James P. Bennett, Jr. and Paula M. Keeney
Neurodegeneration Therapeutics, Inc.
3050A Berkmar Drive
Charlottesville, VA 22901-3450
PH (434) 529-6457
FAX (434) 529-6458
Neuropathological changes of Alzheimer’s disease (AD) and Parkinson’s disease (PD) can coexist in the same sample, suggesting possible common degenerative mechanisms. We used RNA sequencing (RNA-seq) to monitor expressions of thousands of genes in frontal cortex and ventral midbrain tissues of individuals who died with clinical and neuropathological diagnoses of “pure” AD or PD, respectively.
Both AD (frontal cortex, n=10) and PD (ventral midbrain, n=14) samples showed extensive heterogeneity of gene expression when compared to control (9 for AD, 9 for PD) samples from the same brain regions. Hierarchical clustering of gene expression heatmaps revealed two gene populations (AD, 376 genes; PD, 351 genes) that separated AD or PD from control samples at false-discovery rates (q) of <5% and fold changes of at least 1.3 (AD) or 1.5 (PD).
10,124 genes were co-expressed in our AD and PD samples. A very small group of these genes (n=23) showed both low variances (<150; variance=standard deviation squared) and reduced expressions (>1.5-fold under-expression) in both AD and PD. Ingenuity Pathway Analysis (IPA) of this group revealed impairments in NAD+ biosynthesis common to both conditions. Because NAD+ levels can be supplemented, these findings suggest a potential therapy. In addition, manipulation of other genes in this group might benefit most patients with either disease if their expression could be increased and if their reduced expression were causally linked to the disease processes. Within this small group of genes we identified 5 that may play a role in pathogenesis of either or both diseases. Other genes are also potential therapeutic candidates.
AD and PD in vulnerable brain regions appear to arise from and result in independent molecular genetic abnormalities, but we identified several under-expressed genes with potential to treat both diseases. Increased NAD+ levels from supplements containing nicotinamide riboside have been demonstrated and suggest a therapy that could be tested soon.
Alzheimer’s disease (AD) and Parkinson’s disease (PD) pathologies may co-exist in the same brain [1, 2]. This situation begs the questions (among many) of whether the two disease processes are in any way related pathogenically, which process (if either) “came first”, and whether both pathological processes and clinical symptoms could share a common disease-altering treatment.
Answers to the above questions are of necessity speculative at the current level of knowledge. Insight might be gained by examination of “pure” cases of AD and PD to search for common molecular abnormalities, since clinical symptoms alone may not predict underlying pathologies accurately.
We applied RNA sequencing (RNA-seq) of total (ie, non mRNA-enriched) RNA that was rRNA-depleted to this question. RNA sequencing of total RNA allows the estimation of gene expression, which is the net sum of transcription minus degradation of pre- and messenger RNA’s for each gene. In this study we tested the hypothesis that gene expression, assessed by the increasingly popular RNAseq technology, could be used to suggest new treatments for disease, as opposed to “insights” about disease pathogenesis.
When this approach is applied to brain tissue, one has an overall picture of how the cells in that tissue responded to stresses present at the time of death of the subject. In the case of neurodegenerative diseases, one will not observe pathogenic or adaptive gene expression changes that were present in neurons that have died and been removed and may instead observe changes in “survivors”. Pathogenic gene expression changes may or may not be present in surviving vulnerable neurons, and supporting cells (such as astroglia that comprise ~90% of tissue mass) that likely contribute to the majority of expression changes observed, excepting those specifically expressed by neurons.
Bioinformatic approaches to analysis of gene sequencing data are likewise best viewed as estimates that reflect underlying assumptions based on allowable fidelities of alignment, sequencing accuracies and assembly of fragments into whole transcriptomes, most of which are subject in vivo to alternative splicing. Further sorting of individual genes into networks, a procedure still in development, offers the possibility of identifying “hub” genes that control expression of other genes. In addition, humans are not genetically/epigenetically identical, further confounding the interpretation of findings [3, 4]. This fundamental heterogeneity likely contributes to variable responses to treatments, and approaches to embrace this heterogeneity may improve selection of specific therapies for individuals, so-called “personalized (precision) medicine”. 
