Gene expression profiling of prion-infected brains: a novel disease signature for neurodegeneration in non-human primates and in humans

Background: Prion diseases or transmissible spongiform encephalopathies (TSE) are a class of fatal infectious neurodegenerative disorders whose pathogenesis mechanisms are not fully understood. The diseases manifest as sporadic, genetic or acquired. So far, neither specific biomarkers for early diagnosis nor effective therapeutic targets have been identified. The pathological molecular component of the diseases is a misfolded isoform of the prion protein (PrP) denoted as prion. Mounting evidence suggests that in addition to gene coding for the PrP (PRNP) other genes may contribute to the genetic susceptibility of TSE. In this context, microarray-based gene expression analyses offer unique tools to approach neurodegenerative disorders. In particular, transcriptome profiling can be used to identify altered transcripts in response to pathogens, and select potential targets for novel therapeutic approaches. Up to date, a number of studies have been carried out in order to investigate the gene expression alterations occurring in prion-infected organisms, but most of them involved animal models such as mice, sheep and cattle, which are not closely related to humans. Several studies have been performed on non-human primates but none of them have investigated the genomic outcome of prion infection. In this study, we performed the first large-scale transcriptome gene expression analysis on BSE-infected cynomolgus macaques (Macaca fascicularis), which are an excellent model for studying human acquired prion disease. Indeed, cynomolgus macaques are evolutionary very close to humans, have a high degree of amino acid homology in PrP sequence. In addition, like the human sequence, macaque PrP possesses the same polymorphism at codon 129. Furthermore, BSE can be transmitted either intracranially or orally to these animals leading to a disease that is very similar to the human disorder, as regards preclinical incubation time, clinical symptoms and pathophysiology. Aim of the work: The initial objective of the present work was to identify the main genes that are differentially expressed in the frontal cortex of intracranially infected monkeys compared to non-infected ones. This approach could shed some light on the biological processes underlying the pathogenesis of human prion diseases, which may therefore become potential targets for both diagnostic and therapeutic strategies. Following the encouraging results obtained in monkeys, we decided to further confirm the dysregulation pattern highlighted in macaques in human prion disorders. At this step of the study, the final aim was to investigate the specificity of the identified gene signature for CJD in comparison to both healthy subjects and to other neurodegenerative diseases. This further analysis aimed at highlighting not only prion disease specific molecular mechanisms, but also potential common neurodegeneration processes. Methods: Total RNA from the gyrus frontalis superior of 12 animals – 6 intracranially BSE-challenged (A1-A6), 1 orally BSE-infected (B6) and 5 non-infected age- and sex-matched control macaques (CovA, CovB, CovC, CovD1, CovD2) – was isolated homogenizing the material with micro pestles in TRIzol (Invitrogen). DNase I digestion was then performed and RNA was checked for quantity and purity on a NanoDrop 2000 spectrophotometer (Thermo Scientific™) and integrity on a 2100 Bioanalyzer (Agilent Technologies). Samples were labeled using the GeneChip 3’IVT Express Kit (Affymetrix) and hybridized to a GeneChip Rhesus Macaque Genome Array (Affymetrix). The bioinformatics analysis identified 300 probe sets that were up- or down- regulated about twofold (≥|1.95|). Because among them no candidate appeared using FDR 0.05, we chose as criteria an unadjusted p-value of ≤0.005 together with a fold change ≥|2.0|. We then used the Ingenuity Pathways Analysis (IPA) to annotate genes according to their functional relationships and determine potential regulatory networks and pathways. In order to confirm the array results using an independent and more sensitive technique, we performed RT-qPCR for a subset of differentially expressed genes, with GAPDH and ACTB as reference genes. –RT controls were included in the plates for each primer pair and sample. The relative expression ratio was calculated using the 2-∆∆CT method. Statistical significance was calculated with the unpaired student t-test (p<0.05). Regarding human samples, we collected about 120 samples from frontal cortex of frozen postmortem brain tissue, including: prion-infected patients (vCJD, sCJD, iCJD), neurodegeneration affected patients (AD, PD, CBD, tauopathies) and controls (healthy subjects). RNA was extracted using TRIzol with PureLink® RNA Mini Kit (Life Technologies) and on-column DNase I digestion. Quantity and integrity were checked as above, and only samples with a RIN around 4.5 or higher were included in the study. Reverse transcription was then carried out using Superscript III and RT-qPCR was performed for the previously selected gene transcripts, with ACTB and RPL19 as reference genes. In addition, for all macaque and human samples, erythrocyte markers expression analysis was performed in order to exclude any relevant