Staying Young With Old Data

Way back in 2009, a fascinating meta-analysis was performed across 27 prior studies that previously looked for genes significantly overexpressed and underexpressed with advanced age. This study combined datasets from Human, Mouse, and Rat studies, involving a wide variety of tissues.

João Pedro de Magalhães, João Curado, and George M. Church : Meta-analysis of age-related gene expression profiles identifies common signatures of aging

Since this meta analysis was performed, a number of genes have been further investigated. This post documents the results of my deep dive into the literature, further elucidating the mechanisms that the original authors began to piece together. Since this publication, much has been learned about many of these genes, and indeed many of them seem to fit a few core themes that may underlie aging of all tissues, as well as mechanistic drivers of alzheimer’s and cardiovascular disease.

Figure 1 from the paper shows a lovely visualization of the Over (left) and Under (right) expressed genes. What we find in this dataset, after extensive literature review, are a few core themes : Complement cascade of the innate immune system, Collagen degradation and synthesis, mitochondrial function, and defense against Reactive Oxygen Species (ROS).

The rest that follows is copied directly from a google doc, and will be cleaned up later. Feel free to comment there, if you would like : ]

Like many lame websites from the 90s, everything that follows is under construction :

Some preliminary network analysis of the genes over (orange) and under (blue) expressed in aging.

Note to self, follow up on this :

Overexpressed with age

APOD : Apolioprotein D

This is the top overexpressed gene in the paper in many aging tissues.

- Nice review of its role in aging :

- has estrogen/progesterone response elements — because of this, higher amounts are found in women. Women also live longer

- Note : Quercetin is senolytic, and it activates estrogen receptors :

- it also binds and transports progesterone in the body :

- “These reports clearly identify ApoD as a “suspect at many different crime scenes,” including the aging brain”

- Blood concentrations of protein rise with age and Parkinsons :

-APOD has a conserved methionine whos job is to reduce lipid peroxides. The methionine becomes oxidized, and APOD then dimerizes. Alzheimers brains are filled with dimerized and inactive APOD, along with lipid peroxides :

- APOD found in atherosclerotic plaques :

- mutations in APOD associated with CVD :

- Extends lifespan / stress resistance in Drosophila :

- Binds/inhibits osteopontin (SPP1, also in this list, below!) :

- SPP1 is the top overexpressed gene in atherosclerosis in this paper :

- APOD is involved in the mechanisms regulating protection from oxidative stress :

- APOD preferentially binds to Arachidonic acid (an omega-6), which is oxidized in various tissues. Isoprostane is an oxidized product of AA, and can be measured in urine as a way of assessing oxidative stress! :

- This paper provides molecular simulations of APOD binding to a peroxidized eicosanoid 5s-HETE :

-This eicosanoid is synthesized from Arachidonic Acid by the enzyme ALOX5, which is associated with diseases including arthritis, atherosclerosis, IBS, and autoimmune diseases :

-5s-Hete is fascinating : “The selective synthesis of 5(S)-HETE (i.e. synthesis of 5(S)-HETE without concurrent synthesis of 5(R)-HETE) by cells is dependent on, and generally proportionate to, the presence and levels of its forming enzyme, ALOX5. Human ALOX5 is highly expressed in cells that regulate innate immunity responses, particularly those involved in inflammation and allergy. Examples of such cells include neutrophils, eosinophils, B lymphocytes, monocytes, macrophages, mast cells, dendritic cells, and the monocyte-derived foam cells of atherosclerosis tissues

-APOD Binds to the leptin receptor in the brain and may mediate connections between aging and obesity :

-Interestingly, APOD expression in the brain positively correlates with high fat diet. Perhaps its a part of leptin resistance and hypothalamic aging!

-Since fat mass increases with age in practically everyone, this is a neat finding

FCGR2B : Low affinity immunoglobulin gamma Fc receptor II-b

-This is an inhibitory FC region for an antibody. In the body, it attenuates autoimmune reactions.

-In this study, they demonstrated that by knocking it down, autoimmune reactions lead to atherosclerosis! They also demonstrated that macrophages lacking this gene were defective in phagocytosis.

-They classify a set of proteins in plaques that antibodies attacked. These include Fibulin 5 (Fibulin 3 is in this list), and Annexin A1 (A3 and A5 in this list)

C3 : Complement component 3

- Part of innate immunity — lower in this report we document that 3 C1Q genes are represented, along with C4, also overexpressed in aging

- Activated C3b binds to complement-receptor-1 (CR1) — a top alzheimers/CVD gwas hit :

- C3 variants associated with macular degeneration :

- C3 is secreted in proximal tubules in the kidney when transferrin levels are high (note that transferrin receptor is the most underexpressed gene, below) :

CLU : Clusterin

- 4th most important GWAS locus for alzheimers :

- Binds to Ku70, a DNA repair gene that causes premature aging in mice when knocked out. Perhaps it suppresses its activity? : and

