Parvovirus Vectors: The Future of Gene Therapy

Abstract

The unique diversity of parvoviral vectors with innate antioncogenic properties, autonomous replication, ease of recombinant vector production and stable transgene expression in target cells makes them an attractive choice as viral vectors for gene therapy protocols. Amongst various parvoviruses that have been identified so far, recombinant vectors originating from adeno-associated virus, minute virus of mice (MVM), LuIII and parvovirus H1 have shown promising results in many preclinical models of human diseases including cancer. The adeno-associated virus (AAV), a non-pathogenic human parvovirus, has gained attention as a potentially useful vector. The improved understanding of the metabolism of vector genomes and the mechanism of transduction by AAV vectors is leading to advancement in the development of more sophisticated AAV vectors. The in-depth studies of AAV vector biology is opening avenues for more robust design of AAV vectors that have potentially increased transduction efficiency, increased specificity in cellular targeting, and an increased payload capacity. This chapter gives an overview of the application of autonomous parvoviral vectors and AAV vectors, based on our current understanding of viral biology and the state of the platform.

1. Introduction

Parvoviruses are among the smallest of eukaryotic viruses. They are subdivided into three major groups namely densoviruses, autonomous parvoviruses (APV), and dependoviruses [1]. Whereas densoviruses infect only insects, APV and dependoviruses infect vertebrate animals. APV replicate in proliferating target cells without the need of helper viruses but dependoviruses require helper virus functions for replication. Vector development has focused on three rodent APVs that can infect human cells, namely, LuIII, MVM, and H1. Dependovirus is also known as Adeno-associated virus (AAV) because dependovirus cannot replicate and form viral capsids in its host cell without the cell being coinfected by a helper virus such as an adenovirus, a herpesvirus, or a vaccinia virus [2, 3, 4, 5]. AAVs of humans and of numerous other vertebrates are known. More than 90% of human adults have antibodies to AAV, which shows that the virus is common and widely distributed. AAV serotypes 2, 3 and 5 are endemic in humans; AAV-4 infects mainly nonhuman primates and the host for serotypes 1, 6, 7 and 8 is unclear [2, 6, 7, 8, 9, 10, 11, 12]. It is noteworthy that APVs and AAVs do not cause disease in humans. Even though human exposure to AAV and H1 may lead to mild and harmless viraemia, B19 (of the Erythrovirus genus) is the only virus of the Parvovirinae subfamily known to cause pathogenicity in humans [13, 14].

In past few decades, parvoviruses have progressed from a biologically interesting observation into a crucial driver in human gene therapy. Its potential has been displayed in various preclinical and clinical research studies all around the world. Their small size, simple genetic composition and structure, and the high degree of flexibility and amenability of genome and capsid to genetic engineering are some of the key characteristics of these viruses with respect to their development and use as recombinant gene therapy vectors for DNA delivery.

2. Adeno-associated virus (AAV)

Gene therapy protocols using recombinant viral vectors have proven potentially useful in molecular medicine. AAV is one of the most actively investigated gene therapy vehicles. It is a small (25 nm), non-enveloped virus composed by an icosahedral capsid that contains a single-stranded, 4.7-kb DNA genome. AAV genome is comprised of two genes rep and cap that are flanked by two palindromic inverted terminal repeats (ITR). Rep encodes for proteins associated with replication of the viral DNA, packaging of AAV genomes, and viral genome integration in the host DNA [15]. Cap encodes for the three proteins that form the capsid. In recombinant AAV vectors (rAAV), DNA sequences of interest between the AAV inverted terminal repeats (ITRs) are cloned, eliminating the entire coding sequence of the wt AAV genome. In the absence of Rep proteins, ITR-flanked transgenes encoded within rAAV can form circular concatemers that persist as episomes in the nucleus of transduced cells [16]. During AAV assembly, rep and cap genes are provided in trans together with the adenoviral helper proteins required for AAV genome replication and packaging [17, 18]. The most common method of rAAV production is by triple transfection of HEK293 cells with three plasmids: one containing the transgene expression cassette flanked by the viral ITRs, a second packaging plasmid expressing the rep and cap genes and a third plasmid encoding for adenoviral helper genes [17, 19].

To date, 13 different AAV serotypes and 108 isolates have been identified and classified [15, 20]. AAV2 was one of the first AAV serotypes identified and characterized, including the sequence of its genome. As a result of the detailed understanding of AAV2 biology, most rAAV vectors generated today utilize the AAV2 ITRs in their vector designs. The sequences placed between the ITRs will typically include a mammalian promoter, gene of interest, and a terminator. Subtle differences in binding preferences, encoded in capsid sequence differences, can influence cell-type transduction preferences of the various AAV variants [21, 22, 23]. For example, AAV9 has a preference for primary cell binding through galactose [24], AAV2 uses the fibroblast/hepatocyte growth factor receptor and the integrins αVb5 and α5b1; AAV6 utilizes the epidermal growth factor receptor; and AAV5 utilizes the platelet-derived growth factor receptor [25]. A deeper understanding of the AAV capsid properties has made the rational design of AAV vectors that display selective tissue/organ targeting possible, thus broadening the possible applications for AAV as a gene therapy vector. Pseudotyping of rAAV vectors is used to generate tropism-modified vectors. rAAV2 genomes can be packed into capsids derived from other AAV serotypes, thus narrowing or broadening the affinity of the new viral vector for specific cell types.

