Quantifying mRNA in postmortem human brain: influence of gender, age at death, postmortem interval, brain pH, agonal state and inter-lobe mRNA variance

The quantification of mRNA in postmortem human brain is often made complicated by confounding factors. To assess the importance of potential confounders TaqMan real-time RT-PCR was used to measure seven mRNAs (beta-actin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), cyclophilin, microtubule-associated protein (MAP) 2, neuron-specific enolase (NSE), glial fibrillary acidic protein (GFAP), amyloid precursor protein (APP) isoform 770) in cortical samples taken from 90 Alzheimer's disease (AD) and 81 control brains. Demographic data for the brain samples were assessed for interaction between factors and amounts of mRNA. Gender was found to play a role in that females had lower levels of mRNA relative to males; this was consistent in both the AD and control brains. Age at death had inconsistent but significant correlations to amounts of mRNA; male and female controls both had negative correlations, female AD a positive correlation and male AD no correlation. Positive correlations were found between brain pH and amount of mRNA in all genes except glial fibrillary acidic protein (GFAP); correlations were consistent across all groupings of pathology and gender. Mean brain pH was significantly lower in AD (6.4) than in control subjects (6.5, ANOVA, p<0.01), though there was no difference between male and females of either group. No correlation was found between brain pH and age at death. Postmortem interval was correlated with brain pH in Alzheimer's disease brains but not controls. Agonal state was generally a poor predictor of mRNA levels whilst inter-lobe variance of mRNA was found to be non-significant in control brains. Given that gender, age at death and brain pH all have significant effects upon mRNA levels it is recommended that these factors be taken into account when quantifying gene expression in postmortem human brain.     

Normalization of gene expression using SYBR green qPCR: a case for paraoxonase 1 and 2 in Alzheimer's disease brains

Validating the expression stability of reference genes is crucial for reliable normalization of real-time quantitative PCR (qPCR) data, but relatively few studies have investigated this issue in brain human tissues. The present study thus aimed at identifying in human post-mortem brain tissues a set of suitable endogenous reference genes (ERG) for the expression analysis of potential candidate genes associated with Alzheimer's disease (AD). The mRNA levels of ten common ERGs (ACTB, GAPDH, GPS1, GUSB, M-RIP, PGK1, POL2RF, PPIA, UBE2D2, and YES1) were determined in the frontal cortex of autopsy-confirmed AD and non-demented control cases (n=20) using SYBR Green technology. Then, these levels were ranked according to their expression stability using three software applications: geNorm, NormFinder and BestKeeper. Whereas PPIA and UBE2D2 were among the ERGs with the most reliable expression, ACTB was the worst. Subsequently, using PPIA and UBE2D2 as ERGs for normalization, the mRNA levels of paraoxonase 1 (PON1) and paraoxonase 2 (PON2) were quantified in the frontal cortex of AD and control cases (n=80) and analyzed using the REST 2009 program. Our results indicate that both paraoxonases are expressed in the human frontal cortex and that PON2 but not PON1 mRNA levels are up-regulated in AD relative to non-demented controls. However, re-analysis of the results by ANCOVA indicated that the significance of the difference between AD and control groups depended upon the ERG used for normalization. The use of a computational method allowing the inclusion of possible confounding factors is thus recommended for the analysis of data.     

Stress induces changes of internal standard genes of amygdala

In this study, we tested changes of internal standard genes of the brain after electrical tail shock in rats. The level of 28S rRNA of the amygdala was increased significantly after the stress. There was no significant change in GAPDH, beta-actin, 36B4 mRNA and 28S rRNA at the subfornical organ and the hippocampus.     

