15 Jan 2025

Several types of RNA help regulate gene expression, and a number of these transcripts have been linked to Alzheimer’s disease. However, no one has comprehensively examined how the regulatory RNA transcriptome changes in AD brain—until now. In the January 3 Science Advances, scientists led by Leng Han at Indiana University, Indianapolis, presented such an atlas. Drawing from two postmortem cohorts comprising 1,460 tissue samples, the authors surveyed differences in noncoding and post-translationally modified RNAs between control and AD brain. They identified a total of 126,128 noncoding RNAs, of which 3,392 were differentially expressed in AD, and 953,984 post-translational modifications, of which 21,959 were associated with AD. Many of these altered RNAs modify genes or pathways implicated in the disease.

The dataset, dubbed ADatlas, is available online in an interactive format that allows scientists to search for RNAs of interest. The data portal displays RNA alterations across all six brain regions and how these relate to AD traits, as well as how each RNA correlates with gene expression.

Celeste Karch at Washington University in St. Louis called this an important resource for the field. “Across neurodegenerative diseases, we are increasingly appreciating the role of the noncoding genome in regulating cellular functions that are critical for pathogenic processes,” she wrote to Alzforum.

A few previous studies had tied specific long noncoding RNAs to AD, amyloidosis, or tau pathology (Aug 2018 news; Sep 2023 news; Sep 2023 news). Other work implicated particular post-translational RNA modifications (Dickson et al., 2013; Gaisler-Salomon et al., 2014; Khermesh et al., 2016). Nonetheless, a broad survey had not been done.

To cast a wider net, joint first authors Chengxuan Chen and Yuan Liu at Indiana University, and Zhao Zhang and Wei Hong at the University of Texas Health Science Center, Houston, analyzed findings from two large RNA-Seq datasets. The Mount Sinai Brain Bank Study consisted of 915 tissue samples from four brain regions: parahippocampal gyrus, inferior frontal gyrus, superior temporal gyrus, and anterior prefrontal cortex. The Mayo Clinic AD Genetic Study comprised 545 samples from the temporal cortex and cerebellum. The authors used four different traits to compare AD and control brains: global CDR score of the donor, Braak stage, neuritic plaque score, and mean neocortical plaque density.

For noncoding RNAs, they examined two types: long noncoding RNAs and enhancer RNAs. eRNAs are transcribed from … you guessed it, enhancers, and thus can bind those regions to facilitate polymerase recruitment and DNA looping, which brings the transcription machinery into contact with the promoters of target genes. Regarding post-translational modifications, the authors catalogued alternative polyadenylation, which affects transcript length and stability, as well as adenosine to inosine RNA editing. Inosines are typically read as guanosines, so this modification changes the underlying code and the translated protein.

What did they find? Out of 33,321 lncRNAs identified in the six brain regions, 1,462 were differentially expressed across at least one of the four AD traits. The lion’s share of these changes were in the parahippocampal gyrus, a region that includes the entorhinal cortex and hippocampus. These are some of the first areas affected by AD. Many of these lncRNAs affected immune genes and signal transduction. Two lncRNAs of note were LINC02552 and LINC02458, which were suppressed in AD in all regions except the cerebellum. They regulate genes involved in protein homeostasis, immunity, and AD.

Of the 92,897 eRNAs, 1,930 were more or less abundant in AD, predominantly in the parahippocampal gyrus or cerebellum. These eRNAs associated mostly with genes involved in synaptic signaling and ion channels. Two eRNAs came from the inflammatory genes chitinase-3-like protein 2 and mitochondrial antiviral signaling protein, which had been previously linked to AD (Sanfilippo et al., 2020; Eshraghi et al., 2021).

Among the 53,763 alternative polyadenylation (APA) events detected, levels of 556 were linked to AD. The bulk of these were in the parahippocampal gyrus, inferior frontal gyrus, or cerebellum, a region not typically associated with AD pathology. These APAs affected expression of genes associated with synaptic signaling, immunity, and AD. They included voltage-dependent ion channel 2 and the α-synuclein gene SNCA.

A-to-I editing was the most common alteration. Out of 900,221 such events identified, 21,403 were linked to AD. The vast majority of these were in the cerebellum, with parahippocampal gyrus a distant second. Some had been linked to AD before: synaptic gene GRIA2, vesicular coat protein COPA, and small ubiquitin-like modifier 1. COPA associates with fewer plaques, SUMO1 with worse tangles (Dec 2018 conference news; Yang et al., 2019; Apr 2024 conference news).

The authors noted some caveats. For one thing, the cohorts were of predominantly European ancestry, so it is unclear if the findings would apply to other ethnicities. For another, methodological differences in the RNA-Seq techniques used in each cohort may have introduced bias. Finally, the authors cautioned that they do not know if these RNA changes are a cause or consequence of AD.—Madolyn Bowman Rogers

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Largest AD Whole-Exome Study to Date Finds Two New Risk Genes 17 Aug 2018
2. Missing ‘Lnc’? Long Noncoding RNAs Bind Tau, Tame Stress Granules 22 Sep 2023
3. When Faced with Amyloidosis, Human Transplants Die by Necroptosis 19 Sep 2023
4. Tau Silences, Aβ Inflames; Hitting Excitatory Synapses Hardest 20 Dec 2018
5. Could Sumoylation Take Down Tangles? 12 Apr 2024

Paper Citations

1. Dickson JR, Kruse C, Montagna DR, Finsen B, Wolfe MS. Alternative Polyadenylation and miR-34 Family Members Regulate Tau Expression. J Neurochem. 2013 Aug 28; PubMed.
2. Gaisler-Salomon I, Kravitz E, Feiler Y, Safran M, Biegon A, Amariglio N, Rechavi G. Hippocampus-specific deficiency in RNA editing of GluA2 in Alzheimer's disease. Neurobiol Aging. 2014 Aug;35(8):1785-91. Epub 2014 Mar 1 PubMed.
3. Khermesh K, D'Erchia AM, Barak M, Annese A, Wachtel C, Levanon EY, Picardi E, Eisenberg E. Reduced levels of protein recoding by A-to-I RNA editing in Alzheimer's disease. RNA. 2016 Feb;22(2):290-302. Epub 2015 Dec 11 PubMed.
4. Sanfilippo C, Castrogiovanni P, Imbesi R, Di Rosa M. CHI3L2 Expression Levels Are Correlated with AIF1, PECAM1, and CALB1 in the Brains of Alzheimer's Disease Patients. J Mol Neurosci. 2020 Oct;70(10):1598-1610. Epub 2020 Jul 23 PubMed.
5. Eshraghi M, Adlimoghaddam A, Mahmoodzadeh A, Sharifzad F, Yasavoli-Sharahi H, Lorzadeh S, Albensi BC, Ghavami S. Alzheimer's Disease Pathogenesis: Role of Autophagy and Mitophagy Focusing in Microglia. Int J Mol Sci. 2021 Mar 24;22(7) PubMed.
6. Yang Y, Wang X, Ju W, Sun L, Zhang H. Genetic and Expression Analysis of COPI Genes and Alzheimer's Disease Susceptibility. Front Genet. 2019;10:866. Epub 2019 Sep 19 PubMed.

External Citations

1. ADatlas

Further Reading

News

1. MIR-NATs: Noncoding RNAs That Rein in Aggregating Proteins? 22 May 2021
2. Can Non-Coding RNA Tie Inflammation to Alzheimer’s Disease? 5 Jun 2011