Validation of reference genes for quantitative gene expression analysis in experimental epilepsy

Abstract

To grasp the molecular mechanisms and pathophysiology underlying epilepsy development (epileptogenesis) and epilepsy itself, it is important to understand the gene expression changes that occur during these phases. Quantitative real-time polymerase chain reaction (qPCR) is a technique that rapidly and accurately determines gene expression changes. It is crucial, however, that stable reference genes are selected for each experimental condition to ensure that accurate values are obtained for genes of interest. If reference genes are unstably expressed, this can lead to inaccurate data and erroneous conclusions. To date, epilepsy studies have used mostly single, nonvalidated reference genes. This is the first study to systematically evaluate reference genes in male Sprague-Dawley rat models of epilepsy. We assessed 15 potential reference genes in hippocampal tissue obtained from 2 different models during epileptogenesis, 1 model during chronic epilepsy, and a model of noninjurious seizures. Reference gene ranking varied between models and also differed between epileptogenesis and chronic epilepsy time points. There was also some variance between the four mathematical models used to rank reference genes. Notably, we found novel reference genes to be more stably expressed than those most often used in experimental epilepsy studies. The consequence of these findings is that reference genes suitable for one epilepsy model may not be appropriate for others and that reference genes can change over time. It is, therefore, critically important to validate potential reference genes before using them as normalizing factors in expression analysis in order to ensure accurate, valid results.

Reference Gene Validation in the Brain Regions of Young Rats after Pentylenetetrazole-Induced Seizures

“…We could not find in PubMed any previous study analyzing gene stability in the brain using the PTZ model. Only a few works investigated the stability of housekeeping gene expression using other animal models of seizure or epilepsy: after neonatal (P10) febrile seizures [2]; after short (30 min) and long (8 h) perforant pathway stimulation [15]; after systemic/intrahippocampal pilocarpine injection [32]; and in the kainic acid model of temporal lobe epilepsy [7,15]. These works analyzed the entire hippocampus [15,32] or dental gyrus of the hippocampus [2]; an exception was the study by Crans et al (2019) [7], where expression stability was analyzed in the hippocampal tissue and in the neocortex.…”

“…Gapdh and Actb, which are widely used as reference genes [9], are unstable in most brain areas in the PTZ seizure model, according to our data. These genes also demonstrate unstable expression in various experimental conditions, including febrile seizures, the valproic acid model of autism, performant pathway stimulation seizures, and the latent phase of the kainate epilepsy model [2,3,9,15,39]. Therefore, Gapdh and Actb should be avoided for mRNA expression normalization as single reference genes in seizure models.…”

“…Despite differences in the calculations used, these methods are all based on the idea that expressions of stable reference genes relative to other candidate genes should be the least variable across experimental samples. Different algorithms provide slightly different results [2,14,15]. To escape that bias, this paper uses a novel approach by calculating a comprehensive reference gene stability ranking based on four algorithms [3,7,[15][16][17][18][19].…”

“…Different algorithms provide slightly different results [2,14,15]. To escape that bias, this paper uses a novel approach by calculating a comprehensive reference gene stability ranking based on four algorithms [3,7,[15][16][17][18][19]. Among different reference genes, Gapdh, Actb, Ppia, B2m, Hprt1, Ywhaz, Rpl13a, Pgk1, and Sdha are widely used in RT-qPCR experiments on brain tissue or brain-derived cell lines in laboratory rats.…”

Selection of Reliable Reference Genes for RT-qPCR Analysis of Bursaphelenchus mucronatus Gene Expression From Different Habitats and Developmental Stages

“…The RefFinder software comprehensively ranks the candidate reference genes based on the geometric mean (GM) values of the results from the four different statistical algorithms described above ( Xie et al, 2012 ). With the help of these five statistical algorithms, much research on the validation of reference genes under different experimental conditions has been reported ( Liu et al, 2017 ; Sadangi et al, 2017 ; Sprang et al, 2017 ; Mughal et al, 2018 ).…”