Validation of endogenous controls for quantitative gene expression analysis: application on brain cortices of human chronic alcoholics

Real-time PCR is frequently used for gene expression quantification due to its methodological sensitivity and reproducibility. The gene expression is quantified by normalization to one or more reference genes, usually beta-actin (ACTB), glyceraldehyde-3-phosphate dehydrogenase (GAPD) or to ribosomal RNA (18S). However, different environmental or pathological conditions might also influence the expression of normalizing genes, which could severely skew the interpretation of quantitative results. This study evaluates whether 16 genes frequently used as endogenous controls in expression studies, can serve as such for comparison of human brain tissues of chronic alcoholics and control subjects. The prefrontal and motor cortices that are affected differently by chronic alcohol consumption were analyzed. The reference genes that have no or small differences in expression in alcoholics and control subjects, were found to be specific for each region: beta-actin (ACTB) and ribosomal large P0 (RPLP0) for the prefrontal cortex while importin 8 (IPO8) and RNA polymerase II (POLR2A) for the motor cortex. Four out of sixteen analyzed genes demonstrated significant differences in expression between alcoholics and controls: phosphoglycerate kinase (PGK1), hypoxanthine phosphoribosyl transferase (HPRT1) and peptidylprolyl isomerase A (PPIA) in the motor cortex and beta-2-microglobulin (B2M) in the prefrontal cortex. Our study demonstrates the importance of validation of endogenous control genes prior to real-time PCR analysis of human brain tissues. Prescribed and non-prescribed drugs, pathological or environmental conditions along with alcohol abuse may differentially influence expression of reference genes.     

Differential loss of synaptic proteins in Alzheimer's disease: implications for synaptic dysfunction

The objective of our research was to determine synaptic protein levels in brain specimens from AD subjects and age-matched control subjects. Further, to determine whether presynaptic or postsynaptic compartments of neurons are preferentially affected in AD patients, we studied 3 presynaptic vesicle proteins (synaptotagmin, synaptophysin, and Rab 3A), 2 synaptic membrane proteins (Gap 43 and synaptobrevin), and 2 postsynaptic proteins (neurogranin and synaptopodin) in specimens from AD and age-matched control brains. Two brain regions--the frontal and parietal cortices--were assessed for protein levels by immunoblotting analysis. We found a loss of both presynaptic vesicle proteins and postsynaptic proteins in all brain specimens from AD patients compared to those from age-matched control subjects. Further, we found that the loss of synaptic proteins was more severe in the frontal cortex brain specimens than in the parietal cortex brain specimens from the AD subjects compared to those from the control subjects, suggesting that the frontal brain may be critical for synaptic function in AD. Using immunohistochemistry techniques, we also determined the distribution pattern of all synaptic proteins in both the frontal and parietal cortices brain specimens from control subjects. Of the 7 synaptic proteins studied, the presynaptic proteins synaptophysin and rab 3A and the postsynaptic protein synaptopodin were the most down-regulated. Our study suggests that postsynaptic proteins and presynaptic proteins are important for synaptic function and may be related to cognitive impairments in AD.     

Decreased expression of nuclear and mitochondrial DNA-encoded genes of oxidative phosphorylation in association neocortex in Alzheimer disease

We recently reported 50% decreases in mRNA levels of mitochondrial DNA (mtDNA)-encoded cytochrome oxidase (COX) subunits I and III in Alzheimer disease (AD) brains. The decreases were observed in an association neocortical region (midtemporal cortex) affected in AD, but not in the primary motor cortex unaffected in AD. To investigate whether the decreases are specific to mtDNA-encoded mRNA, we extended this analysis to nuclear DNA (nDNA)-encoded subunits of mitochondrial enzymes of oxidative phosphorylation (OXPHOS). Brains from five AD patients showed 50–60% decreases in mRNA levels of nDNA-encoded subunit IV of COX and the β-subunit of the F₀F₁-ATP synthase in midtemporal cortex compared with mRNA levels from midtemporal cortex of control brains. In contrast, these mRNAs were not reduced in primary motor cortices of the AD brains. The amount of nDNA-encoded β-actin mRNA and the amount of 28S rRNA were not altered in either region of the AD brain. The results suggest that coordinated decreases in expression of mitochondrial and nuclear genes occur in association cortex of AD brains and are a consequence of reduced neuronal activity and downregulation of OXPHOS machinery.     