In spite of these limitations, RNA-seq studies of postmortem brain tissue may offer insights into similarities and differences among conditions that have similar pathological or clinical presentations. For that reason, we undertook a study of gene expression in “pure” AD and PD samples available to us.
RNA-seq followed procedures described in our prior publications [6, 7] (see also http://www.NDTherapeutics.org). Briefly, extraction of total RNA from frozen sections of frontal cortex or ventral midbrain was performed using Qiagen miRNeasy kits according to manufacturer’s instructions (www.qiagen.com). Data regarding the ages, RNA quality and postmortem intervals of our samples are provided in Table 1. Many of our samples were obtained prior to staging systems developed by the Braak group and instead used CERAD criteria (Consortium to Establish a Registry for Alzheimer’s Disease) . The two samples that did have Braak staging carried out indicated advanced AD. All postmortem samples were examined by a neuropathologist.
On-column DNAase treatments and extra buffer washes were routinely carried out during total RNA isolation. Multiplex Illumina® sequencing libraries, quantitation and sequencing were performed by CoFactor Genomics, using paired-end approach and ~60 million reads/sample.
Compressed (gz) sequencing files in fastq format were downloaded, examined with FastQC and Illumina® sequencing adapters removed with Trimmomatic®. Reads were aligned against the current (hg38) version of the human genome using Tophat2/Bowtie2. The resulting binary sequencing files (*.bam) were processed by either Cufflinks (using FPKM normalization) or Qlucore Omics Explorer® (QOE, www.qlucore.com) containing the NGS plug-in (FPKM or TMM normalization).
We chose to use gene expression variance (= standard deviation squared) as a quantitative measure of gene expression heterogeneity. Figure 1 shows plots of variances for averaged (from Cufflinks) gene expression of the AD or PD samples on the y-axes against variances of the respective CTL samples on the x-axes. There is extensive scatter within both data sets and no clear relationships among variances of AD or PD samples and their respective CTLs’ variances.
Figure 1 Variances of gene expression, on a gene-by-gene basis, for AD samples (frontal cortex, top row) and PD samples (ventral midbrain, bottom row) are plotted against variances for CTL samples for AD (frontal cortex, top row) and PD (ventral midbrain, bottom row). Variances for gene expression were calculated from FPKM estimates of expressions derived from Cufflinks analyses, using Excel.
Figure 2 shows the relationships among variances in expression of the AD or PD samples on the y-axes compared to AD/CTL or PD/CTL expression ratios on the x-axes. The datasets suggest a Gaussian-type distribution, implying both that the data are normally distributed and that the genes with mean expression ratios closest to unity also can have the highest variances.
Figure 2 Variances of gene expression, on a gene-by-gene basis, for AD samples (frontal cortex, top row) and PD samples (ventral midbrain, bottom row) plotted against AD/CTL expression ratios (top row) or PD/CTL expression ratios (bottom row). Gene expressions were calculated from FPKM estimates of expressions derived from Cufflinks analyses
Figure 3 shows a plot of averaged gene expression of PD/CTL samples on the y-axis compared to averaged gene expression of the AD/CTL samples on the x-axis on a gene-by-gene basis, for genes (n=10,124) that are co-expressed in both PD and AD samples. There is no obvious single relationship among these 10,124 genes expressed in both PD and AD samples, but the majority of these co-expressed genes appear to have similar expressions in CTL’s and AD or PD samples. It is possible to define subgroups of genes that are more expressed in PD compared to AD (blue rectangle) , and vice-versa (orange rectangle).
Figure 3. Plot of expression ratios (PD/CTL vs. AD/CTL) in postmortem samples of 10,124 co-expressed genes. Note that most genes are near ratio values of 1.0.
Table 2 shows that among these 10,124 co-expressed genes in both PD and AD samples, there is a small number that have both small variances (<150) in PD and AD populations and are under-expressed in both PD and AD populations (Fold Change (FC) < 0.67). For these genes, a metric is calculated, the “gene expression index”, that is the product of variance X expression ratio (AD/CTL, or PD/CTL). Genes with the smallest gene expression index would be predicted to influence the largest number of subjects with either condition (or both conditions). This would occur under at least two conditions: 1. The expression of the gene could be increased; and 2. The reduced expression is causal to the disease process and not solely secondary to the disease process.
