Normal Aging and Alzheimer's Disease

 

We wondered to what extent transcripts altered in the normally aged brain overlap with those altered in the Alzheimer’s brain. We thus constructed lists of upregulated/downregulated transcripts in aging brains based on the following datasets: GSE53890, GSE46193, GSE22569, GSE71620, GSE127710, GSE11882, GSE1572 and a Biorxiv paper, Atlas of genetic effects in human microglia transcriptome across brain regions, aging and disease pathologies. The database IDs of these two lists are 125583121 and 125584121. We required that transcripts be altered in at least 3 of these 8 studies, resulting in rather short upregulation (17 members) and downregulation (35 members) lists. The small size of these lists means that really extreme P-values cannot be generated against other datasets.

A single transcript was identified as upregulated in 5 of the 8 studies: SLC14A1, while three transcripts were downregulated 5 times: PLPPR4, NECTIN3, and LPAR1. Interestingly, PLPPR4 and LPAR1 both relate to lysophosphatidic acid metabolism.

Below, we detail the extent to which selected studies offer parallels with canonical aging in the brain.

Study	Log p-value against canonically upregulated in aging brain
canonical up in human Alzheimer's brain (found in at least 5 of 35 studies)	-11.61
canonical up in alcoholic brain (4 studies; n>=3)	-10.07
upregulated in mouse 12-20m brain on hiv gp120 oe (CCR5 knockout prevents neuronal injury)	-9.94
downregulated in kidney biopsies of older (vs  younger) subjects (GSE60807: Metallothioneins and renal ageing)	-8.5
upregulated in glioblastoma stem-cell like tissue (GSCs) vs GSC culture-spheres (GSE60705)	-8.3
upregulated in a.sw mouse brain on mog treatment (induces neuroinflammation; vs no treatment)(fc/p: GSE99300)	-6.38
canonical up in schizophrenia brain (17 studies; n>=3)	-5.69
downregulated in Alzheimer's signature B1 (s1: Molecular subtyping of Alzheimer’s disease)	-3.25
downregulated in Alzheimer's signature C2 (s1: Molecular subtyping of Alzheimer’s disease)	0.07
upregulated in Alzheimer's signature A (s1: Molecular subtyping of Alzheimer’s disease)	0.07
downregulated in Alzheimer's signature B2 (s1: Molecular subtyping of Alzheimer’s disease)	0.07
downregulated in Alzheimer's signature A (s1: Molecular subtyping of Alzheimer’s disease)	0.07
upregulated in Alzheimer's signature B1 (s1: Molecular subtyping of Alzheimer’s disease)	0.07
upregulated in Alzheimer's signature B2 (s1: Molecular subtyping of Alzheimer’s disease)	0.07
downregulated in Alzheimer's signature C1 (s1: Molecular subtyping of Alzheimer’s disease)	0.07

 

 

 

Study	Log p-value against canonically downregulated in aging brain
downregulated in prefrontal cortex of alcoholics (GSE161986)	-10.67
human embryo ventral midbrain dopaminergic 1 cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-9.71
downregulated in basal ganglia of HIV patients (vs healthy)(GSE35864: two types of HIV-associated neurocognitive impairment)	-9.4
human embryo ventral midbrain serotonergic cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-8.4
downregulated in mouse postnatal brainstem on hoxa5 ko (s3: Conditional Loss of Hoxa5 Function Early after Birth)	-8.09
human embryo ventral midbrain dopaminergic 2 cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-7.84
extended list of upregulated in postmortem brains of hiv patients w/neurocognitive disorders who took antiretrovirals (vs didn't)(s: fc/p: Significant effects of antiretroviral therapy)	-7.8
mouse olfactory bulb neuron cluster 7 (top 200 by p-values) (s1: Single-Cell Profiling and SCOPE-Seq Reveal)	-7.37
fetal interneurons cluster 1 (s: landscape of the human prefrontal cortex)	-7.14
human embryo ventral midbrain mediolateral neuroblast 5 cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-6.99
human embryo ventral midbrain NBgabaergic cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-6.55
50% human embryo ventral midbrain gabaergic cluster (s2: Molecular Diversity of Midbrain Development in Mouse Human)	-6.28
upregulated in postmortem brains of hiv patients w/neurocognitive disorders who took antiretrovirals (vs didn't; rep1 only)(fc/p: GSE28160: Significant effects of antiretroviral therapy)	-6.27
upregulated in adult mouse hippocampus on cannabis feeding (GSE58437)	-6.22
upregulated in postmortem brains of hiv patients w/neurocognitive disorders who took antiretrovirals (vs didn't)(fc/p: GSE28160: Significant effects of antiretroviral therapy)	-6.21
upregulated in sh-sy5y on rest kd (raw data: GSE98871)	-6.03
downregulated in Alzheimer's signature C2 (s1: Molecular subtyping of Alzheimer’s disease)	-1.37
downregulated in Alzheimer's signature B1 (s1: Molecular subtyping of Alzheimer’s disease)	-0.61
upregulated in Alzheimer's signature A (s1: Molecular subtyping of Alzheimer’s disease)	-0.61
downregulated in Alzheimer's signature B2 (s1: Molecular subtyping of Alzheimer’s disease)	-0.61
downregulated in Alzheimer's signature A (s1: Molecular subtyping of Alzheimer’s disease)	0.15
upregulated in Alzheimer's signature B1 (s1: Molecular subtyping of Alzheimer’s disease)	0.15
upregulated in Alzheimer's signature B2 (s1: Molecular subtyping of Alzheimer’s disease)	0.15
downregulated in Alzheimer's signature C1 (s1: Molecular subtyping of Alzheimer’s disease)	0.15