CTSS : Cathepsin S — lysosomal cysteine protease

- “a lysosomal cysteine proteinase that may participate in the degradation of antigenic proteins to peptides for presentation on MHC class II molecules” —

- Collagen connections (see underexpressed genes list) : “Secreted cathepsin S cleaves some extracellular matrix (ECM) proteins. Cathepsin S may be considered the most potent elastase known. The list of proposed cathepsin S substrates includes laminin, fibronectin elastin, osteocalcin and some collagens. It also cleaves chondroitin sulfate, heparan sulfate and proteoglycans of the basal membrane. Cathepsin S plays an active role in blood vessels permeability and angiogenesis due to its elastolytic and collagenolytic activities.” —

LYZ : lysozyme

- degrades bacterial cell walls by breakdown of N-acetylmuramic acid and N-acetyl-D-glucosamine (NAG)

- Note: NAG is also known as GlcNac, and o-GlcNac is a protein modification is associated with aging pathologies. Could LYZ modulate it too? :

- O-GlcNAc and aging: C. elegans as a genetic model to test O-GlcNAc roles in type II diabetic insulin resistance:

“An insulin receptor-like mutant allele (daf-2) with reduced signaling induces dauer formation, and this was enhanced by an OGA null mutant which elevated O-GlcNAc [3], and inhibited by an OGT null mutant which reduced O-GlcNAc levels”

- According to wikipedia, when NAG polymerizes with glucoronic acid, you get hyaluronan, the molecule found to be upregulated in naked mole rats :

- Glucosamine residues form the core of the LPS o-antigen, which causes LPS toxicity

IGJ : J-chain of immunoglobulin

- Connects Fc regions of antibodies in IGA and IGM particles

- Important in Complement inhibition : ”The J-chain also plays a role in the activation of complement. J-chain negative IgM hexamers are 15–20 times more effective at activating complement than J-chain positive IgM pentamers.[11] A consequence of this lack of complement activation is it allows J-chain positive pIgM to bind antigens without causing excessive damage to epithelial membranes through complement activation.” —

- Perhaps the increased expression of this gene is a way of defending itself from elevated complement. There is evidence for this. B-cells have been shown to secrete IGM molecules that bind to LDL particles when they become oxidized :

- Do IgM antibodies protect against atherosclerosis in Lupus? :

- IGM is Required for Protection Against Atherosclerosis in LDL Receptor-Deficient Mice :

- Could there be a treatment here? Could IgM be injected that would recognize oxidized LDL, preventing it from binding free IgG and causing a complement cascade?

Complement Component 1q : multiple genes (C1QA, C1QB, C1QC)

Initiates the classical complement cascade (Ive moved B and C up in ordering)

“The antibodies IgM and all IgG subclasses except IgG4 are able to initiate the complement system.”

“C1q is composed of 18 polypeptide chains: six A-chains, six B-chains, and six C-chains. Each chain contains a collagen-like region located near the N terminus and a C-terminal globular region. The A-, B-, and C-chains are arranged in the order A-C-B on chromosome 1”

“The 80-amino acid helical component of each triple peptide contain many Gly-X-Y sequences, where X and Y are proline, isoleucine, or hydroxylysine; they, therefore, strongly resemble collagen fibrils.”

- C1Q causes multiple aging related phenotypes:

C1QA : Complement component 1Q, subunit c

C1QB : Complement component 1Q, subunit c

C1QC : Complement component 1Q, subunit c


-Member of s100 family of proteins that contain 2 Calcium binding EF-hand motifs. See also S100A4 (below)

-Binds to S100B protein, which binds calcium and zinc. S100B binds to Tau and controls its phosphorylation (alzheimers and other dementias)

MT1F : Metalothionen

- Has lots of cycstein residues and soaks up heavy metals : Copper, Cadmium, zinc — upregulated by metals and glucocorticoids

-This gene is so metal

- Note, Copper and Zinc are required for SOD1 to work properly to clear free radicals

- Some evidence that metalothionen may enhance SOD activity by gathering needed cofactors? :

“The combination of metallothionein and superoxide dismutase protects pancreatic β cells from oxidative damage.”

- Antioxidative roles of metallothionein and manganese superoxide dismutase induced by tumor necrosis factor-alpha and interleukin-6. :

“These data strongly suggest that MT and Mn-SOD in the liver cooperatively play antioxidative roles.”

Excess copper might be bad!

