Section 3.1.3.2. Creating an expression construct

The first step in preparing an expression system is obtaining the gene of interest. This is not nearly as daunting a task as it once was; an intense effort is now being directed at genome sequencing and the preparation of cDNA clones from a number of prokaryotic and eukaryotic organisms. There are also a large number of cloned viral genes and genomes. This means that, in most cases, an appropriate gene or cDNA can be obtained without the need to prepare a clone de novo. If the nucleic sequence is available, but the corresponding cloned DNA is not, it is usually a simple matter to prepare the desired DNA clone using the polymerase chain reaction (PCR). If the relevant genomic or cDNA clone is not available and there is no obvious way to obtain it, there are established techniques for obtaining the desired clone; however, these methods are often tedious and labour intensive. They also constitute a substantial field in their own right and, as such, lie beyond the scope of this chapter (for an overview, see Sambrook et al., 1989).

In higher eukaryotes, most mRNA strands are spliced. With minor exceptions, mRNA strands are not spliced in E. coli. In yeast, the splicing rules do not match those used in higher eukaryotes. If one expects to express a protein from a higher eukaryote in one of these systems, a cDNA must be prepared or obtained. Because some introns are large, cDNA clones are often used as the basis of expression constructs in baculovirus systems, as well as in cultured insect and mammalian cells.

In all subsequent discussions, we will assume that the experimentalist possesses both a cDNA that encodes the protein that will be expressed and an accurate sequence. If a genomic clone is available, it can be converted to cDNA form by PCR methods or by using a retroviral vector. Retroviral vectors, by nature of their life cycle, will take a gene through an RNA intermediate, thus removing unwanted introns (Shimotohno & Temin, 1982; Sorge & Hughes, 1982). If a good sequence is not available, one should be prepared. In general, expression constructs are based, more or less exclusively, on the coding region of the cDNA. The flanking 5′ and 3′ untranslated regions are not usually helpful, and if these untranslated regions are included in an expression construct, they can, in some cases, interfere with transcription, translation or both. With some knowledge of the organization of the protein, it is sometimes helpful to express portions of a complex protein for crystallization. This will be discussed in more detail later in this chapter and in Chapter 4.3 .

Optimizing the expression of the protein is extremely important. The amount of effort required to get an expression system to produce twice as much protein is usually less than that required to grow twice as much of the host; moreover, the effort to purify a recombinant protein is inversely related to its abundance, relative to the proteins of the host. There are specific rules for expressing a recombinant protein in the different host–vector systems; these will be discussed in the context of using various hosts (E. coli, yeast, baculoviruses and cultured insect and mammalian cells).

Although the precise nature of the modifications necessary to obtain efficient expression of a protein is host dependent, the tools used to produce the modified cDNA and link it to an appropriate expression plasmid or other vector are reasonably standard. In recent years, PCR has become the method of choice for manipulation of DNA; it is a relatively easy and rapid method for altering DNA segments in a variety of useful ways (Innis et al., 1990; McPherson et al., 1995). For most construction projects, the ends of the cDNA are modified, using PCR with appropriate oligonucleotide primers that have been designed to introduce useful restriction sites and/or elements essential for efficient transcription and/or translation. Since it can often be advantageous to try the expression of a given protein construct in a number of different vectors, it is useful to incorporate carefully chosen restriction sites that will enable the fragment to be inserted simultaneously, or transferred seamlessly, into different plasmids or other vectors (Fig. 3.1.3.1). PCR can also be used to create mutations in the interior of the cDNA. For some projects where large-scale mutagenesis is planned, other mutagenic techniques are particularly helpful (for example, site-directed cassette mutagenesis using Bsp MI or a related enzyme; Boyer & Hughes, 1996). Ordinarily, however, these alternative strategies are only useful if a relatively large number of mutants are needed for the project.

Figure 3.1.3.1 | top | pdf | Creating an expression construct. PCR can be used to amplify the coding region of interest, providing that a suitable template is available. PCR primers should be designed to contain one or more restriction sites that can be conveniently used to subclone the fragment into the desired expression vector. It is often possible to choose vectors and primers such that a single PCR product can be ligated to multiple vectors. The ability to test several expression systems simultaneously is advantageous, since it is impossible to predict which vector/host system will give the most successful expression of a specific protein.

If PCR is used either to modify the ends of a DNA segment or to introduce specific mutations within a segment, it should be remembered that the PCR can introduce unwanted mutations. PCR conditions should be chosen to minimize the risk of introducing unwanted mutations (start with a relatively large amount of template DNA, limit the number of amplification cycles, use relatively stringent conditions for hybridization of the primers, choose solution conditions that reduce the number of errors made in copying the DNA and use enzymes with good fidelity, such as Pfu or others that have proofreading capabilities). It is also important to sequence all of the DNA pieces generated by PCR after they have been cloned.