GMP AAV | PackGene Biotech

Differences in AAV Administration Routes in Clinical Applications for Neurological Diseases

Gene therapy has been a focal point in the biopharmaceutical industry over the past five years. Initially considered for the treatment of rare genetic diseases, gene therapy has gradually been applied to neurological diseases such as Alzheimer’s and Parkinson’s due to ongoing advancements in gene therapy and neurological research. However, there are still technical bottlenecks that need to be overcome in terms of safety and technology.

Origin and Biological Characteristics of AAV

The concept of gene therapy was first clearly proposed in 1947 (Keeler, 1947). Two factors prompted Clyde Keeler to distinguish the gene therapy model from other treatment strategies: its ability to fundamentally address the causes of diseases and its potential to induce “permanent correction” of genes. These two factors remain the main reasons researchers pursue gene therapy. Neurological diseases may benefit more from gene therapy’s unique advantages than other types of diseases. For monogenic diseases, gene therapy has shown clinical effectiveness, and it has now expanded to treat common complex diseases through various genetic interventions. Long-term expression efficacy of gene therapy is particularly beneficial for clinical applications involving the spinal cord, brain, and peripheral sensory organ systems, as most drugs cannot cross the blood-brain barrier (BBB) to reach these organs. For organs or tissues like the brain, cerebrospinal fluid, retina, or cochlea, traditional drugs with short half-lives requiring frequent injections are clinically prohibited. Gene therapy, requiring only a single injection, can achieve long-lasting effects, thus providing a safe and convenient treatment method.

Fig-1-AAV-Virus-and-Genomic-Vector

Fig 1: AAV Virus and Genomic Vector

Adeno-associated virus (AAV) is a member of the Parvoviridae family, characterized by its non-enveloped icosahedral capsid carrying a 4.7 kb single-stranded (ss) genome (Figure 1). Initially, it was believed that AAV needed “help” from co-infection with helper viruses like adenovirus or herpesvirus to complete its replication cycle. The AAV genome is flanked by inverted terminal repeats (ITRs), encoding three gene families: replication-related genes (Rep), structural capsid genes (Cap VP1, -2, and -3), and a small virus-associated assembly-activating protein (AAP) necessary for assembly (Figure 1A) (Knipe and Howley, 2013).

AAV has been naturally isolated from multiple species, including non-human primates (NHP), cows, goats, mice, and even birds (Gao et al., 2005). Wild-type AAV can integrate into the human AAVS1 19q3 gene site (Kotin et al., 1990; Linden et al., 1996). This relative specificity is due to the sequence homology of the ITR regions and the integration of Rep gene products into the host genome (Linden et al., 1996). Recombinant AAV is produced by removing all open reading frames (orf) from the viral genome, making it a replication-defective viral vector (Figure 1B). To produce recombinant AAV, the 145 bp ITR sequences need to be retained (Hastie and Samulski, 2015). The absence of viral genomes in the vector prevents replication, significantly reducing safety concerns while freeing up gene loading capacity for therapeutic purposes.

The most common vector production method is triple-plasmid transient transfection of HEK293 cells, involving one plasmid with the ITR-flanked cis transgene, another plasmid encoding AAV rep and cap genes, and a set of minimal adenovirus genes that assist AAV replication.

**Selection of AAV Administration Routes in Clinical Applications for Neurological Diseases**

The choice of AAV administration routes is crucial for the safety and efficacy of gene transfer. Transferring genes to the central nervous system or isolated sensory organs, such as the eyes and ears, poses significant challenges. Correct selection of vectors and delivery routes can enhance cell infection efficiency while ensuring drug safety (Figure 2). For example, the physiological barriers, such as the BBB, greatly limit AAV drug entry into the brain, eyes, cochlea, and spinal cord.

For treating diseases related to the nervous system or sensory organs, local injection methods (e.g., to the eyes or cochlea) are preferred because the targeted areas are relatively small. Although delivering drugs to cerebrospinal fluid (intraventricular and intrathecal routes) or blood may be more beneficial for treating multifocal diseases, the most commonly used method to deliver therapeutic genes to the brain remains intracerebral parenchymal infusion of AAV. Intranasal administration is another non-invasive method that can be used for treatments aimed at restoring lysosomal enzyme expression levels, as lysosomal enzymes can diffuse through the central nervous system. Finally, intramuscular or spinal injections of AAV are generally chosen to improve the treatment of motor neuron-related diseases.