AAV has been shown to be safe and effective in preclinical and clinical settings. Due to their oncogenic and immunogenic properties [26, 27], retroviral and adenoviral vectors may be associated with certain complications, but AAV has not been proven to cause any such pathological symptoms. Additionally, AAV possesses many desirable features like its ability to transduce nondividing cells [28, 29], broad host range [30], and the ability of the wild-type (wt) AAV genome to integrate site specifically into chromosome 19 in human cells [31, 32]. Besides, wt AAV has also been shown to possess antioncogenic properties [33]. AAV can infect not only actively dividing cells, but also quiescent cells, which makes it particularly valuable for many cell populations where viral and non-viral vectors are not sensitive to gene delivery, such as retinal cells and neuronal cells. The natural ability of AAV to infect quiescent cells has contributed to many significant advances in gene therapy, such as Luxturna (Spark Therapeutics) approved by the FDA for the treatment of Leber’s congenital amaurosis [34].

In the past 20 years, the relevance of AAV vector-based therapy in clinical transformation has continued to increase, and it currently accounts for 8.1% of global gene therapy clinical trials. There are currently 17 gene therapies approved by the US FDA, including the AAV vector voretigene neparvovec rzyl (VN), which was developed by Spark Therapeutics in 2017 under the trade name Luxturna [34]. VN contains an AAV2 that wraps the RPE6 gene, which is used to treat biallelic RPE65-related retinal dystrophy, a rare genetic disease that leads to impaired visual function, declines with age, and ultimately leads to blindness. The second AAV-based gene therapy approved by the FDA in 2019 is Onasemogene abeparvovec xioi (OA), developed by AveXis under the trade name Zolgensma. OA uses AAV9 expressing a functional SMN1 transgene to treat type I spinal muscular atrophy (SMA1) in children under 2 years of age [35].

Most AAV successfully used in preclinical and clinical research is limited to natural capsid serotypes. The existence of neutralizing antibodies against AAV is still an important obstacle to systemic delivery [36]. These neutralizing antibodies interfere with the entry of AAV into target cells, intracellular transport and unpacking in the nucleus, thereby preventing transduction. Epidemiological studies have shown that neutralizing antibodies with different seropositivity rates can be found in 30–60% of the population. The most popular of these neutralizing antibodies is against AAV2, followed by AAV1. Another problem of AAV-mediated gene therapy is the size limit of the genome (4.7 kbp), including ITRs, leaving only a ∼4.5 kbp size space for the transferred gene. Engineered AAV can be designed through capsid modification, surface coupling and encapsulation to solve the limitations of natural AAV [37]. A common goal of AAV engineering is to avoid inactivation by neutralizing antibodies in the blood circulation after systemic administration. Another benefit of AAV engineering is to improve targeted delivery and activation by binding tissue-specific ligands to the capsid, surface coupling and encapsulating materials. Engineered AAV can also be used to overcome the limited genome size and combine multiple treatment modalities for multimodal therapy.

3. Autonomous parvovirus vectors

Autonomously replicating parvovirus (ARP) can replicate in proliferating cells without the need for a helper virus. This is one feature that makes the ARPs attractive for potential vector production. ARPs are found in many species; they do not require a helper virus for replication, but they do require proliferating cells (S-phase functions) and, in some cases, tissue-specific factors [38]. Most vector work has been focused on autonomous parvoviruses that can infect human cells, namely, LuIII, MVM (minute virus of mice), and H1, which are members of the rodent group of APVs. Autonomous parvoviruses were first isolated from human tumor tissue and it was then observed that they possess an onco-suppressive potential, inhibiting the formation of spontaneous and chemically or virally induced tumors in vivo and in vitro [39, 40, 41]. Autonomous parvoviruses express preferentially in cancerous cells and possess oncolytic activity that has led to their implication in potential use as vectors for cancer gene therapy. Moreover, these viruses do not cause pathogenicity in adult animals and they seem to be associated with low or no immunogenicity. Additional features that make APVs interesting candidates for gene therapy are their episomal replication and high stability [42]. APV vectors have packaging capacity for foreign DNA of approximately 4.8 kb, a limit that probably cannot be exceeded by more than a few percent.