Table 2. Genes expressed in AD samples (left group) or PD samples (right group) from among those co-expressed (n=10,124) that had both low variances (<150) in their respective sample populations and under-expression (disease/CTL < 0.67). These genes are potential target for therapy development (see Text for details).
Figure 4 shows a more traditional heatmap of gene expression, hierarchically clustered, where false discovery rates (FDR, q) are < 5% and fold-changes in expression are 1.3 for the AD and 1.5 for the PD samples. In both cases it is possible to define gene groups that allow separation of the samples into disease compared to CTL. There is no overlap among these gene groups (data not shown).
Figure 4. Heatmaps of genes hierarchically clustered from the AD population (A) (q<5%, Fold Change > 1.3); and the PD population (B) (q<5%, Fold Change > 1.5). Note that these analyses separated the AD and PD populations from their respective CTL samples. Both heatmaps were created in Qlucore using FPKM normalization.
Ingenuity Pathways Analyses (IPA, Qiagen) was carried out on the small group of genes shown in Table 2. These are genes that have both low variances in both AD and PD tissues and are under-expressed in AD and PD tissues. IPA revealed that the core pathways significantly disrupted in our samples were NAD biosynthesis and NAD salvage. (p values were 3.7-4.8E-03)
In this paper we have presented both a comparison of RNA-seq datasets between PD and AD samples and a novel approach to using this data to predict treatment paradigms to be tested in these populations. We found extensive heterogeneity within both PD and AD groups and a small group of co-expressed genes that demonstrated both low variation among subjects (manifested as low variance) and reduced expression in both conditions (manifested as averaged disease/ctl ratios of 0.67 or less).
We propose that traditional analytic approaches to RNA-seq data reduction are of limited help in providing directions for therapeutic trials, likely as a result of the extensive heterogeneity in the samples. We found that developing gene lists based on false discovery rate estimation and hierarchical clustering, while separating the two populations (PD or AD vs CTL), did not provide insights into therapeutic development likely to help most subjects.
We propose a new approach, based on the gene expression index (GEI), that is a mathematical product of gene expression variance (square of standard deviation) and gene expression ratio. Genes with the lowest GEI would be worthy of increased expression, either directly (e.g., by vector-mediated expression) or indirectly (e.g., by small molecule inducers or alteration of microRNA expression modifiers). Low GEI value genes would, by virtue of their low variance, be predicted to have effects on the greatest numbers of afflicted persons.
This argument assumes that genes with low GEI values are causally related to the disease process and not simply adaptive changes to the disease process. This may require empirical testing of many different genes. Low GEI value genes could potentially also serve as biomarkers to be followed both as response metrics to therapies that alter these genes and as selection criteria to be used for genes to be altered in a given individual. Such a “personalized therapy” approach does not follow the usual single agent model of drug development that has been so successful for agents of certain classes, but which has failed so far to provide disease alteration in more complex conditions such as neurodegeneration.
