November 1, 2011 (Vol. 31, No. 19)
Robert Possee, Ph.D.
Progress and Prospects for Assembling Multicomponent Structures in Insect Cells
Baculovirus expression vectors are commonly used to synthesize large quantities of recombinant proteins in insect cells. Unlike prokaryotic expression systems, recombinant baculovirus-infected cells can perform any post-translational modifications required to yield functional proteins.
The baculovirus most commonly used for recombinant gene expression is the Autographa californica multiple nucleopolyhedrovirus (AcMNPV). This has a large, circular DNA genome (134 kilobase pairs), which can be manipulated using a variety of methods.
Oxford Expression Technologies (OET) has developed a strategy in which a replication-deficient AcMNPV (flashBacTM) is rescued after transfection of insect cells with a transfer vector containing the foreign gene of choice (Figure 1). The system permits nonvirologists to utilize baculovirus expression vectors with minimal training or expertise.
While most examples of genes expressed in baculovirus-infected cells feature single proteins, there is an increasing need for the production of multicomponent structures such as sub-viral particles or enzyme complexes.
Perhaps one of the most exciting areas is the use of virus-like particles in vaccine development. This approach originated with the observation that sub-viral particles composed of a surface antigen were present in the blood of patients infected with hepatitis B virus and could be used as a vaccine (HbsAg). It was shown subsequently that eukaryotic expression systems, such as yeast and insect cells, could make HbsAg without the risk of contamination with other human viruses.
Baculovirus expression systems have proven particularly versatile for making subviral particles with multiple protein components. Examples include parvoviruses (adeno-associated virus), papilloma viruses, caliciviruses, hepatitis E virus, and reoviruses. The use of one virus to make sub-viral particles of another is appropriate since the vector already performs this function to ensure its own replication.
The native baculovirus particles are at least as complex as any other structure that it might be required to synthesize. With this approach it is also feasible to replicate noninfectious components of highly pathogenic viruses (e.g., avian/swine influenzas) or ones that are difficult to propagate in vitro (e.g., hepatitis C) for use as candidate vaccines.
In addition to the high levels of recombinant protein production possible with baculoviruses, it is also feasible to insert multiple foreign gene coding regions into the virus genome. The polyhedrin and p10 gene promoters can be duplicated and inserted at different locations within AcMNPV without apparent effect on their efficacy for driving recombinant protein gene expression.
Consequently, rather than making a recombinant virus to express each component of a sub-viral particle and then co-infecting insect cells to assemble the final structure, one virus can be made to vector all genes simultaneously.
A feature of many insect and mammalian cell culture systems has been the increasing use of serum-free media. This absence of high-serum concentrations in baculovirus-infected cells simplifies protein purification and subsequent acceptance by regulatory authorities for use of the product in humans. The disadvantage is that the inhibition of protease activity during the synthesis of recombinant protein is lost and some degradation of product may be observed. This degradation can be partly ameliorated by harvesting recombinant virus-infected cultures earlier but this often decreases overall yields.
For the baculovirus system the problem is exacerbated by the presence of a cysteine protease or cathepsin (v-cath) encoded in the genome of the virus expression vector. Fortunately, v-cath can be deleted from AcMNPV without effect on virus replication. This greatly improves the yield and quality of the recombinant product and will be an important factor when trying to assemble complex multicomponent structures in insect cells, as exemplified by OET’s flashBACULTRA vector system (Figure 1).
The successful uptake of any recombinant gene-expression system depends greatly on how simple and convenient it is to use. It is now much easier to make recombinant baculoviruses compared to the early years of the system’s development. In a similar manner, the long and laborious process required to assemble multiple gene constructs into baculovirus transfer vectors looks to be drastically curtailed for future users.
Previously, the polyhedrin gene locus was the primary target for the insertion of foreign genes. For multiple gene constructs it was necessary to insert each coding region stepwise into the transfer vector. This might have required up to five sequential additions, with each one becoming more difficult as the size of the transfer plasmid increased. There was also a risk of mutations occurring within the genes already inserted, which might negate expression of one or more proteins.
The preferred strategy for such plasmid assemblies now is the one pioneered by ATG-Biosynthetics in which pairs of target gene-coding regions are assembled in parallel in baculovirus transfer vectors under polyhedrin and p10 gene promoter control (Figure 2). These pairs of foreign genes, also with transcription terminator signals, are then integrated into a single vector. The process of multiplication can be repeated iteratively so that 2, 4, 6, or even more genes may be assembled in a single vector.
This approach is particularly useful for multisubunit virus-like particles. ATG-Biosynthetic’s system is marketed as Multi-Bac and has been used successfully to express human kinase complex, gamma-secretase complex and heptameric coatomer complexes. The system uses Tn7 transposition in E. coli to insert genes at the polyhedrin gene locus but can also insert multiple gene-expression cassettes at the AcMNPV v-cath locus via the cre-lox recombination system, usually in a two-step process.
Recent work by OET has shown that it is possible to insert foreign genes simultaneously at multiple locations within the baculovirus genome by co-transfection of insect cells with flashBac DNA and two transfer vectors. This negates the step in E. coli and means that recombinant viruses can be obtained more quickly.
There also doesn’t seem to be any limitation in the amount of foreign DNA that can be inserted into the AcMNPV genome. This may be a consequence of a rod-shaped virus particle that does not have the structural constraints of some other viruses.
Baculoviruses have for some time afforded us the technology to assemble complex, multi-component protein structures in insect cells. While the use of baculoviruses for single gene expression is now widespread, multigene expression is less common owing to the perceived difficulties in making recombinant transfer vectors and viruses.
These problems have now largely been solved, and the way is open to using insect cell-based expression systems for both fundamental research into the assembly of multicomponent structures and the production of sub-viral particles for safe and effective vaccines.