The genome of ARP comprises of two nonstructural proteins and viral capsid proteins. The non-structural proteins, NS-1 and NS-2 are highly conserved among the rodent parvoviruses that lead to cross-reactivity in serological assays utilizing whole virus antigen. The viral capsid proteins, VP-1 and VP-2 are specific to the virus and form the basis for serological differentiation. All currently proposed MVM and H-1 vectors retain the palindromes and NS1-coding sequences [42]. Other than its role in viral DNA replication, NS1 also possess the cytotoxic activity in tumor cells which should contribute to the destruction of tumor cells directly, through oncolysis, and indirectly through the induction of an immune response via the presentation of tumor-associated antigens by APCs to lymphocytes [43]. In addition, the late promoter P38 that regulates the expression of capsid proteins is transactivated by NS1. This promoter is used for the expression of transgenes to ensure that their expression occurs only in cells that also express NS1. VP proteins are generally expressed from P38, either on a helper plasmid or in packaging cells, thereby linking the expression of capsid proteins to the viral life cycle [43].

The major problems encountered with these vectors are their low titers and the generation of wild-type or replication-competent virus (RCV) through recombination with helper plasmids [42]. Over the time, advancements have been made to enhance the titers of recombinant virus and to reduce the contamination by RCV. Genetic engineering of vector has led to their enhanced production after transfection [44]. The reduction of homology between vector and helper sequences as well as integration of helper sequences into host cell genomes have greatly reduced the generation of RCV [45, 46].

Similarly, production of LuIII transducing virus has been accomplished by co-transfection of plasmid-based helper and transducing genome constructs [47]. In general, during co-transfections to generate transducing virus, recombination between helper and transducing genomes can regenerate infectious virus with variable frequencies [47]. Elimination of DNA-DNA recombination can be achieved by providing one of the components of the packaging system in RNA form. Sindbis, the plus-strand RNA virus that can express large amounts of protein from foreign genes in a variety of vertebrate cell types [48], were used for providing components of the LuIII packaging system in RNA form. Sindbis replicon vector was used to express NS1, the major non-structural protein of LuIII. Sindbis-expressed NS1 RNA and protein were readily detectable in cultured cells; this NS1 was able to mediate production of LuIII-luciferase transducing virus [48].

4. Use of parvovirus vectors in cancer gene therapy

Gene therapy is one of the most promising approaches for cancer treatment because it has the potential to provide tumor cell selectivity and/or protection of untransformed cells of the body. In order to transduce the gene of interest, either nonviral vectors or viral vectors are used. Nonviral vector strategies include naked plasmid DNA, liposome-DNA complexes, peptide-bound DNA and electroporation [49]. The most widely used viral vectors are retroviruses, adenoviruses and herpesviruses. The preliminary data looks very promising, but most vectors suffer from downsides that limit their utility for gene therapy. While nonviral vector systems have low transfection efficiencies, most of the viral systems have the problems of poor tumor targeting, immunogenicity and low transduction efficiencies. Certain parvoviruses are characterized by their oncotropism, oncosuppression and ability to mediate long-term gene expression. Together with their human apathogenicity, these characteristics make them very interesting vector systems for cancer gene therapy. Viruses of the Parvovirinae subfamily, of the Parvoviridae family, have the ability to infect a variety of different vertebrates. Although the natural hosts of parvovirus H1, MVM and LuIII are rodents, these parvoviruses can also infect human cells [50]. Similarly, many AAV serotypes are endemic in humans and non-human primates. Despite this, neither of these viruses are pathogenic in humans.

Gene therapy strategies for tumor cells have to be highly specific, particularly when the vector is to be used systemically, in order to prevent damage to healthy tissues. The therapeutic index of most existing vectors is low [51]. They non-specifically transduce normal cells as well along with targeting the tumor cells resulting in undesired damage and cell death. In general, there are two different ways to achieve specificity of a gene therapy vector: transductional targeting and transcriptional targeting [51, 52].

Transductional targeting describes the selective uptake of the vector into the cells of interest, where the transgene is transcribed. Selective uptake can be achieved by various strategies, such as modification of the viral capsid or pseudotyping of viruses. AAVs have serotype-specific tissue tropism; thus, one approach to achieving tissue-specific transduction with a therapeutic gene is the use of different AAV serotypes [53, 54]. For example, AAV-2 preferentially transduces the liver, AAV-1 transduces the muscle and AAV-5 transduces airway epithelium [53, 55]. However, serotype-specific tissue tropism enhances transgene expression in certain tissues but does not provide absolute specificity of transgene expression in other tissues. Various re-targetting strategies have been tried to enhance the specificity, efficiency and safety of AAV vectors, in particular: (1) direct re-targeting by modification of the viral capsid using the optimal insertion site that ensures the presentation of a targeting peptide on the viral surface but does not interfere with packaging [56, 57, 58] and (2) indirect re-targeting using a molecule bound to the viral surface that binds specifically and stably to the target cell (e.g. glycoside molecules and bispecific antibodies) [59, 60]. In another study, two unique features of AAV and B19 virus were exploited to create a chimeric recombinant vector system to specifically target the primitive erythroid progenitors in human bone marrow cells [61]. Recombinant B19 virus vectors are much more efficient than the recombinant AAV vectors in transducing primary human erythroid progenitor cells. Further refinement of this vector system can be useful in cancer gene therapy applications for erythroid cell lineage in the human hematopoietic system.