Based on our results and the apparent involvement of these genes in PD and/or AD, we propose increased expression in CNS of the following genes in PD and AD subjects, who might share both disease processes. Other genes from Table 1 could also be proposed:
- NR4A2: (GEI’s PD=1.79; AD=0.58). NR4A2 (nuclear receptor subfamily 4 group A member 2), aka Nurr1, is a member of the steroid-thyroid hormone-retinoid receptor superfamily. It may also act as a transcription factor, and it has relevance to both PD and AD. In PD, NR4A2 gene is expressed in lower levels in PD subjects’ peripheral blood mononuclear cells (PBMC) ( ) and exhibits polymorphisms, one of which (rs35479735) tracks with sporadic PD ( ). In AD, a deficit of hippocampal expression of miR-184, assayed by RT-qPCR, has been described . A luciferase assay experiment confirmed that miR-184 can interact with the 3’-UTR of NR4A2 , which accumulates in brain regions that also accumulate Aβ plaques in the 5X FAD mouse model of AD . Nurr1 protein has been reported to be induced in brain by an herbal combination product 
- PBX1: (GEI’s PD=9.61; AD=1.05). PBX1 (PBX homeobox 1, pre B-cell leukemia transcription factor 1) is a transcription factor and one of four members of the PBX homeobox family. Its expression is impaired in dopaminergic neurons of PD subjects , and it is expressed in Nurr1 (+) neurons in substantia nigra as they mature into tyrosine hydroxylase (+) midbrain dopaminergic neurons . PBX 1 is induced by retinoic acid [12, 13]
- GAP43: (GEI’s PD=70.43; AD=0.27). GAP43 (growth associated protein 43) gene encodes a protein (Neuromodulin) that is expressed in high levels in neural growth cones and is reduced in brains from individuals with Parkinson’s disease dementia compared to brains from individuals with Alzheimer’s disease . GAP43 is reported to be induced by: a) the combination of Y27632 (a ROCKII inhibitor) and 4-benzyl-2 methyl-1,2,4-thiadiazolidine-3,5-dione (TDZD-8; a GSK 3β inhibitor) in rats with spinal cord injury ; b) FK506, an immune modulator , and c) an essential oil 
- NRN1: (GEI’s PD=2.31; AD=30.91) NRN1 (Neuritin 1) is expressed in postmitotic neurons of the developing nervous system and is believed to be important for neural plasticity. Increased expression of NRN1 promotes maturation of synapses in hippocampal neurons prepared from Tg2576 mice, and Neuritin infusion into Tg2576 mice normalizes synaptic plasticity in hippocampal slices . NRN1 expression is increased by glial cell-derived neurotrophic factor (GDNF, ), and GDNF improves survival of nigral dopamine neurons in a chronic model of Parkinson’s disease (MitoPark mice, ).
- NMNAT2; (GEI’s PD=4.57; AD=0.24). NMNAT2 (nicotinamide nucleotide adenylyltransferase 2) is a member of the NMNAT family that catalyzes an essential step in NAD/NADP biosynthesis. NMNAT2 is predominantly expressed in brain and its expression levels correlate positively with cognitive function and negatively with AD pathology . Its mRNA protein levels are reduced in AD brain 
 Mattila PM, Roytta M, Torikka H, Dickson DW, Rinne JO (1998) Cortical Lewy bodies and Alzheimer-type changes in patients with Parkinson’s disease. Acta Neuropathol 95, 576-582.
 Smith MZ, Nagy Z, Barnetson L, King EM, Esiri MM (2000) Coexisting pathologies in the brain: influence of vascular disease and Parkinson’s disease on Alzheimer’s pathology in the hippocampus. Acta Neuropathol 100, 87-94.
 Horvath S, Gurven M, Levine ME, Trumble BC, Kaplan H, Allayee H, Ritz BR, Chen B, Lu AT, Rickabaugh TM, Jamieson BD, Sun D, Li S, Chen W, Quintana-Murci L, Fagny M, Kobor MS, Tsao PS, Reiner AP, Edlefsen KL, Absher D, Assimes TL (2016) An epigenetic clock analysis of race/ethnicity, sex, and coronary heart disease. Genome Biol 17, 171.
 Quintana-Murci L (2016) Genetic and epigenetic variation of human populations: An adaptive tale. C R Biol 339, 278-283.
 Hampel H, Vergallo A, Aguilar LF, Benda N, Broich K, Cuello AC, Cummings J, Dubois B, Federoff HJ, Fiandaca M, Genthon R, Haberkamp M, Karran E, Mapstone M, Perry G, Schneider LS, Welikovitch LA, Woodcock J, Baldacci F, Lista S, Alzheimer Precision Medicine I (2018) Precision pharmacology for Alzheimer’s disease. Pharmacol Res.
 Bennett J, Keeney P (2017) Micro RNA’s (mirna’s) may help explain expression of multiple genes in Alzheimer’s Frontal Cortex. Journal of Systems and Integrative Neuroscience 3, 1-9.
 Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, Vogel FS, Hughes JP, van Belle G, Berg L (1991) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41, 479-486.