Amyloid precursor protein gene isoforms in Alzheimer's disease and other neurodegenerative disorders

Differential expression of the amyloid precursor protein gene (APP) may be important in the development of amyloidosis in Alzheimer's disease (AD) and experimentally in the brain's response to injury. Controversial data suggests that APP isoforms containing the Kunitz protease inhibitor isoform (APP KPI+) are over expressed in the brains of patients with AD when compared to the non-Kunitz protease inhibitor containing isoforms (APP KPI-). We have investigated this hypothesis using a quantitative analysis of gene expression on brain tissue collected at post-mortem. In situ hybridization has been used with synthetic oligonucleotide probes labelled with 35S to detect the two principal splice variants of APP: APP 695 (KPI-) and APP 751 (KPI+). A prospective brain bank of frozen brain specimens has been established and includes pathologically proven AD (n=15) and other neurodegenerative disorders as controls (n=18). The controls consist of frontal lobe atrophy (n=4), Huntington's disease (n=5), Parkinson's disease (n=4), motor neuron disease (n=2), multi-infarct dementia (n=1), multisystem atrophy (n=1), and subacute sclerosing panencephalitis (n=1). We have observed no significant differences in the expression of APP 695 KPI- mRNA in frontal lobe: 17.49+/-3.26 optical density (OD) units of mRNA expression in AD vs. 16.13+/-1.76 OD units mRNA in controls (P=0.80, linear regression); or temporal lobe: 14.73+/-2.96 in AD vs. 16.49+/-2.15 in controls (P=0.55). No significant differences have been found in APP 751 KPI+ in frontal lobe: 12.86+/-2.98 in AD vs. 13.70+/-2.88 in controls (P=0.97); and temporal lobe: 13.31+/-4.93 in AD vs. 11.07+/-1.99 in controls (P=0. 65). Analysis of the ratios of APP 751 KPI+ OD units of mRNA to APP 695 KPI- mRNA revealed a trend to an increased ratio which did not reach statistical significance: frontal lobe APP 751 KPI+/APP 695 KPI- 1.92+/-1.04 in AD vs. 0.86+/-0.17 in controls (P=0.54); temporal lobe 2.54+/-1.59 in AD vs. 0.96+/-0.11 controls (P=0.34). Our data has not revealed differential expression of APP mRNA isoforms in AD and supports the hypothesis that post-translational events in APP metabolism are important in amyloidogenesis and the pathogenesis of AD.     

Increased 5S rRNA oxidation in Alzheimer's disease

It is widely accepted that oxidative stress is involved in neurodegenerative disorders such as Alzheimer's disease (AD). Ribosomal RNA (rRNA) is one of the most abundant molecules in most cells and is affected by oxidative stress in the human brain. Previous data have indicated that total rRNA levels were decreased in the brains of subjects with AD and mild cognitive impairment concomitant with an increase in rRNA oxidation. In addition, level of 5S rRNA, one of the essential components of the ribosome complex, was significantly lower in the inferior parietal lobule (IP) brain area of subjects with AD compared with control subjects. To further evaluate the alteration of 5S rRNA in neurodegenerative human brains, multiple brain regions from both AD and age-matched control subjects were used in this study, including IP, superior and middle temporal gyro, temporal pole, and cerebellum. Different molecular pools including 5S rRNA integrated into ribosome complexes, free 5S rRNA, cytoplasmic 5S rRNA, and nuclear 5S rRNA were studied. Free 5S rRNA levels were significantly decreased in the temporal pole region of AD subjects and the oxidation of ribosome-integrated and free 5S rRNA was significantly increased in multiple brain regions in AD subjects compared with controls. Moreover, a greater amount of oxidized 5S rRNA was detected in the cytoplasm and nucleus of AD subjects compared with controls. These results suggest that the increased oxidation of 5S rRNA, especially the oxidation of free 5S rRNA, may be involved in the neurodegeneration observed in AD.     

