Recent advances in cloning herpesviral genomes as infectious bacterial artificial chromosomes

Abstract

Herpesviruses are common but important pathogens in humans and animals. These viruses have large complex genomes encoding genes with diverse functions in different phases of their life cycle and associated diseases. In the last decade, genomes of herpesviruses cloned as infectious bacterial artificial chromosomes (BACs) have become powerful tools for delineating the functions of viral genes and understanding the pathogenesis of their associated diseases. Here we review the history of herpesviral genetics and recent advances in methods for cloning herpesviral genomes as infectious BACs.

Introduction

The International Committee on Taxonomy of Viruses (ICTV) updated the taxonomy for herpesviruses in 2009.¹ The former Herpesviridae family has been split into three families including Herpesviridae, Alloherpesviridae and Malacoherpesviridae, which make up the new Herpesvirales order. The current Herpesviridae includes the mammal, bird and reptile herpesviruses; the Alloherpesviridae consists of the fish and frog herpesviruses; and the Malacoherpesviridae contains a bivalve herpesvirus. Morphologically, herpesviruses are distinct from other viruses. A mature herpesvirus virion has a capsid containing a large double-stranded DNA genome, an outer glycoprotein-embedded envelope and the tegument space between the capsid and envelope.² The life cycle of herpesviruses typically consists of latent and lytic replication phases.² During latent infection, herpesviral genomes persist as circular episomes in the nucleus expressing a limited number of latent genes/products. During lytic infection, viral lytic genes are expressed accompanying the production of infectious virions that contain encapsidated linear viral genomes. The large DNA genomes of herpesviruses encode genes with diverse functions in distinct steps of virus life cycle including attachment, penetration, replication, assembly, egress and latency as well as in the pathogenesis of their associated diseases. One of the best approaches to delineate the function of a viral gene is reverse genetics. Once a viral mutant is generated, its phenotype can be compared with that of the wild-type virus in appropriate cell culture and animal models. In the last four decades, various methods have been developed for generating herpesviral mutants and used for defining the functions of viral genes.³ The recent adaptation of the bacterial artificial chromosome (BAC) technology for cloning herpesviruses as infectious clones have revolutionized the field and accelerated our understanding of herpesviral infections and their associated diseases.

A Brief History of Herpesviral Reverse Genetics

Mutagenesis of herpesviruses in early studies was achieved by treating the infected cells with a mutagen such as bromodeoxyuridine (BrdU) and isolating the resulting random mutants that are sensitive to a specific physical, chemical or biological condition. Following the isolation of temperature-sensitive mutants of herpesvirus simplex virus-1 (HSV-1),⁴ mutants of various herpesviruses were generated using the same or similar methods.^5–15 While this is a useful method for obtaining herpesviral mutants, the mutation frequency is usually low, and the isolation and purification steps are time-consuming and labor-intensive. Furthermore, the mutation is often scattered through the entire viral genome and mutants with multiple mutations are common, making it difficult to precisely map the mutation site(s) and elucidate the function of a viral gene.

More specific mutation in a herpesviral genome became possible when the recombination strategy was applied to generate a viral mutant by co-transfection into cells with viral genomic DNA and a DNA fragment carrying the specific allele of mutation.¹⁶ Many herpesviral mutants have been obtained with this approach and used to characterize the functions of herpesviral genes. However, because of the low recombination efficiency, this approach usually generates a mixture of mutant and wild-type viruses. Subsequent selection and purification of the mutant becomes time-consuming and tedious. A similar approach was developed using overlapping inserts of cosmids covering the entire viral genome with one of the inserts carrying a specific mutation.¹⁷ Recombination of the overlapping inserts results in the production of a viral mutant without the need for further selection and purification. This later approach has been used to generate mutants for HSV-1, pseudorabies virus (PRV), Epstein-Barr virus (EBV), varicella zoster virus (VZV), murine cytomegalovirus (MCMV) and rhesus rhadinovirus (RRV).^17–23 Nonetheless, the procedures for cloning a herpesviral genome as overlapping cosmids remain challenging. Furthermore, the large cosmid inserts are usually unstable and multiple recombination events in cells are required to generate a viral mutant, rendering it prone to the introduction of undesirable mutations.