Fig-2-Pathways-of-AAV-Delivery-to-the-Nervous-System-in-Vivo

Fig 2: Pathways of AAV Delivery to the Nervous System in Vivo

Local administration of drugs has significant advantages compared to systemic injections into the venous system or other fluid-filled compartments. It maximizes the concentration and retention time of gene transfer drugs near target cells, limiting widespread biological distribution and thereby reducing the risk of immunogenicity and toxicity from AAV components or transgene ectopic expression.

Intracerebral Stereotactic Injection

Local delivery to the central nervous system is complex, involving surgical procedures where subjects are anesthetized and restricted in a stereotactic frame while a needle or flexible quartz catheter is directly inserted into the parenchyma through a cranial drill hole. In small animal models (mice and rats), AAV is injected directly, usually with stereotactic coordinates calculated from reference points on the skull (bregma) based on brain atlases. Real-time MRI-guided systems can improve accuracy in larger species (e.g., NHPs and humans) (Eberling et al., 2008; Fiandaca et al., 2009; Hadaczek et al., 2006). Most AAV serotypes exhibit good neuron affinity, e.g., AAV5 effectively targets astrocytes (Davidson et al., 2000), while capsid-modified AAV6 can transduce microglia (Rosario et al., 2016). Preclinical studies in mouse models of lysosomal storage diseases (LSD), Alzheimer’s disease (AD), Parkinson’s disease (PD), and Huntington’s disease (HD) have successfully used AAV for lung injection (Hocquemiller et al., 2016; Choudhury et al., 2017). In small animal models, only a few microliters of vector are needed for extensive transduction. Studies in larger species also show significant safety and higher transduction efficiency, such as in cat models of GM1 and GM2 gangliosidosis (using AAV1 or AAVh.8; McCurdy et al., 2015), dog models of mucopolysaccharidosis type I and IIIB (using AAV5; Ellinwood et al., 2011), and MPTP-induced NHP models simulating Parkinsonian neurodegeneration (observing sustained clinical improvement 8 years after AAV2-hAADC injection; Hadaczek et al., 2010). Safety and transduction efficacy studies following intracerebral AAV injection in healthy NHPs also demonstrate good tolerance (Colle et al., 2010; McBride et al., 2011; Zerah et al., 2015, Ciron et al., 2009; Sondhi et al., 2012). For motor neuron diseases, multi-level spinal parenchymal injections have been successfully applied in ALS mouse models (Azzouz et al., 2000; Franz et al., 2009; Hardcastle et al., 2018).

High-risk surgical interventions and resulting focal lesions from transgene expression make stereotactic AAV brain injections highly challenging for large mammals and humans. Besides inherent risks of viral or bacterial infections, bleeding, and edema from neurosurgical interventions, key factors associated with stereotactic AAV injection include limited vector spread within the parenchyma. Some serotypes (e.g., AAV1, -6, -8, and -9) exhibit long-distance anterograde and retrograde axonal transport, potentially facilitating viral particle spread within brain anatomical connection regions (Castle et al., 2014; Lo¨w et al., 2013; San Sebastian et al., 2013; Zingg et al., 2017). Convection-enhanced delivery (CED) is another injection option, applying pressure during infusion to enhance vector diffusion (Bankiewicz et al., 2000; Hadaczek et al., 2006). Other methods to enhance AAV transduction after local brain delivery include co-infusion of factors like heparin (Mastakov et al., 2002) or glycol (Mastakov et al., 2001) (Carty et al., 2010).

Intrathecal Injection

Compared to the ventricular route, lumbar-level intrathecal (IT) injection is more feasible clinically. Studies have shown that a single IT injection of AAV.hGAA significantly improves neurological and cardiac functions in Pompe disease (Hordeaux et al., 2017). When comparing cerebrospinal fluid delivery routes in different species, transduction is more effective in adult mice (Bey et al., 2017), while lumbar puncture IT administration yields stronger transduction in the brain and spinal cord of cynomolgus monkeys or dogs (Hinderer et al., 2018a).