Amyloid precursor protein mRNA levels in Alzheimer's disease brain

Insoluble beta-amyloid deposits in Alzheimer's disease (AD) brain are proteolytically derived from the membrane bound amyloid precursor protein (APP). The APP gene is differentially spliced to produce isoforms that can be classified into those containing a Kunitz-type serine protease inhibitor domain (K(+), APP(751), APP(770), APRP(365) and APRP(563)), and those without (K(-), APP(695) and APP(714)). Given the hypothesis that Abeta is a result of aberrant catabolism of APP, differential expression of mRNA isoforms containing protease inhibitors might play an active role in the pathology of AD. We took 513 cerebral cortex samples from 90 AD and 81 control brains and quantified the mRNA isoforms of APP with TaqMan real-time RT-PCR. After adjustment for age at death, brain pH and gender we found a change in the ratio of KPI(+) to KPI(-) mRNA isoforms of APP. Three separate probes, designed to recognise only KPI(+) mRNA species, gave increases of between 28% and 50% in AD brains relative to controls (p=0.002). There was no change in the mRNA levels of KPI-(APP 695) (p=0.898). Therefore, whilst KPI-mRNA levels remained stable the KPI(+) species increased specifically in the AD brains.     

Expression profiles of cytokines in the brains of Alzheimer's disease (AD) patients compared to the brains of non-demented patients with and without increasing AD pathology

Neuroinflammation is involved in the pathology of Alzheimer's disease (AD). Our major focus was to clarify whether neuroinflammation plays an important role in AD pathogenesis, particularly prior to the manifestation of overt dementia. We analyzed cytokine expression profiles of the brain, with focus on non-demented patients with increasing AD pathology, referred to as high pathology control (HPC) patients, who provide an intermediate subset between AD and normal control subjects, referred to as low pathology control (LPC) patients. With real-time PCR techniques, we found significant differences in interleukin (IL)-1β, 10, 13, 18, and 33, tumor necrosis factor-α (TNFα) converting enzyme (TACE), and transforming growth factor β1 (TGFβ1) mRNA expression ratios between HPC and AD patients, while no significant differences in the expression ratios of any cytokine tested here were observed between LPC and HPC patients. The cytokine mRNA expression ratios were determined as follows: first, cytokine mRNA levels were normalized to mRNA levels of a housekeeping gene, peptidyl-prolyl isomerase A (PPIA), which showed the most stable expression among ten housekeeping genes tested here; then, the normalized data of cytokine levels in the temporal cortex were divided by those in the cerebellum, which is resistant to AD pathology. Subsequently, the expression ratios of the temporal cortex to cerebellum were compared among LPC, HPC, and AD patient groups. Our results indicate that cytokines are more mobilized and implicated in the later AD stage when a significant cognitive decline occurs and develops than in the developmental course of AD pathology prior to the manifestation of overt dementia.     

Evidence for the association of the S100beta gene with low cognitive performance and dementia in the elderly

Variations in the S100beta gene may be instrumental in producing a continuum from mild cognitive decline to overt dementia. After screening 25 single nucleotide polymorphisms (SNPs) in S100beta, we observed association of the rs2300403 intron 2 SNP with poorer cognitive function in three independent populations. Moreover, we detected a significant association of this SNP with increased risk of developing dementia or Alzheimer's disease (AD) in six independent populations, especially in women and in the oldest. Furthermore, we characterised a new primate-specific exon within intron 2 (the corresponding mRNA isoform was called S100beta2). S100beta2 expression was increased in AD brain compared with controls, and the rs2300403 SNP was associated with elevated levels of S100beta2 mRNA in AD brains, especially in women. Therefore, this genetic variant in S100beta increases the risk of low cognitive performance and dementia, possibly by favouring a splicing event increasing S100beta2 isoform expression in the brain.     

