Partial genome content within rAAVs impacts performance in a cell assay-dependent manner.

Recombinant adeno-associated viruses (rAAVs) deliver DNA to numerous cell types. However, packaging of partial genomes into the rAAV capsid is of concern. Although empty rAAV capsids are studied, there is little information regarding the impact of partial DNA content on rAAV performance in controlled studies. To address this, we tested vectors containing varying levels of partial, self-complementary EGFP genomes. Density gradient cesium chloride ultracentrifugation was used to isolate three distinct rAAV populations: (1) a lighter fraction, (2) a moderate fraction, and (3) a heavy fraction. Alkaline gels, Illumina Mi-Seq, size exclusion chromatography with multi-angle light scattering (SEC-MALS), and charge detection mass spectrometry (CD-MS) were used to characterize the genome of each population and ddPCR to quantify residual DNA molecules. Live-cell imaging and EGFP ELISA assays demonstrated reduced expression following transduction with the light fraction compared with the moderate and heavy fractions. However, PCR-based assays showed that the light density delivered EGFP DNA to cells as efficiently as the moderate and heavy fractions. Mi-Seq data revealed an underrepresentation of the promoter region for EGFP, suggesting that expression of EGFP was reduced because of lack of regulatory control. This work demonstrates that rAAVs containing partial genomes contribute to the DNA signal but have reduced vector performance.

Application of Size Exclusion Chromatography with Multiangle Light Scattering in the Analytical Development of a Preclinical Stage Gene Therapy Program.

To provide safe recombinant adeno-associated viruses (rAAV) to patients, scalable manufacturing processes are required. However, these processes may introduce impurities that impact the performance and quality of the final drug product. Empty rAAV capsids are product-related impurities. Regulatory guidance requires that accurate analytical methods be implemented early in product development to measure the level of empty capsids. A process confirmation vector, produced from 200 L production, was used to develop and optimize a size exclusion chromatography with UV and multiangle light scattering (SEC-MALS) method. Vector produced from a 500 L production was used to assess the full-to-empty ratio using the following analytical methods: sedimentation velocity analytical ultracentrifugation (SV-AUC), droplet digital PCR (ddPCR) with capsid enzyme-linked immunosorbent assay (ELISA), bulk absorbance at 260/280 nm, cryogenic electron microscopy, and SEC-MALS. This test article was used for a 30-day, non-Good Laboratory Practices animal study that assessed biodistribution of the product (STRX-330). SEC-MALS outperformed the other methods and correlated well with SV-AUC values of full-to-empty particles. In addition, SEC-MALS agreed with ddPCR and ELISA measurements for vector genomes/mL and capsid particles/mL, respectively. SEC-MALS was linear, accurate, and precise while achieving chromatography quality control (QC) recommendations. Compared to other stability-indicating assays, SEC-MALS performed similarly to ddPCR, capsid ELISA, and infectivity assays in accelerated stress studies. In response to alkaline, but not acidic stress, SEC-MALS indicated distinct changes in the DNA content of the monomer Adeno-associated viruses (AAV) peak for STRX-330, which was supported by ddPCR data. Conversely, acidic treatment resulted in more aggregated vector, but did not impact the DNA content. This work indicates that SEC-MALS is a valuable analytical tool in the analytical development and QC testing of AAV. In addition, this work suggests that SEC-MALS can provide fundamental understanding of AAV in response to environmental stress. This may impact steps of the manufacturing process to minimize conditions that reduce performance.

Turn-taking: a case study of early gesture and word use in answering WHERE and WHICH questions.

When young children answer questions, they do so more slowly than adults and appear to have difficulty finding the appropriate words. Because children leave gaps before they respond, it is possible that they could answer faster with gestures than with words. In this study, we compare gestural and verbal responses from one child between the ages of 1;4 and 3;5, to adult Where and Which questions, which can be answered with gestures and/or words. After extracting all adult Where and Which questions and child answers from longitudinal videotaped sessions, we examined the timing from the end of each question to the start of the response, and compared the timing for gestures and words. Child responses could take the form of a gesture or word(s); the latter could be words repeated from the adult question or new words retrieved by the child. Or responses could be complex: a gesture + word repeat, gesture + new word, or word repeat + new word. Gestures were the fastest overall, followed successively by word-repeats, then new-word responses. This ordering, with gestures ahead of words, suggests that the child knows what to answer but needs more time to retrieve any relevant words. In short, word retrieval and articulation appear to be bottlenecks in the timing of responses: both add to the planning required in answering a question.