Use of a three-dimensional humanized liver model for the study of viral gene vectors.

Reconstituted three-dimensional (3D) liver models obtained by engrafting hepatic cells into an extracellular matrix (ECM) are valuable tools to study tissue regeneration, drug action and toxicology ex vivo. The aim of the present study was to establish a system for the functional investigation of a viral vector in a 3D liver model composed of human HepG2 cells on a rat ECM. An adeno-associated viral (AAV) vector expressing the Emerald green fluorescent protein (EmGFP) and a short hairpin RNA (shRNA) directed against human cyclophilin b (hCycB) was injected into the portal vein of 3D liver models. Application of the vector did not exert toxic effects, as shown by analysis of metabolic parameters. Six days after transduction, fluorescence microscopy analysis of EmGFP production revealed widespread distribution of the AAV vectors. After optimization of the recellularization and transduction conditions, averages of 55 and 90 internalized vector genomes per cell in two replicates of the liver model were achieved, as determined by quantitative PCR analysis. Functionality of the AAV vector was confirmed by efficient shRNA-mediated knockdown of hCycB by 70-90%. Our study provides a proof-of-concept that a recellularized biological ECM provides a valuable model to study viral vectors ex vivo.

A novel method for the quantification of adeno-associated virus vectors for RNA interference applications using quantitative polymerase chain reaction and purified genomic adeno-associated virus DNA as a standard.

Recombinant adeno-associated virus (rAAV) vectors are promising tools in gene therapy, but accurate quantification of the vector dose remains a critical issue for their successful application. We therefore aimed at the precise determination of the titer of self-complementary AAV (scAAV) vectors to improve the reliability of RNA interference (RNAi)-mediated knockdown approaches. Vector titers were initially determined by quantitative polymerase chain reaction (qPCR) using four primer sets targeting different regions within the AAV vector genome (VG) and either coiled or linearized plasmid standards. Despite very low variability between replicates in each assay, these quantification experiments revealed up to 20-fold variation in vector titers. Therefore, we developed a novel approach for the reproducible determination of titers of scAAV vectors based on the use of purified genomic vector DNA as a standard (scAAVStd). Consistent results were obtained in qPCR assays using the four primer sets mentioned above. RNAi-mediated silencing of human cyclophilin B (hCycB) by short hairpin RNA-expressing scAAV vectors was investigated in HeLa cells using two independent vector preparations. We found that the required vector titers for efficient knockdown differed by a factor of 3.5 between both preparations. Hence, we also investigated the number of internalized scAAV vectors, termed transduction units (TUs). TUs were determined by qPCR applying the scAAVStd. Very similar values for 80% hCycB knockdown were obtained for the two AAV vector preparations. Thus, only the determination of TUs, rather than vector concentration, allows for reproducible results in functional analyses using AAV vectors.