Intranasal Injection

Local administration can also be achieved through intranasal delivery of AAV for brain transduction. Although primarily targeting respiratory-related diseases, this non-invasive

method is clinically feasible and safe (Al-Qtaibi et al., 2022). In animal models, nasal administration of AAV9 encoding the lysosomal enzyme tripeptidyl peptidase 1 (TPP1) demonstrates significant improvement in lifespan, behavior, and CNS tissue biomarkers in CLN2 disease mice (Tardif et al., 2019).

Summary

AAV has been confirmed to be safe for clinical use, but current research on different administration routes remains limited to small animal models and a few clinical trials. Future research should focus on comparing the effects of different administration routes in larger animal models or human studies, which may promote the development of gene therapy for neurological diseases.

aav-applying-in-CNS-disease-clinical-trial

References
1. Azzouz, M., et al. (2000). Multicistronic lentiviral vector-mediated striatal gene transfer of aromatic l-amino acid decarboxylase, tyrosine hydroxylase, and GTP cyclohydrolase I in a rat model of Parkinson’s disease. *Journal of Neuroscience Research, 60*(2), 142-151.
2. Bankiewicz, K. S., et al. (2000). Convection-enhanced delivery of AAV vector in Parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function using pro-drug approach. *Experimental Neurology, 164*(1), 2-14.
3. Choudhury, S. R., et al. (2017). In Vivo Selection Yields AAV-B1 Capsid for Central Nervous System and Muscle Gene Therapy. *Molecular Therapy, 25*(10), 2503-2513.
4. Hordeaux, J., et al. (2017). Intrathecal AAV Injection of AAV9 in the Human Cervical Spinal Cord: Phase 1/2 Clinical Trial in Giant Axonal Neuropathy. *Molecular Therapy, 25*(2), 379-391.
5. Tardif, C., et al. (2019). Intranasal delivery of AAV9 in mice to CNS-targeted tripeptidyl peptidase 1, extending lifespan and improving behavior and CNS tissue biomarkers in CLN2 disease. *Journal of Neuroscience Research, 97*(11), 1435-1451.

Innovations in AAV Production: Leveraging High-Yield Cell Lines, novel RC plasmid, and Dual-Plasmid Systems

Abstract

PackGene has pioneered innovative approaches to meet the escalating demand for high-quality adeno-associated virus (AAV) vectors across the dynamic landscape of gene therapy. For example, PackGene achieves remarkable enhancements in AAV packaging efficiency with a 3 to 8-fold increase in AAV production yield across various serotypes through their proprietary π-Alpha 293 AAV High-Yield Platform. The process began with the development of the PCS3.0 suspension cell line, which exhibits substantial improvements in production yield compared to its predecessors. PackGene meticulously engineered the PCS3.0 cell line using cutting-edge technologies to optimize for unparalleled productivity, stability, and quality. This PCS3.0 suspension cell line was then coupled with comprehensive process optimization and the introduction of a novel dual-plasmid system to further enhance AAV productivity and quality while minimizing the risk of contamination. Overall, PackGene’s commitment to innovation has revolutionized AAV production by driving transformative advancements in gene therapy.

Introduction

π-Alpha 293 AAV High-Yield Platform

PackGene’s proprietary π-Alpha 293 AAV High-yield Platform uses a combination of technological advancements to increase AAV production yield and quality. First, the platform incorporates PackGene’s PCS3.0 suspension cell line and uniquely designed RC plasmid in the triple-plasmid transfection system to increase production yield 3 to 8 times for various AAV serotypes. These technologies are coupled with both in-process upstream and downstream QbD optimizations to increase total AAV yield up to 10-fold. A single batch of AAV production delivers up to 1E+17vg virus particles, which is enough to meet the needs of most clinical and commercial level of AAV production requirements.

Advantages of the Platform

Productivity: Proprietary technologies elevate AAV yield in HEK293 serum-free cell suspension systems by over 10-fold.

