Many reference genes are used by different laboratories for gene expression analyses to indicate the relative amount of input RNA/DNA in the experiment. These reference genes are supposed to show least variation among the treatments and with the control sets in a given experiment. However, expression of reference genes varies significantly from one set of experiment to the other. Thus, selection of reference genes depends on the experimental conditions. Sometimes the average expression of two or three reference genes is taken as standard. This review consolidated the details of about 120 genes attempted for normalization during comparative expression analysis in 16 different plants. Plant species included in this review are Arabidopsis thaliana, cotton (Gossypium hirsutum), tobacco (Nicotiana benthamiana and N. tabacum), soybean (Glycine max), rice (Oryza sativa), blueberry (Vaccinium corymbosum), tomato (Solanum lycopersicum), wheat (Triticum aestivum), potato (Solanum tuberosum), sugar cane (Saccharum sp.), carrot (Daucus carota), coffee (Coffea arabica), cucumber (Cucumis sativus), kiwi (Actinidia deliciosa) and grape (Vitis vinifera). The list includes model and cultivated crop plants from both monocot and dicot classes. We have categorized plant-wise the reference genes that have been used for expression analyses in any or all of the four different conditions such as biotic stress, abiotic stress, developmental stages and various organs and tissues, reported till date. This review serves as a guide during the reference gene hunt for gene expression analysis studies.

Real-time PCR is always the method of choice for expression analyses involving comparison of a large number of treatments. It is also the favored method for final confirmation of transcript levels followed by high throughput methods such as RNA sequencing and microarray. Our analysis comprised 16 different permutation and combinations of treatments involving four different Agrobacterium strains and three time intervals in the model plant Arabidopsis thaliana. The routinely used reference genes for biotic stress analyses in plants showed variations in expression across some of our treatments. In this report, we describe how we narrowed down to the best reference gene out of 17 candidate genes. Though we initiated our reference gene selection process using common tools such as geNorm, Normfinder and BestKeeper, we faced situations where these software-selected candidate genes did not completely satisfy all the criteria of a stable reference gene. With our novel approach of combining simple statistical methods such as t test, ANOVA and post hoc analyses, along with the routine softwarebased analyses, we could perform precise evaluation and we identified two genes, UBQ10 and PPR as the best reference genes for normalizing mRNA levels in the context of 16 different conditions of Agrobacterium infection. Our study emphasizes the usefulness of applying statistical analyses along with the reference gene selection software for reference gene identification in experiments involving the comparison of a large number of treatments.

No abstract

No abstract

No abstract