18.104.22.168. Internal reference standards
Endogenous internal reference standards (commonly called ‘housekeeping genes’) are relatively invariant host mRNA targets present in every sample that can be used to normalize quantitative data for minor variations between samples in RNA quality and quantity (Bustin et al., 2009; Radonić et al., 2004). The problem is that it is impossible to prove categorically that the expression of any candidate ‘invariant’ gene is not affected by the expression of the target gene (Radonić et al., 2004). For this reason it is currently recommended to use a battery of 3 or 4 internal controls, from different classes of genes (metabolic enzymes, structural proteins, transcription factors, ribosomal proteins etc.) and construct a control-gene index, with which to normalise between samples (Bustin, 2000). Common internal reference standards for honey bee research are β-actin (Chen et al., 2005a; Shen et al., 2005a; 2005b; Locke et al., 2012), rRNA (Chantawannakul et al., 2006), microsomal glutathione-S transferase (Evans and Wheeler, 2000; Gregory et al., 2005); ribosomal proteins RP-S5 (Evans, 2004; 2006; Wheeler et al., 2006), RP49 (Corona et al., 2005; Yañez et al., 2012), RP-S8 (Kucharski and Maleszka, 2002), and transcription factors eIF3-S8 (Grozinger et al., 2003) and eF1α (Toma et al., 2000; Yamazaki et al., 2006).
One technical difficulty with endogenous internal reference standards is the presence of contaminating genomic DNA in a sample, which could be amplified instead of the mRNA. There are two solutions to this:
- Digest the RNA sample with DNAse prior to RT-PCR.
Many RNA purification kits come with this option.
- Design the RT-PCR assay such that the primers are separated by a (large)
intron in the genomic copy of the gene.
Only the spliced mRNA will be amplified by the assay (Bustin, 2000). Such intron-spanning primers have been designed for the honey bee RP49 mRNA and B-actin-isoform-2 mRNA (de Miranda and Fries, 2008; Yañez et al., 2012; Locke et al., 2012).
As stated above, all internal reference standards also require their own external standards for accurate quantification. The inclusion of internal reference standards obviously greatly increases the cost of a project. The inclusion of internal controls is therefore one of several parameters to be decided on when starting a project, based on the projects’ objectives, requirement for quantitative precision and available finances. Generally, the need for internal controls is greater for fully quantitative experiments with highly detailed analysis of relatively few samples. The need is much less for semi-quantitative survey-type studies, with fewer specific analyses and large numbers of samples.