Since a BAC vector is capable of carrying a large insert (up to 300 kb), it can easily accommodate large sizes of viral genomes.²⁴ Luckow and colleagues first adapted this technology for virologic studies by cloning the circular genome of baculovirus as an infectious BAC.²⁵ As a result, BAC-to-BAC systems derived from their works have been widely used for gene expression in baculoviruses. The pioneering work of cloning the linear genome of MCMV as an infectious BAC has revolutionized the genetics of herpesviruses.²⁶ Since the initial publication of this method, the genomes of several herpesviruses have been cloned as infectious BACs.^27–49 Once an infectious BAC is obtained, it can be stably maintained in E. coli, in which genetic mutations such as point mutation, deletion and insertion can be easily introduced into the viral genome by different mutagenesis methods including transposon insertion-mediated mutagenesis,^50,51 RecA-mediated recombination,⁵² or Red/ET recombination.⁵³ Multiple specific mutations can be introduced in the genome without the need to reconstitute the infectious viral intermediates and revertant genomes can be easily obtained in E. coli by two-step Red-mediated recombination⁵⁴ or using a specific selection marker such as galK gene.⁵⁵ Importantly, genetic characterization of the mutant can be easily achieved prior to their reconstitution in cells, thus avoiding any adventitious deletions and illegitimate insertions often observed in the conventional methods. Because the construction of a mutant genome is completely independent of the biological properties of the mutant virus, this approach has the unique advantage for generating mutant viruses with growth disadvantages that are often difficult to construct with conventional methods. In the following sections, we review the procedures for cloning herpesviral genomes as infectious BACs, as well as the recent developments (Fig. 1).

Figure 1.

Methods for cloning genomes of herpesviruses as infectious bacterial artificial chromosomes (BACs). (A) Recombination of episome.²⁷,³² A BAC cassette is transfected into cells harboring viral episome. Following recombination and drug selection, a herpesviral BAC is directly recovered in E. coli²⁷ or after passage in cells, which serves to ensure their infectiousness.³² (B) Recombination of linear genome. Linear viral genome and a BAC cassette are delivered into cells by transfection of linear genome isolated from virions,²⁶ infection with virions⁴⁰ or infection by direct transfer of infected cells.³⁸ Following recombination, drug selection and/or plaque purification, a herpesviral BAC is recovered in E. coli after passage in cells and isolation of circular viral genomes²⁶ or end repair and ligation of linear recombinant herpesviral BAC genome from virions.³⁶,³⁸ (C) Recombination of overlapping cosmid inserts.⁴¹,⁵⁶ Inserts of cosmid clones covering entire herpesviral genome with one of them containing the BAC cassette are delivered into cells. Following recombination, episomal herpesviral BAC genome is recovered in E. coli. (D) Direct ligation.⁴⁸ One unit of viral genome from concatemeric viral DNA is directly cloned into a BAC cassette using a unique restriction enzyme site. In theory, the unit of viral genome can also be derived from linear viral genome following end repair or from viral episome following digestion with a unique restriction enzyme. (E) In vitro transposon-mediated cloning.⁴⁶,⁴⁹ Linear viral genome is randomly inserted with a BAC cassette by in vitro transposition. Following passage in cells, episomal herpesviral BAC genome is recovered in E. coli. In theory, in vitro transposition can also be carried out with viral episome. The episomal herpesviral BAC genome can either be directly recovered in E. coli or after passage in cells. (Areas in dashed line have not been experimentally tested).

BAC Vectors for Cloning Herpesviral Genomes as Infectious BACs

The first BAC vector, which was used to clone MCMV as an infectious BAC, consists of a mini-F factor and a selection marker guanine phosphoribosyltransferase (gpt) flanked by two loxP sites.²⁶ The gpt can efficiently convert xanthine (IMP) to xanthine monophosphate (XMP), the immediate precursor to guanine monophosphate (GMP). When IMP is provided as a precursor, the gpt gene can be used as a selectable marker for resistance to mycophenolic acid, which inhibits the conversion of IMP to XMP. In the presence of mycophenolic acid and IMP, gpt facilitates the selection and purification of viral mutants. To avoid the potential side effect on the virus, the BAC vector can be removed with aid of two loxP sites by Cre-mediated recombination. To monitor infection in cells, reporter markers such as green fluorescence protein (GFP) cassette are incorporated into the BAC vector.²⁷ Mammalian selection markers such as hygromycin-resistant gene are also included to facilitate selection of infected cells.²⁷ In some cases, dual markers such as neomycin (G418)-resistant gene and enhanced GFP (Neo/EGFP) are used in a single mammalian cassette and serve as both reporters and selection markers.³⁹ While the most common selection marker in E. coli for herpesviral BACs is the chloramphenicol-resistant gene, kanamycin-resistant gene has also been successfully used.⁴⁹