blood contamination. Results: The microarray-based transcriptome analysis of brains from BSE-infected macaques revealed 300 transcripts with expression changes greater than twofold. Among these, the bioinformatics analysis identified 86 genes with known functions, most of which are involved in cellular development, cell death and survival, lipid metabolism and transport and acute phase response signaling. RT-qPCR was performed on selected gene transcripts in order to validate the differential expression in infected animals versus controls. The results obtained with the microarray technology were confirmed and a five-gene signature was identified. In brief, HBB (hemoglobin, beta chain) and HBA2 (hemoglobin, alpha chain 2) were down-regulated in intracranially infected macaques, whereas TTR (transthyretin), APOC1 (apolipoprotein C1) and SERPINA3 (serpin peptidase inhibitor 3) were up-regulated. Interestingly, we found a completely different expression pattern for B6, the only orally-infected sample available, in comparison to the intracranially infected animals, for three genes (USP16, NR4A2, HBB), suggesting that the route of infection might play a substantial role in determining the gene expression regulation. Given that the autopsy procedure could have led to the presence of some blood in the brain material, we analyzed all the samples also for expression of two specific erythrocyte markers, ALAS2 (5'-aminolevulinate synthase 2) and RHAG (Rh-associated glycoprotein), in order to assess the reliability of the results related to the regulation of both chains of hemoglobin and exclude any major influence of potential blood contamination. RT-qPCR analysis for both markers revealed negligible blood contamination (CT ≥ 35) within some samples. Given the encouraging results found in macaques, we decided to investigate if this BSE-infection gene signature was reliable also in discriminating CJD patients from healthy ones. In humans, the disease that corresponds to BSE infection in macaques would be vCJD, which arose in human population in late 90s after consumption of BSE contaminated bovine meat. However, given the limited numbers of definitive diagnosed vCJD patients (slightly more than 200 worldwide, two of which are in Italy) and considered their reduced accessibility, we decided to extend our analysis also to sCJD patients. This would also allow us to shed some light on the possible differences in gene regulation mechanisms between acquired and sporadic human prion disorders. In addition, to better investigate the influence of different etiologies, we also included some patients with iatrogenic CJD (iCJD), an acquired prion disease -as vCJD- but with a different origin, in this case patients that followed treatment with growth hormone derived from prion contaminated cadavers. Regarding control samples, we had to face the very limited availability of brain samples from healthy subjects, either age-matched with vCJD (around 30 years) or with sCJD (around 65 years). Therefore, we decided to introduce in our study some samples from patients with non-CJD neurodegenerative disorders as an additional “control” group; this would also enable the identification of possible prion-specific gene expression alterations. In general, the gene expression trend observed in macaques was confirmed in humans, with similar FC values, for four out of five genes: HBA1/2 is down-regulated in both sCJD cases and also in patients affected by other non-CJD neurodegenerative diseases, while APOC1, TTR and SERPINA3 are up-regulated in CJD patients, but not in patients affected by other neurodegenerative diseases, as they show levels of expression not different from that of the healthy controls (FC < |2|). Conclusions: In our work we used both microarray and RT-qPCR technologies that allowed us to identify a gene signature able to distinguish BSE-infected macaques from control animals. The identified genes are involved in oxygen transport and iron homeostasis (HBB, HBA2), cholesterol metabolism and lipid transport (APOC1, SERPINA3) as well as acute phase response (SERPINA3, TTR). Therefore, these results suggest that, in order to identify potential biomarkers and drug targets for prion diseases and other neurodegenerative disorders, a combination of various pathways has to be targeted, including oxygen homeostasis, cholesterol metabolism and inflammation response. Importantly, the dysregulation of four of these genes (HBA2, APOC1, TTR, SERPINA3) has been validated with similar FC values also in CJD affected human samples, confirming the reliability of our previous analysis on BSE-infected monkeys and providing important hints on some prion-specific alterations in CJD disease. These results could be extremely helpful in understanding the mechanism underlying the progression of the disease, allowing for the identification of some key players which, if not being the cause of the onset, could however be some of the target genes affected by the disease. In addition, some of our findings support the hypothesis of a potential shared mechanism underlying the onset and the development of all neurodegenerative disorders. This is in agreement with recent data supporting the idea of a unifying role of prions in these diseases in general and maybe a prion-like behavior for most neurodegenerative disorder.