 A handful of observations:

1) Alcoholism does a great job of generating the fingerprint of brain aging! Is it the alcohol or the addiction? Previously, we noted that cocaine abuse strongly overlaps with aspects of Alzheimer's.

2) While our “canonically upregulated in Alzheimer’s” set offers the single best parallel with the aging brain, our “canonically downregulated in Alzheimer’s” set is not to be found (the log p-value is actually about -3, if you’re interested). Thus, while upregulated genes in Alzheimer’s could be examples of hyper-aging, that’s not so much the case for downregulated transcripts. The aberrancies in Alzheimer’s may lie more on the side of downregulation.

3) Genes downregulated in Alzheimer’s signature B1 are upregulated in normal aging! In some weird sense, these Alzheimer’s brains/tissues are under-aged!

4) While mouse neuroinflammation studies failed to match up with our Alzheimer’s lists, we do note that genes upregulated in a mouse MOG study are also upregulated in the aging brain.

5) Oddly, genes downregulated in kidneys of elderly individuals tended to be upregulated in the brain of elderly individuals.

6) Once again, HIV and/or antiretrovirals are prominent.

We also constructed lists of blood transcripts that are altered in aging: IDs 125596121 and 125596122. These are based on a mere 3 studies (GSE123698/GSE27458/GSE68526), with transcripts needing to appear in at least two of those studies to make the blood lists. Again, the lists are rather short (21 up and 79 down). One transcript was upregulated in all three studies (VCAN), while 9 transcripts were downregulated in all three studies (ABLIM1, GPR18, CCR7, LEF1, FAIM3, FAM102A, LRRN3, NELL2, RPS23).

Study	canonical up in blood of elderly
downregulated in blood of Buruli ulcer patients 90d after mycolactone-containing vesicles injection (GSE157350: Mycolactone toxin induces an inflammatory)	-9.57
upregulated in blood of SLE patients (raw data w/ttest: Ro60 autoantigen binds endogenous retroelements and regulates)	-9.34
upregulated in neutrophils from psoriasis patients (GSE106087)	-8.27
downregulated in PBMs in familial combined hyperlipidemia (familial combined hyperlipidemia and its modification)	-8.25
upregulated in blood of lupus patients taking high (vs. low) levels of steroids (GSE110174)	-7.95
upregulated in whole blood in acute (vs. convalescent) Kawasaki disease (GSE73463: Kawasaki Disease Using a Minimal Whole-Blood)	-7.86
upregulated in HIV-infected cd4+ t (tcm) cells (GSE73968)	-7.84
upregulated in erythroid progenitors on ezh2 kd (s2: Polycomb subunit composition by GATA)	-7.59
upregulated in blood of individuals w/reactions to nevirapine (antiretroviral; vs tolerant)(GSE42436)	-7.45
new canonical cytokine storm up (GSE95233 GSE43777 GSE17924 GSE97287 GSE101702 GSE111368 GSE111368)	-7.39
upregulated in blood of septic shock patients (GSE4607: Interleukin-27 is a novel candidate diagnostic biomarker)	-6.61
upregulated in MDA-MB-231 cells after treatment with SAHA (deacetylase inhibitor)(GSE77200)	-6.61
upregulated in blood of HIV patients (both w/ART therapy) on 12w chloroquine treatment (GSE71063: Chloroquine on Immune Activation and Interferon Signatures)	-6.2
upregulated in blood of HIV patients on 12w chloroquine treatment (GSE71063: Chloroquine on Immune Activation and Interferon Signatures)	-4.51
downregulated in blood of individuals w/reactions to nevirapine (antiretroviral; vs tolerant)(GSE42436)	-3.08
upregulated in Alzheimer's signature C1 (s1: Molecular subtyping of Alzheimer’s disease)	-2.98
canonical up in human Alzheimer's brain (found in at least 5 of 35 studies)	-2.89