Procedures for Cloning Herpesviral Genomes as Infectious BACs

The first step in cloning a herpesviral genome as an infectious BAC is to introduce a BAC cassette, which contains the mini-F factor and selection marker(s) in E. coli, into the viral genome. This can be achieved through recombination between the viral genome and the BAC vector flanked by viral sequences surrounding the integration site,^{26–30,32–40,42–45,47} recombination of overlapping cosmid inserts,^41,56 direct ligation of viral genomic fragments with the BAC vector,⁴⁸ and direct in vitro transposition^46,49 (Fig. 1). Once the BAC vector is introduced into the herpesviral genome, the recombinant BAC can be recovered in E. coli, typically DH10B or another recombination-deficient strain, by electroporation. The recombinant BAC can be isolated by drug selection in E. coli, and the integrity of the genome and correct BAC vector insertion site can be confirmed by genetic analyses, such as restriction digestion, Southern hybridization and DNA sequencing. The verified recombinant genome can be reconstituted in mammalian cells by electroporation, nucleofection, calcium phosphate transfection or lipid-mediated transfection, and recombinant infectious virions can be recovered for phenotypic characterizations.

Cloning a Herpesviral Genome as an Infectious BAC through Recombination between Viral Genome and BAC Vector Flanked by Viral Sequences

Following the cloning of MCMV as an infectious BAC through recombination between the viral genome and BAC vector flanked by viral sequences,²⁶ most herpesviral genomes have been cloned as infectious BACs using the same or slightly modified methods (Fig. 1A and B). This approach is feasible for most herpesviruses; however, for those that do not form any plaques or are slow growing such as Kaposi's sarcoma-associated herpesvirus (KSHV) and EBV, it remains challenging. Both KSHV and EBV are gammaherpesviruses generally maintained as circular viral genomes, also known as episomes, in the latently infected cells. EBV and KSHV infectious BACs were obtained by electroporation of their latently-infected B-cells harboring high copy numbers of episomes with BAC vectors.^27,32 While RRV, also a gammaherpesvirus, can grow in primary rhesus fibroblasts, transfection efficiency in these cells with a large DNA fragment is low, making it difficult to obtain an infectious BAC. Nevertheless, an infectious RRV BAC was generated using infectious virions to infect primary rhesus fibroblasts transfected with the BAC vector.⁴⁰ Similar strategy was used to obtain an infectious BAC of VZV, a cell-associated virus, by directly transferring the infected cells onto the BAC vector-containing target cells to achieve infection.^38,57

Two strategies are generally used to recover recombinant viruses in E. coli. If the recombinant virus is in latent status with a circular viral genome, the viral DNA is isolated from cells and electroporated into E. coli.²⁷ If the recombinant virus undergoes lytic replication generating a linear viral genome, the circular viral DNA is obtained following infection of cells with the infectious virions or by direct ligation of the linear viral DNA isolated from purified virions. Both the circular or ligated viral DNA is electroporated into E. coli.^26,35

Cloning a Herpesviral Genome as an Infectious BAC through Recombination of Overlapping Cosmid Inserts

While it is possible to clone a herpesviral genome as overlapping cosmids and rescue the recombinant viral genome following delivery of the inserts in cells, this strategy is not always straightforward. However, if cosmids with overlapping inserts are already available, this approach might be an obvious choice (Fig. 1C). One of the cosmids must have an insert containing the BAC vector. Following transfection of the cosmid inserts into mammalian cells, the recombinant virus can be recovered in E. coli. This strategy has been used to clone VZV as infectious BACs.^41,56

Cloning a Herpesviral Genome as a BAC by Direct Ligation

Theoretically, both the linear and circular viral genome can be cloned as BAC clones by direct ligation (Fig. 1D). Linear viral genome can be cloned into a BAC vector by ligation following treatment of both ends. If a unique restriction enzyme site is present in the viral genome, the cicular viral genome or the viral concatemeric DNA generated during lytic replication can also be cloned into a BAC vector by direct ligation following digestion of the DNA with the unique enzyme. Human herpesvirus 6A (HHV6A) was cloned as a BAC using this strategy though the infectiousness of the BAC was not confirmed.⁴⁸ The reliance of a unique enzyme site in the viral genome determines that this approach requires prior sequence knowledge of the entire viral genome. Because herpesviruses have large complex genomes with rich repeat sequences, this information might not be always available or accurate. Furthermore, the position of the unique site is likely important as it might influence the cloning and infectiousness of the virus. Analysis of all published KSHV genome sequences identified PmeI as a unique restriction site. However, attempt to clone KSHV from the BCBL-1 cell line, whose entire viral genome sequences have not been determined, as a BAC by ligating the Pme I-digested BAC vector and the KSHV episomal DNA was unsuccessful (unpublished data). The presence of a possible second Pme I site in this genome might account for the failure of the initial attempt.³²

