developed. It led to the identification of optimal operating conditions, in terms of loading duration, gradient flowrate and corresponding duration that ensured that quality and safety specifications were respected, while maximizing productivity. Such parameters are duration of loading of 24 minutes, 15 column volumes of elution and elution flowrate of 0.4 mL/min, providing Productivity=1.73 1015 VPart/Lresin/h, Yield=57.3% and Purity=80.1%.
I virus adeno-associati ricombinanti (rAAVs) sono tra i vettori più promettenti nell’ambito della terapia genetica, a causa della loro elevata efficienza nella trasfezione e bassa immunogenicità. Durante la loro produzione, tuttavia, solo una porzione delle capsidi virali espresse sono riempite in modo efficace con il gene di interesse, dando origine a miscele di rAAVs pieni e vuoti. Poiché le capsidi vuote non contribuiscono all’azione terapeutica, ma competono con quelle piene nel legarsi ai recettori cellulari, esse devono essere rimosse. La cromatografia è la tecnica prescelta per quest’operazione, dal momento che è in grado di raggiungere i livelli di purezza richiesti per un prodotto farmaceutico. In particolare, la cromatografia a scambio anionico è prediletta, perché la presenza di materiale genetico nelle capsidi piene aumenta la loro carica negativa superficiale, con un punto isoelettrico di 5.9 rispetto a 6.3 nel caso di capsidi vuote. Ciononostante, la differenza tra le due specie è molto sottile e il design di un processo in grado di garantire la loro separazione e il recupero di AAVs pieni ad elevata purezza è complesso e costoso, se basato sui tradizionali esperimenti di laboratorio. Per questo motivo, l’obiettivo di questa tesi è lo sviluppo di un modello dettagliato, in grado di descrivere fedelmente il trasporto di AAVs in una colonna a scambio anionico, e il suo utilizzo per un’ottimizzazione in silico del processo, riducendo, in definitiva, il carico sul laboratorio, i costi e il tempo necessario, affidandosi a simulazioni. Il modello si basa su un general rate model, descrivendo la distribuzione radiale di AAVs nelle particelle della resina come conseguenza della competizione tra adsorbimento, convezione verso la superficie e diffusione intraparticellare. Il modello è stato calibrato attraverso esperimenti dedicati e la sua affidabilità è stata dimostrata simulando operazioni cromatografiche sperimentali. Infine, è stata sviluppata una procedura di ottimizzazione. Questa ha condotto all’identificazione delle condizioni operative ottimali, in termini di durata della fase di loading, portata volumetrica adottata nel gradiente e la corrispondente durata, che garantiscano che qualità e sicurezza siano rispettate, massimizzando la produttività. Tali parametri sono durata di loading pari a 24 minuti e 15 volumi di colonna di gradiente a 0.4 mL/min, che forniscono Produttività=1.73 1015 VPart/Lresin/h, Resa=57.3% and Purezza=80.1%.
Modelling and optimization of the chromatographic separation of full/empty adeno-associated virus mixtures
La Torre, Martina
2024/2025
Abstract
Recombinant adeno-associated viruses (rAAVs) are among the most promising vectors in the field of gene therapy, due to their high transfection efficiency and low immunogenicity. During their manufacturing, however, only a portion of the expressed viral capsids are efficiently filled with the gene of interest, leading to full/empty rAAVs mixtures. Since empty capsids do not contribute to the therapeutic action, while competing with the full ones for binding to the cell receptors, they need to be removed from the product formulation. Chromatography is the technique of choice for this purpose, as it is able to reach the purity levels required for a pharmaceutical product. Notably, anion exchange chromatography is selected, since the presence of genetic material in full capsids enhances their negative surface charge, with an isoelectric point of 5.9 compared to 6.3 in the case of empty ones. Nevertheless, the difference between the two species is very subtle and the design of a process which is able to guarantee their separation and recovery of full AAVs at high purity is challenging and cost-intensive, if based on traditional laboratory experiments. For this reason, the aim of this thesis is the development of a detailed model able to reliably describe the transport of AAVs in an anion exchange column and to exploit it for an in silico optimization of the process, ultimately reducing the laboratory burden, the costs and the time needed. The model relies on a general rate model, describing the radial distribution of AAVs in the resin particles as a result of the competition between adsorption, convection to the surface and intraparticle diffusion. The model was calibrated through dedicated experiments and its reliability was demonstrated by simulating experimental chromatography runs. Finally, an optimization procedure was developed. It led to the identification of optimal operating conditions, in terms of loading duration, gradient flowrate and corresponding duration that ensured that quality and safety specifications were respected, while maximizing productivity. Such parameters are duration of loading of 24 minutes, 15 column volumes of elution and elution flowrate of 0.4 mL/min, providing Productivity=1.73 1015 VPart/Lresin/h, Yield=57.3% and Purity=80.1%.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_12_La Torre_Tesi.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
7.63 MB
Formato
Adobe PDF
|
7.63 MB | Adobe PDF | Visualizza/Apri |
|
2025_12_LaTorre_Executive Summary.pdf
non accessibile
Descrizione: Executive summary
Dimensione
1.38 MB
Formato
Adobe PDF
|
1.38 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/246417