 

 

Study	canonical down in blood of elderly
downregulated in whole blood in acute (vs. convalescent) Kawasaki disease (GSE73463: Kawasaki Disease Using a Minimal Whole-Blood)	-33.94
downregulated in resting peripheral blood cd28- (vs. +) cells (GSE105150)	-32.83
downregulated in whole blood of tuberculosis patients (fc/p: GSE42830: Transcriptional blood signatures distinguish pulmonary tuberculosis)	-31.56
upregulated in blood on non-tuberculous mycobacterial infection with high (vs. low) albumin levels (GSE97298)	-31.49
downregulated in MS patient's blood on siponomod (vs no treatment)(fc/p: GSE141381: Siponimod enriches regulatory T and B lymphocytes)	-31.42
downregulated in blood of septic shock patients (GSE4607: Interleukin-27 is a novel candidate diagnostic biomarker)	-30.04
downregulated in pbmcs of SLE patients w/high (vs. low) neutrophil counts (Stratification of Systemic Lupus Erythematosus Patients Into)	-28.97
upregulated in pbmcs of dual HCV/HIV patients after 24w anti-retroviral therapy (GSE56619: Modulation of HCV replication after combination antiretroviral)	-28.18
downregulated in cd14+ monocytes of macular degeneration patients (Monocytes from Patients with Neovascular Age-Related Macular Degeneration)	-25.52
downregulated in blood of HIV patients (both w/ART therapy) on 12w chloroquine treatment (GSE71063: Chloroquine on Immune Activation and Interferon Signatures)	-21.85
upregulated in cd8+/3+ t-cells with high (vs. low) pd1 (gene expression profiles of programmed death-1(hi) CD8 T)	-17.75
downregulated in pediatric sepsis blood (GSE145227)	-16.51
upregulated in blood on vitamin D supplementation (GSE94138)	-13.84
downregulated in whole blood of HIV patients w/acute infection (vs healthy)(fc/p: GSE29429)	-11.2
downregulated in whole blood in HIV patients (fc/p: GSE100151)	-9.6
new canonical cytokine storm down (GSE95233 GSE43777 GSE17924 GSE97287 GSE101702 GSE111368 GSE111368)	-9.55
downregulated in whole blood of HIV progressors (vs. non progressors)(GSE56837: IFN-stimulated gene LY6E in monocytes regulates the CD14/TLR4)	-8.52
downregulated in pbmcs of HIV patients with high (vs. low) viral counts (GSE140713)	-7.31
downregulated in alzheimer's mouse blood (52w)(fc/p: GSE144459: Blood Transcriptome Biomarkers of Alzheimer's)	-2.94
downregulated in blood of individuals w/reactions to nevirapine (antiretroviral; vs tolerant)(GSE42436)	-2.29
canonically upregulated in blood of alzheimer's patients (n=2)	0.02
canonically downregulated in blood of alzheimer's patients (n=2)	0.03

A few more points:

1) HIV and antiretrovirals, again!

2) There’s an obvious involvement of the immune system here, with two studies relating to lupus/SLE, and several relating to sepsis and cytokine storms, as well as psoriasis.

3) We see Vitamin D upregulating the transcripts that are downregulated in aging; more evidence for our hypothesis that Vitamin D is perfectly wonderful.

4) Take chloroquine if you seek accelerated aging. Note also that chloroquine usage is wedged between multiple studies in which sepsis and cytokine storms are seen, with the direction of alteration being the same in all these studies. I’ll pass on the chloroquine for the severe Covid-19 infection, thank you.

5) Alzheimer studies don’t intersect strongly with the above blood studies. Note that we have “canonical Alzheimer’s” lists relating to both brain and blood.

Again, we’ve pruned the above tables to make some points. As always, if you’d like to see the complete, unadulterated results, just load the relevant study IDs into our “Fisher” tool.

Do you think we’ll see future studies showing that aging itself falls into multiple transcriptomic clusters?

whatismygene.com 

Become a Patron!