Cloning a Herpesviral Genome as an Infectious BAC by in vitro Transposition

Transposons are mobile DNA sequences in the genomes of prokaryotes and eukaryotes. Transposon Tn5 was one of the first identified transposon.⁵⁸ Tn5-mediated transposition has become a powerful tool for randomly distributing primer binding sites, creating gene “knockouts” and introducing a physical tag or a genetic tag into large target DNA.⁵⁹ Tn5 transposase, which is a small, single subunit enzyme, has been cloned and purified to high specific activity. Tn5, which carries out transposition without the need for host cell factors, can transpose any DNA sequences containing a short 19 bp Mosaic End (ME) Tn5 transposase recognition sequences. Because Tn5-mediated insertions into target DNA are highly random, this method has become one of the most frequently used systems. We have recently adapted this system and developed a Tn5-based BAC vector for cloning herpesviral genomes as infectious BACs (Fig. 1E).⁴⁶ Through an in vitro transposition reaction, the BAC vector is easily introduced into a viral genome, and the recombinant BAC can be quickly recovered in E. coli. Compared to other existing methods, this new approach is highly efficient and does not require any information of viral sequences, cloning of viral DNA fragments and plaque purifications. It is particularly useful for cloning herpesviral genomes without prior knowledge of the genomic sequences. As a consequence, this method is potentially useful for discovering previously unidentified viruses as well as isolation and characterizations of natural and clinical strains or isolates of herpesviruses that often have diverse sequence variations.

Cloning a Herpesviral Genome as an Infectious Auto-excisable BAC

While infectious herpesviral BACs have become important tools for studying the functions of viral genes,^60,61 the insertion of a BAC vector cassette into the viral genome often leads to genetic and phenotypic alterations of the virus. This is mainly due to the strict size requirement for viral genome packaging and stability, possible disruption of the viral genome resulting from the insertion event or introduction of special sequences such as promoter and enhancer in the cassette that might affect the expression of neighboring viral genes.^32,62,63 Several methods have been developed to excise the BAC vector cassette. One of them is to generate a BAC cassette flanked by identical viral sequences, which allows recombination between the sequences and excision of the cassette in cells.⁶³ While it has been shown to be successful for MCMV, this method is time-consuming and requires multiple viral passages and plaque purification in mammalian cells.⁶³ The Cre-loxP system has also been successfully adapted to excise the BAC cassette.⁶⁴ However, it leaves a 34 bp loxP site in the viral genome, which is sufficient to cause phenotypic alterations to the virus.⁶⁵ Two new approaches have recently been developed to remove the BAC vector cassette. The first approach is to introduce an inverted duplication of viral genomic sequences within the mini-F replicon, which can mediate the recombination and markerless excision of vector sequence upon reconstitution in mammalian cells.⁴¹ The second approach is to insert the BAC vector cassette into the genomic region containing the terminal repeats (TR) or internal repeats. Recombination of the repeats leads to auto-excision of the cassette and generation of a traceless recombinant genome.^49,66 Because repeat sequences are commonly present in herpesviruses, this approach can be easily adapted. While both methods are efficient, each has its disadvantages. The introduction of the inverted viral sequences makes it difficult to generate viral mutants with mutations adjacent to the BAC vector. For TR-mediated auto-excision, screening for an infectious BAC with the cassette inserted into the TR region is required.⁴⁹ The in vitro transposition cloning method should facilitate cloning and selection of herpesviral BACs with insertions in the repeat regions of the genomes.⁴⁶

Conclusions

While much has been learnt, the biology of herpesviruses remains to be further explored in the years to come. The application of BAC technology in herpesviral reverse genetics has accelerated the functional delineation of herpesviral genes. The recent developments of novel methods for cloning herpesviral genomes as BACs should further facilitate not only the understanding of the infection and replication of herpesviruses and their associated diseases but also the cloning and characterizations of diverse herpesviral genomes.