Quality: Unique process development procedures significantly diminish key impurities such as Host Cell DNA (HCD) and endotoxin.

Viability: Key technological innovations reduce the rate of empty capsids and enhance infection titers.

High-Yield AAV Production Cell Line: From PCS 2.0 to PCS 3.0

PackGene’s pursuit of high-yield AAV production began with the development of an optimized single clonal suspension cell line. Derived from the ATCC HEK293 cells (Grieger JC et al.), PackGene’s PCS 3.0 monoclonal cell line yields impressive results. With Production yield enhancements of PCA (adherent cell line) up to 7x titer and PCS 2.0 (suspension cell line) up to 4x titer.

PackGene recognized the crucial role of cell substrates in AAV manufacturing, and thus began the process of optimizing AAV production by generating a cell line that is specifically engineered for high efficiency AAV production. To this end, PackGene designed a stage-wise cell engineering protocol to identify cells that are optimized for AAV production. The PCS2.0 monoclonal cell line was chosen as the initial cell line due to its suspension growth capability, high efficiency in producing AAV, and widespread use in AAV production. Multiple highly uniform monoclonal cells lines were generated from PCS2.0 cultures using Dispencell in-situ cell dispensing and a CloneSelect Imager for clonality verification. Each generated cell line was then screened for AAV production yield and Population Doubling Time (PDT) in order to identify the most efficient clone (Figure 1). The most efficient clone was then selected and further refined through gradient-decreasing serum cultures and two rounds of single-cell sorting (Figure 2). The resultant PCS3.0 monoclonal cell line is a robust suspension cell line with unparalleled AAV productivity, stability, and quality.

PCA-genomic-titer

PCS-genomic-titer

Figure 1. PackGene’s monoclonal cell line shows improved AAV production yield

Overview-of-PCS3.0-cell-line-development

Figure 2. Overview of PCS3.0 cell line development

PackGene conducted several tests to compare the performance of the PCS3.0 cell line with that of the widely used VPC2.0 cell line for AAV production, focusing on the ssAAV9 serotype. A standard suspension growth protocol in a 125ml shake flask was employed for AAV production with the VPC2.0 cell line. In contrast, testing the PCS3.0 cell line involved multiple growth apparatus, including a 125ml shaker flask, a 3L bioreactor, and an ambr250 bioreactor. Furthermore, PackGene evaluated the efficiency of PCS3.0 in the ambr250 bioreactor using four cis-plasmids, each containing a distinct gene of interest (GOI).

The results of these studies indicated comparable AAV yields between VPC2.0 and PCS3.0 (Table 1, Figure 3A). However, PCS3.0 exhibited several advantages over VPC2.0. Notably, PCS3.0 demonstrated significantly lower levels of residual host cell DNA (HCD) and residual plasmid DNA (pDNA) compared to VPC2.0. Encapsulated pDNA levels were found to be reduced up to threefold ( 50~60 ng/E+12vg). Additionally, under the 3L growth condition, PackGene observed that PCS3.0 displayed reduced glucose consumption and slower accumulation of metabolic waste (lactate).

Figure-3-table

Table 1. VPC2.0 and PCS3.0 comparison parameters and results.

Figure-3.-PCS3.0-cell-line-in-a-ssAAV9-gene-therapy-project-3

Figure 3. PCS3.0 cell line in a ssAAV9 gene therapy project

𝝅-Alpha 293 High-yield AAV Production Platform Process Optimization

In addition to selecting the idea cell line, PackGene recognized that optimization of Critical Process Parameters (CPP) can substantially impact AAV production efficiency and quality. PackGene performed parametric optimization analysis across (1) transfections conditions (2) engineering parameters and (3) lysis and harvest conditions (Table 2). Combined these CPP optimizations ensure maximal AAV production efficiency.