The measurement of adenosine and estrogen receptor expression in rat brains following ovariectomy using quantitative PCR analysis

In our laboratory we have developed a quantitative-polymerase chain reaction (Q-PCR) strategy to examine the differential expression of adenosine receptor (ADOR), A(1), A(2A), A(2B) and A(3), and estrogen receptors (ER) alpha and beta. Brain and uterine mRNA were first used to optimise specific amplification conditions prior to SYBR Green I real time analysis of receptor subtype expression. SYBR Green I provided a convenient and sensitive means of examining specific PCR amplification product in real time, and allowed the generation of standard curves from which relative receptor abundance could be determined. Real time Q-PCR analysis was then performed, to examine changes in receptor expression levels in brains of adult female Wistar rats 3-month post ovariectomy. Comparison with sham-operated age-matched control rats demonstrated both comparative and absolute-copy number changes in receptor levels. Evaluation of both analytical methods investigated 18S rRNA as an internal reference for comparative gene expression analysis in the brain. The results of this study revealed preferential repression of ADORA(2A) (>4-fold down) and consistent (>2-fold) down-regulation of ADORA(1), ADORA(3), and ER-beta, following ovariectomy. No change was found in ADORA(2B) or ER-alpha. Analysis of absolute copy number in this study revealed a correlation between receptor expression in response to ovariectomy, and relative receptor subtype abundance in the brain.     

Clusterin mRNA and protein in Alzheimer's disease

Clusterin, a multifunctional lipoprotein is expressed in a number of tissues but expression is particularly high in the brain, where it binds to amyloid-β (Aβ), possibly facilitating Aβ transport into the bloodstream. Its concentration in peripheral blood was identified as a potential biomarker for Alzheimer's disease (AD) and predicted retention of (11)C-Pittsburgh Compound B in the temporal lobe. Single-nucleotide polymorphisms in the clusterin gene, CLU, are associated with the risk of developing AD. We measured clusterin mRNA levels in control and AD brains and investigated the relationship of the clusterin protein to soluble, insoluble, and plaque-associated Aβ. Clusterin mRNA levels were unchanged when normalized to GAPDH but modestly increased in the frontal and temporal cortex in AD in relation to NSE and MAP-2. Levels of NSE and MAP-2 mRNA were reduced in the AD frontal cortex. Clusterin protein concentration was unchanged and did not correlate with the amount of Aβ present. In the frontal cortex, clusterin concentration was higher in APOE ε4-negative brains but no effect of APOE was detected in the temporal cortex or thalamus. Overall clusterin mRNA and protein levels are unaltered in the neocortex in AD and clusterin concentration does not reflect Aβ content. The increase in clusterin noted in peripheral blood in AD may reflect increased passage of this chaperone protein across the blood-brain barrier but further work is needed to determine how CLU variants influence the development of AD.     

Reference genes for normalization: a study of rat brain tissue

Quantitative real-time polymerase chain reaction (qPCR) has become a widely used tool in the search for disease genes. When examining gene expression with qPCR in psychiatric diseases, endogenous reference gene(s) must be used for normalization. Traditionally, genes such as beta-actin (ActB), Gapd, and 18s rRNA, assumed to be stably expressed, have been used for normalization. However, it has become clear that expression of these genes is influenced by different experimental paradigms. Since stable gene expression of houskeeping genes (HKGs), therefore, cannot be expected, alternative strategies are warranted. With the overall aim to inspect the gene expression of three target genes, NMDAR1, SORT, and CREB, in rat hippocampus, we tested a panel of eight HKGs, 18s rRNA, ActB, CycA, Gapd, Hmbs, Hprt1, Rpl13A, and Ywhaz in order to select the most stably expressed gene, using the NormFinder and geNorm software applets. Additionally, we have applied four different normalization approaches for normalization of the three target genes. We found using the NormFinder software that Ywhaz is the most stably expressed gene among the eight tested HKGs. However, the results of the analysis of the target genes are highly dependent on the choice of normalization approach. Moreover, the number of HKGs, used for selection of the most stable HKG, also influences on the result of the gene expression analysis of target genes. These results underline the importance of choosing a proper normalization strategy when analyzing gene expression with qPCR. The strategy should be unbiased and carried out in every specific experimental setting.