In transcriptional targeting, even though the transgene might be taken up by many different cells, it is transcribed only in the target cells. Using this approach, transgenes are expressed selectively by replacing the natural promoter or by modifying the transcription-factor-binding sites within a promoter. Transcriptional targeting of AAV is mostly used to enhance transgene expression in a tissue-specific manner rather than to restrict its expression to certain tissues. There are various promoters that have been used successfully like an albumin gene promoter and a retroviral long terminal repeat promoter to express human α1 -antitrypsin in hepatocytes [62], a myelin basic protein (MBP) gene promoter to direct MBP expression specifically to oligodendrocytes [63, 64] and regulatory elements of the F4/80-gene promoter for specific expression in primary microglia [65]. Transcriptional targeting of ARPs has the benefit that the vectors are already selective for cancer cells. Transcriptional targeting of ARPs has been used to achieve cell-type specific transgene expression of the parvovirus LuIII. rLuIII vectors expressing the luciferase marker gene under the control of a chimeric promoter containing a liver-specific enhancer. It directed the preferential expression of the luciferase marker in transduced human hepatoma cells [66, 67]. Another approach targeted colon carcinoma by using hybrid H-1-MVM parvovirus vectors carrying binding sites for the heterodimeric β-catenin/Tcf transcription factor in the P4 promoter; this transcription factor functions in the wnt signalling pathway, which is constitutively activated in colon carcinoma.

Gene therapy strategies for cancer can be grouped as follows: (1) Immunogene therapy with the aim of achieving either an antitumor vaccine effect or enhancing T-cell antitumor effector capability; (2) anti-angiogenic gene therapy to reduce the supply of oxygen and nutrients to the tumor; (3) cytoreductive gene therapy by gene transfer to a large number of tumor cells in situ to achieve nonimmune tumor reduction by direct cytotoxicity or by an indirect bystander effect; and (4) transduction of HSCs with drug-resistance genes to enhance their resistance to cytotoxic drugs. Depending on the desired gene therapy approach, there are different requirements the vector must fulfil with regard to safety and efficiency of the vector, specific targeting of gene transduction, expression level of the transduced gene and ease of manufacture [68].

5. Conclusion

It has become increasingly clear that parvovirus-based vectors are a potentially safe and useful alternative to the more commonly used retroviral and adenoviral vectors. Gene transfer vectors based on the replication-defective (adeno-associated virus) and autonomous parvoviruses are emerging as promising vehicles for gene therapeutic approaches. AAV has been exploited as a gene delivery vector due to its unique characteristics like small size, simple genetic composition, lack of inflammatory response, and ability to transduce both dividing and non-dividing cells followed by persistence for the lifetime of the cell. AAV-based vectors are nonpathogenic and possess an extremely wide host and tissue range. Unlike AAV, autonomous parvoviruses do not integrate. However, their tropism for transformed tissues and innate oncolytic properties may permit rapid in situ therapies. As a consequence, APVs, including MVM and H-1 virus, have been developed as antitumor vectors with the aim of strengthening the antineoplastic effect of the natural parvoviruses. With the emergence of clinically approved products in the global market and more and more successful clinical trials being conducted, AAV is at the forefront of gene therapy, but its smaller genome and neutralizing antibodies limit its application in many diseases. Present research suggests that the genetic modification of AAV vectors may further increase the success of AAV gene therapy. Vector can be engineered to increase AAV transduction efficiency by optimizing the transgene cassette. Moreover, capsid engineering can enhance vector tropism and the ability of the capsid and transgene to avoid the host immune response. Genetic manipulation of these components to optimize the large-scale production of AAV is also being explored.

APVs can prove to be superior alternative to more established vectors for gene transfer, particularly with respect to their potential use in cancer therapy. Based on the natural diversity of APVs and the ability to generate pseudo types with capsids from closely related members of the group, they should complement AAV vectors as well as offer various advantages like circumventing immune responses and exploiting tissue tropisms. However, substantial work is required to completely explore the pros and cons of these vectors, especially in context of mechanisms of transduction and the range of tissues that can be transduced in vivo by vectors with alternative, or modified, capsids. A significant preclinical evaluation of these vectors should lead to their application in future clinical cancer gene therapy trials.