Critical Process Parameters
Transfection Conditions	Engineering Parameters	Harvest conditions
Cell density	Stir speed	Stir speed
Total plasmid amount	pH	Cell density
Plasmid ratio	Dissolved oxygen (dO2)	Time
DNA/PEI ratio	Temperature	Temperature

Table 2. Optimized Critical Process Parameters

With Critical Process Parameters optimized, PackGene’s suspension production process demonstrates robust amplification across varying scales with an overall recovery rate that exceeds 30%. For example, scaling up to 200L AAV production volume with stable genomic titer and residual hcDNA resulted in 1.0E17 vg at harvest. Critically, Key Product Quality Attributes, including significant full capsid enrichment, were maintained in the 76-92% range during scale-up. Thus, up-scaling production using PackGene’s optimized 𝝅-Alpha 293 High-yield AAV Production Platform meet even stringent GMP production requirements.

Figure-4.-Total-AAV-Yields-and-Scalability-During-Process-Optimization-Of-CPP

Figure 4. Total AAV Yields and Scalability During Process Optimization Of CPP

Uniquely Designed RC Plasmid Increases AAV Yield by 3-8 Times

PackGene designed and validated a proprietary non-coding regulatory element with the goal of enhancing AAV yield. Incorporation of this regulatory element into a novel Rep-Cap plasmid enhanced AAV yield in both adherent and suspension cell lines by 3-8 times across various serotypes (Figure 5). In addition, PackGene found that AAV production using their novel RC Plasmid reduced the observed plasmid DNA encapsulation ratio from 3% to between 0.1% and 0.2% (Data not shown).

novel-repcap-plasmid

Figure-5.-PackGenes-novel-RC-plasmid-significantly-increases-AAV-production-yield-1

Figure 5. PackGene’s novel RC plasmid significantly increases AAV production yield

Novel Dual-Plasmid System: Next-Generation AAV Production Strategy

The traditional triple-plasmid transfection system is a cornerstone in industrial AAV manufacturing due to its scalability. An alternative dual-plasmid system has emerged as a cost-saving and simplified procedure for AAV manufacturing (Matsushita T et al., Grimm D et al., Tang Q et al.). While there has been some adoption of the dual-plasmid approach, persistent challenges including low productivity, quality issues, and the risk of generating replication-competent AAV (rcAAV) necessitated further innovation.

PackGene sought to design and validate a dual-plasmid AAV production system that mitigates the challenges that are common to other dual-plasmid AAV production systems. Toward this end PackGene designed a novel TP-plasmid that integrates replication-competent (RC) genes, and helper functions into a single plasmid. Additional experiments were performed to integrate this novel TP-plasmid with PackGene’s 𝝅-Alpha 293 High-yield AAV Production Platform by parametrically testing transfection reagents. The resultant TP011-dual plasmid AAV production system was tested against traditional triple plasmid AAV production methods and was found to show increases yield and reductions in host cell DNA (HCD) residue. This result was consistent across production scales, which positions PackGene’s TP011 dual-plasmid system as a superior AAV production platform.

Non-rcAAV design of packing plasmids for dual-plasmid system.

Triple-plasmid AAV production systems require three separate plasmids, one plasmid that holds the gene of interest (GOI), a second plasmid that contains replication-competent (RC) genes, and a third plasmid that contains helper genes required for AAV packaging. PackGene’s dual-plasmid system requires only two plasmids, one plasmid that holds the gene of interest (GOI) and a second TP plasmid that holds both replication-competent (RC) genes and helper genes. PackGene’s initial attempts to design a TP plasmid resulted in the TP007 plasmid. AAV generation with the TP007 plasmid was successful, however, rcAAV detection by Rep-2 specific qPRC showed less than ideal rcAAV generation rates (Figure 6). To mitigate this issue, TP007 was modified to include a proprietary non-coding regulatory element that was designed by PackGene (Figure 6A). Inclusion of this regulatory element in the new TP011 vector showed a substantial reduction in rcAAV generation (Figure 6B). AAV generation with a dual-plasmid system that shows low rcAAV generation levels marks a significant leap forward in AAV production technology.

Figure-6.-Non-rcAAV-design-of-packing-plasmids-for-dual-plasmid-system

Figure-6-table-1

Figure 6. Non-rcAAV design of packing plasmids for dual-plasmid system

Design of Experiments paired with the Dual-Plasmid system to drive high productivity

PackGene paired their innovative dual-plasmid system with their 𝝅-Alpha 293 High-yield AAV Production Platform Process Optimization to maximizes AAV yield and to ensures consistent and high-quality vector production. This alone enhances the efficacy and reliability of PackGene’s AAV production platform. Nevertheless, PackGene went several steps further to fully integrate these two innovations. Transfection reagents (TR) were screened for use within the dual-plasmid system (Figure 7A) and identified a small molecule compound that significantly enhanced AAV9 productivity on different TRs (Figure 7B).