 Annese A, Manzari C, Lionetti C, Picardi E, Horner DS, Chiara M, Caratozzolo MF, Tullo A, Fosso B, Pesole G, D’Erchia AM (2018) Whole transcriptome profiling of Late-Onset Alzheimer’s Disease patients provides insights into the molecular changes involved in the disease. Sci Rep 8, 4282.
 Moon M, Jeong I, Kim CH, Kim J, Lee PK, Mook-Jung I, Leblanc P, Kim KS (2015) Correlation between orphan nuclear receptor Nurr1 expression and amyloid deposition in 5XFAD mice, an animal model of Alzheimer’s disease. J Neurochem 132, 254-262.
 Sim Y, Park G, Eo H, Huh E, Gu PS, Hong SP, Pak YK, Oh MS (2017) Protective effects of a herbal extract combination of Bupleurum falcatum, Paeonia suffruticosa, and Angelica dahurica against MPTP-induced neurotoxicity via regulation of nuclear receptor-related 1 protein. Neuroscience 340, 166-175.
 Villaescusa JC, Li B, Toledo EM, Rivetti di Val Cervo P, Yang S, Stott SR, Kaiser K, Islam S, Gyllborg D, Laguna-Goya R, Landreh M, Lonnerberg P, Falk A, Bergman T, Barker RA, Linnarsson S, Selleri L, Arenas E (2016) A PBX1 transcriptional network controls dopaminergic neuron development and is impaired in Parkinson’s disease. EMBO J 35, 1963-1978.
 Qin P, Haberbusch JM, Soprano KJ, Soprano DR (2004) Retinoic acid regulates the expression of PBX1, PBX2, and PBX3 in P19 cells both transcriptionally and post-translationally. J Cell Biochem 92, 147-163.
 Qin P, Haberbusch JM, Zhang Z, Soprano KJ, Soprano DR (2004) Pre-B cell leukemia transcription factor (PBX) proteins are important mediators for retinoic acid-dependent endodermal and neuronal differentiation of mouse embryonal carcinoma P19 cells. J Biol Chem 279, 16263-16271.
 Bereczki E, Branca RM, Francis PT, Pereira JB, Baek JH, Hortobagyi T, Winblad B, Ballard C, Lehtio J, Aarsland D (2018) Synaptic markers of cognitive decline in neurodegenerative diseases: a proteomic approach. Brain 141, 582-595.
 Zhang G, Lei F, Zhou Q, Feng D, Bai Y (2016) Combined application of Rho-ROCKII and GSK-3beta inhibitors exerts an improved protective effect on axonal regeneration in rats with spinal cord injury. Mol Med Rep 14, 5180-5188.
 Huang TQ, Song JN, Zheng FW, Pang HG, Zhao YL, Gu H, Zhao JJ (2017) Protection of FK506 against neuronal apoptosis and axonal injury following experimental diffuse axonal injury. Mol Med Rep 15, 3001-3010.
 Villareal MO, Ikeya A, Sasaki K, Arfa AB, Neffati M, Isoda H (2017) Anti-stress and neuronal cell differentiation induction effects of Rosmarinus officinalis L. essential oil. BMC Complement Altern Med 17, 549.
 An K, Jung JH, Jeong AY, Kim HG, Jung SY, Lee K, Kim HJ, Kim SJ, Jeong TY, Son Y, Kim HS, Kim JH (2014) Neuritin can normalize neural deficits of Alzheimer’s disease. Cell Death Dis 5, e1523.
 Chen C, Li X, Ge G, Liu J, Biju KC, Laing SD, Qian Y, Ballard C, He Z, Masliah E, Clark RA, O’Connor JC, Li S (2018) GDNF-expressing macrophages mitigate loss of dopamine neurons and improve Parkinsonian symptoms in MitoPark mice. Sci Rep 8, 5460.
 Ali YO, Allen HM, Yu L, Li-Kroeger D, Bakhshizadehmahmoudi D, Hatcher A, McCabe C, Xu J, Bjorklund N, Taglialatela G, Bennett DA, De Jager PL, Shulman JM, Bellen HJ, Lu HC (2016) NMNAT2:HSP90 Complex Mediates Proteostasis in Proteinopathies. PLoS Biol 14, e1002472.