Expressed Alu repeats as a novel, reliable tool for normalization of real-time quantitative RT-PCR data

Abstract

We describe a novel strategy for mRNA normalization in quantitative real-time PCR that is based on expressed Alu repeat amplification as a measure for the mRNA fraction. We show that expressed Alu repeat amplification is a fast, accurate normalization tool that can be successfully used for quantification of selected mRNA in the human transcriptome. This result is particularly important for clinical diagnosis and biomarker validation studies based on mRNA detection in human blood.

Figure 1

BDNF mRNA in human blood normalized to GAPDH and β2M. Median BDNF mRNA levels in the blood of eight control participants from the Institute of Neurology, MRC Prion Unit, and University College London (UK). The BDNF level was determined by means of quantitative real-time PCR and normalized to the level of GAPDH mRNA and β2M. The values are the averages of six independent PCR experiments. The boundary of the box closest to zero indicates the 25th percentile, the line within the box marks the median, and the boundary of the box farthest from zero indicates the 75th percentile.

Figure 2

EAR amplification protocol in RNA from total blood. (a) BLAST analysis for Alu repeat distribution in the human genome. Alu repeat sequences are present in the UTRs of many genes. The small blue points indicate the homologous sequences at the Alu human consensus repeat. (b) Alu human interspersed repetitive sequence and primers used in this study for EAR amplification. (c) C_q of EAR mRNA plotted against the log of the relative initial amount of the pooled cDNA. (d) Melt-curve diagram generated from the amplification as described in (c).

Figure 3

Validation of EAR as a tool for normalization in RT-qPCR. (a) Application of the GeNorm algorithm to eight genes to identify the three most stable housekeeping genes in blood. Average expression stability values (M) of remaining control genes during stepwise exclusion of the least stable control gene in the human blood of eight control participants. (b) Blood BDNF mRNA relative to EAR and to the three most stably expressed housekeeping genes. Pearson's correlation r = 0.56; P < 0.05. (c) Median blood BDNF mRNA levels versus GAPDH, versus the three most stable control genes and versus EAR in ten control participants. Samples used in (a-c) were from the National Neurological Institute-IRCCS "Carlo Besta", Milan (Italy). (d) Median blood BDNF mRNA levels versus GAPDH, versus β2M, and versus EAR mRNA in eight control participants from the Institute of Neurology, MRC Prion Unit, University College London (UK).

Figure 4

Application of the EAR normalization tool for the amplification of a specific blood mRNA (IL8). (a) Amplification of blood IL8 mRNA relative to EAR mRNA and to the three most stably expressed housekeeping genes. Pearson's correlation r = 0.95; P < 0.0001. (b) Median blood IL8 mRNA levels versus EAR and versus the three most stable control genes in ten control participants. Samples from the National Neurological Institute-IRCCS "Carlo Besta", Milan (Italy).

Figure 5

BDNF mRNA level normalized to EAR in the human post-mortem cortex (a) Application of the GeNorm algorithm to six genes to identify the three most stable housekeeping genes in human post-mortem cortex. Average expression stability values (M) of remaining control genes during stepwise exclusion of the least stable control gene in the human cortex of seven control participants. (b) BDNF mRNA levels in the human cortex of seven control participants and seven HD patients normalized to GAPDH (Zuccato et al. [8]) and EAR mRNA.

Figure 6

Normalization to EAR and GeNorm to evaluate TNNC1 mRNA levels, a gene specifically expressed in human skeletal muscle. (a) Application of the GeNorm algorithm to six genes to identify the three most stable housekeeping genes in human skeletal muscle. Average expression stability values (M) of remaining control genes during stepwise exclusion of the least stable control gene in the human skeletal muscle of four control participants. (b) Median skeletal muscle TNNC1 mRNA levels versus EAR and versus the three most stable control genes in four control participants. Samples from the National Neurological Institute-IRCCS "Carlo Besta", Milan (Italy). Pearson's correlation r = 0.94; P < 0.05