Figure-7.-DoE-optimization-enabling-high-productivity1

Figure-7.-DoE-optimization-enabling-high-productivity-2

Figure 7. DoE optimization enabling high productivity

A-B. Different transfection reagent and effect of a small molecule on AAV9 productivity. C. Contour plots of a Box-Behnken design (BBD) DoE experiment on TP011 dual-plasmid system to optimize transfection viable cell density (VCD), plasmid DNA and TR amount. For AAV2-EA0218K, a platform GOI, DoE optimized dual-plasmid transfection process can harvest ＞5.0×1011 vg/mL AAV lysate; D. CPP designed space of VCD/DNA/TR can outcome a stable AAV2-EA0218K yield between 4.0~6.0 ×1011 vg/mL.

TP011 dual-plasmid system with higher yield and lower HCD residue.

PackGene tested the efficiency and quality of their TP011 dual-plasmid system by comparing AAV9 production titer and residual host cell DNA (HCD) residue to AAV generated using traditional triple-plasmid production methods. They found that AAV9 yield was more than 3x higher when using the TP011 dual-plasmid system when compared to the traditional triple plasmid methods (Table 3; Figure 8A). In addition, they found that residual HCD was more than 3x lower in TP011 dual-plasmid system samples relative to triple-plasmid methods while residual pDNA was comparable (Table 3; Figure 8B, C).

PackGene also sought to determine if improvements in AAV production titer were stable regardless of the GOI plasmid used for production. To test this they compared TP011 dual-plasmid system AAV9 production titer to titers of AAV generated by triple-plasmid production methods using 4 distinct GOI plasmids. For two of the GOI plasmids it was found that AAV9 generated by the TP011 dual-plasmid system showed elevated titer relative to triple-plasmid production methods. AAV9 titers were comparable between the two methods for the remaining two GOI plasmids (Figure 8D). This data serves as evidence that PackGene’s TP011 dual-plasmid system represents an advancement in AAV production technology by showing superior yield with reduced HCD residue. Thus, the TP011 dual-plasmid system streamlines the production process, resulting in heightened yield and purity.

Figure-8.-TP011-dual-system-with-higher-yield-and-lower-HCD-residue-table

Table 3. TP011 dual-plasmid system and traditional triple-plasmid method comparison parameters

Figure-8.-TP011-dual-system-with-higher-yield-and-lower-HCD-residue-1

Figure 8. TP011-dual system with higher yield and lower HCD residue.

Application of dual plasmid system in a therapeutic case study

PackGene has used their 𝝅-Alpha 293 High-yield AAV Production Platform with their TP011 dual-plasmid system to generate AAV for numerous labs across the world. Here we present one case study in which PackGene generated scAAV9 for a real-world application, with a comparison to the same AAV9 generated by traditional triple-plasmid methods (Table 5). PackGene found similar scAAV9 AAV lysate yield, residual HCD and pDNA impurities when their TP011 dual-plasmid system with the traditional triple-plasmid generation method (Table 5; Figure 9). Additionally, there was no significant difference in AAV empty capsid ratio the two approaches, and Analytical Ultracentrifuge (AUC) shows a relative smaller ssAAV peak and larger scAAV peak compared to triple-plasmid system, with empty AAV peak as low as 1.50%. The higher scAAV ratio indicates the therapeutic gene can express directly and more rapidly, with higher expression levels after entering the cell.

Figure-10.-Application-of-dual-system-in-a-gene-therapy-case-based-on-scAAV9-table

What are you looking for?

Differences in AAV Administration Routes in Clinical Applications for Neurological Diseases

Innovations in AAV Production: Leveraging High-Yield Cell Lines, novel RC plasmid, and Dual-Plasmid Systems

Download

Login

New User Registration

Reset Password

Google Account Binding