For Reference Only

Reference genes (also more loosely termed “housekeeping genes”) have been used extensively for many years to normalize results from a variety of gene expression assays, including quantitative RT-PCR. More recently is has been shown that these genes may not be as consistently and stably expressed as was originally thought. Levels of GAPDH and β-actin, for example, can vary significantly between different tissues or under different experimental conditions. One solution has been to prescreen a panel of reference genes to find those most suitable for the particular conditions under which the experiment will be run. This scenario, however, can be time-consuming and tedious, particularly if many different tissue types or experimental conditions are used. In microarray experiments, work has shown that in vitro synthesized nonmammalian RNA can act as a robust reference source; Gilsbach and colleagues (p. 173) utilize this paradigm and demonstrate that similarly produced cRNA can be spiked into experimental quantitative RT-PCR samples to act as an internal control for normalization. Using the jellyfish protein, aequorin, the authors showed that, when compared with other standard reference genes, normalization with the aequorin cRNA performed identically. This methodology can be applied in almost any quantitative RT-PCR assay and abrogates the need for the validation of a reference gene for each experimental protocol.

BAC Recombineering

Straightforward modification of BAC-cloned sequences is critical for generating transgenic animals and for functional studies of DNA viruses. Although a number of methods for markerless mutagenesis have been described, existing strategies can be limited by cumbersome requirements for negative selection markers or by inefficient screening procedures. Tischer et al. (p. 191) ran up against these limitations in trying to manipulate infectious BAC clones of equine herpes virus and Marek's disease virus. They report a revised strategy for Red-mediated recombination. The process involves a so-called universal transfer construct, a PCR product containing the desired sequence modification (which for later use is partially duplicated), a positive selection marker, a strategically placed I-SceI site, and termini that are homologous to the target sequence. A first recombination inserts this construct at the target site. Next, I-SceI cleavage makes possible an intramolecular Red recombination, permitting the full removal of the selection marker. The process is cleaner than excision using Flp/FRT or Cre/lox, and is extremely flexible. The authors show the applicability of the method for long or short insertions or deletions, as well as point mutations. Overall efficiencies range from 24% for particularly challenging applications to 70%–100% for more straightforward modifications.

Bacterial Two-Hybrids Get More Picky

The basic two-hybrid system, originally developed in yeast for detection of protein-protein interactions, has been duplicated and adapted to analyze a broad range of intracellular relationships, including the disruption of protein-protein or protein-nucleic acid interactions by small molecules or other proteins. More recent transfer of these assays into bacteria has allowed scientists to take advantage of their rapid growth and more efficient transformation characteristics. A common problem that plagues both yeast and bacterial hybrid assays is the occurrence of so-called self-activating bait-prey interactions, which can produce false positive results. Negative selection markers utilizing, for example, the URA3 gene are useful for both counter-selection experiments as well as to reduce unwanted background bait-prey binding. However, thus far they have not been available in bacterial hybrid systems. In this issue, Meng et al. (p. 179) describe such a system customized for E. coli, in which the pyrF gene, the bacterial homolog of URA3, is deleted, and the expression of a episome-borne URA3 gene (which can substitute for pyrF) is introduced that is dependent on bait-prey interaction. URA3 expression, in the presence of suicide substrate 5-fluoro-orotic acid (5-FOA) will kill those bacteria exhibiting bait-prey self-activation. The authors went further by developing a co-cis-tronic HIS3-URA3 reporter able to be used under both positive (HIS3) and negative (URA3) selection conditions. Successful use of both of these new selection methods further broadens the capabilities and flexibility of the bacterial two-hybrid system.

Genetic Immunization for Antibody Generation

The antigen requirements of antibody generation services can seem downright daunting if the protein of interest proves difficult to express or purify. Even if a sufficient amount of suitable protein can be produced, the process can often be a frustrating time sink. Synthetic peptides are one alternative, but cost and immunogenicity remain important concerns. A totally different approach is to produce the antigen in the same animal from which the antibody is to be derived. In these genetic immunization procedures, the researcher need simply clone and purify an appropriate plasmid construct, which can then be directly injected into the experimental animal. In a comprehensive study, Bates et al. (p. 199) describe their findings on the best approach for antibody generation by genetic immunization. Their work reveals that hydrodynamic tail vein (HTV) and hydrodynamic limb vein (HLV) delivery in mice are efficient methods for inducing antibody responses. Compared to gene gun approaches, higher antibody levels can be generated in a shorter time frame. For added convenience, the authors show that the same plasmid construct can be transfected into Cos cells and the lysate used for screening the sera from immunized animals. The procedure appears to be equally efficient in rats and rabbits and should be readily adaptable for use in goats and sheep.

Pre-Microarray RNA Amplification

Regular readers of BioTechniques will be well aware of the ongoing advances in the reliable amplification of RNA for use in microarray experiments. These procedures allow submicrogram amounts of RNA (such as that obtained when small biological samples are collected by microdissection) to be faithfully amplified to levels appropriate for microarray analysis. Although there has been substantial progress in development and validation of these wet-lab techniques, fewer studies have examined optimal data processing procedures for such situations. To address this gap, Cope et al. (p. 165) examined the performance of four different processing algorithms on RNA prepared for Affymetrix GeneChips® using either the standard method or a two-cycle amplification procedure. The findings are reassuring, in that by every measure tested, the two-cycle amplification protocol does not significantly lessen the quality of the hybridization. In addition, analytical methods that work well for the standard protocol also perform well when amplification has been performed. While an algorithm that weights the contribution of each probe by position (to compensate for possible 3′ bias) yields slightly improved results, standard data processing is likely to be sufficient in most cases.

Making cDNA Ends Meet

RACE, or the rapid amplification of cDNA ends, has long been the method of choice for obtaining the unknown sequence at the 5′ and 3′ ends of partially characterized cDNAs. The technique has been modified in a variety of ways to increase its sensitivity, speed, and robustness. In this issue, Huang and Chen (p. 187) describe yet another tweak to this method that allows for the simultaneous amplification of both ends of a cDNA molecule, eliminating the need for performing two separate 5′ and 3′ RACE reactions. This clever trick is achieved through a combination of a standard RT template switching reaction—in which a so-called TS-oligo is added that allows the reverse transcriptase enzyme to switch templates from the mRNA to the oligonucleotide, thus creating a double-stranded molecule—and inverse PCR, with a critical ligation step between them. The ligation reaction circularizes the double-stranded cDNA, allowing primers that are directed away from the unknown sequence to be used. The method compares very favorably to standard RACE techniques, showing increased sensitivity and specificity, and is claimed by the authors to be both reliable and straightforward.