- Copper‐Dependent Oxidative Stress and Neurodegeneration : “This review will cover some aspects of the involvement of copper‐mediated oxidative stress in degenerative processes in the central nervous system, with special focus on the familial form of amyotrophic lateral sclerosis (FALS)”

-”Mitochondrial Dysfunction in Neurodegenerative Diseases Associated with Copper Imbalance“

“Copper is an essential transition metal ion for the function of key metabolic enzymes, but its uncontrolled redox reactivity is source of reactive oxygen species. Therefore a network of transporters strictly controls the trafficking of copper in living systems. Deficit, excess, or aberrant coordination of copper are conditions that may be detrimental, especially for neuronal cells, which are particularly sensitive to oxidative stress. Indeed, the genetic disturbances of copper homeostasis, Menkes’ and Wilson’s diseases, are associated with neurodegeneration. Furthermore, copper interacts with the proteins that are the hallmarks of neurodegenerative disorders, such as Alzheimer’s disease, Parkinson’s disease, prion diseases, and familial amyotrophic lateral sclerosis. In all cases, copper-mediated oxidative stress is linked to mitochondrial dysfunction, which is a common feature of neurodegeneration. In particular we recently demonstrated that in copper deficiency, mitochondrial function is impaired due to decreased activity of cytochrome c oxidase, leading to production of reactive oxygen species, which in turn triggers mitochondria-mediated apoptotic neurodegeneration.”

LGALS3 : Galectin-3

-Galactose-specific lectin that binds IgE

-”Galactose-specific lectin which binds IgE. May mediate with the alpha-3, beta-1 integrin the stimulation by CSPG4 of endothelial cells migration. Together with DMBT1, required for terminal differentiation of columnar epithelial cells during early embryogenesis (By similarity). In the nucleus: acts as a pre-mRNA splicing factor. Involved in acute inflammatory responses including neutrophil activation and adhesion, chemoattraction of monocytes macrophages, opsonization of apoptotic neutrophils, and activation of mast cells.”

-Galectin-3 gene (LGALS3) expression in experimental atherosclerosis and cultured smooth muscle cells : “The galectin-3 gene (LGALS3) encodes a β-galactose binding lectin.” —

-With other markers, is a strong biomarker of cardiac stress

-Is upregulated in pulmonary fibrosis, a disease characterized by abnormal collagen accumulation :

-Inhibition of Galectin-3 significantly reduced portal hypertension and fibrosis in mice (also wikipedia ^)

-Along with MPEG1, this gene is overexpressed in muscular dystrophy mice:

-Note, the fact that LGALS3 binds to beta-galactose is interesting. Senescence is associated with increased beta-galactosidase activity :

-In fact, this paper says this gene is secreted in the SASP state! :

-Galectin3 is alternatively named the “AGE receptor 3” :

-Note: evidently galactosyl-hydroxylysine is a common modification in collagen! :

-From 1971: collagen has 12% glucose+galactose content :

-Galectin 3 mediates adhesion of cells to collagens 1 and 4, by mediating interactions with the collagen receptor integrin α2β1, which is primarily expressed in platelets and epithelial cells :

MGST1 : Microsomal glutathione s-transferase 1

-Cleans up lipid peroxides, localizes to the mitochondrial membrane and ER :

ANXA3 : Annexin A3

-Member of family of calcium-dependent phospholipid-binding protein regulates cell growth and signal transduction pathways. Inhibits phospholipase A2 and cleaves inositol 1,2-cyclic phosphate to 1-phosphate. May also play a role in anti-coagulation :

-Inihibition of phospholipase A2 may cause less arachidonic acid to be formed. That would block phospholipids in the membrane from being converted to arachidonic acid. Note that aspirin effect is one step later killing COX-1 and 2 enzyme that convert arachidonic acid to pro-inflammatory prostaglandin.

ANXA5 : Annexin A5

-Used in assays to detect apoptotic cells by binding phosphatidylserine (marker of apoptosis, important in signaling). May play role in inhibition of blood coagulation by competing for phosphatidylserine binding sites with prothrombin. Also inhibits activity of phospholipase A1.

S100A4 : calcium binding protein

-” considered a specific fibroblast maker, and thus is frequently used to monitor or predict the mechanism of tissue fibrosis” — causes pulmonary fibrosis and directly signals fibroblasts to secrete fibronectin and collagen

-associates with the RAGE (Receptor of advanced glycation endproducts)

- which senses AGE’s and activates NFKb pathway, which causes TNF-alpha to be secreted and lead to an inflammatory cascade

-The RAGE receptor becomes activated by N-Acetyl-Glucosamine and other glycosaminoglycans, including heparan sulfate!

- WTF biology …. The basement membrane of blood vessels is made of GAGs and NAGs! WHY YOU ATTACKING YOU!?!?

-A drug exists which blocks this binding between S100A4 and RAGE :

-This drug is used for autoimmune conditions in farm animals … and is available in bulk on alibaba …..

GPNMB : Osteoactivin

-Highly expressed during chondrogenesis (cartilage formation) and osteogenesis :

-Note: cartilage is composed of collagen and proteoglycans in differing amounts :

-Osteoactivin has a heparan binding domain, and is HEAVILY glycosylated with NAG modifications:

-Elevated and cytoprotective in ALS motor neurons :

GSTA1 : Glutathione S-transferase 1

-Ros scavenger in cytosol

VWF : Von Willebrand factor

-Important in clotting and wound healing — defects in this gene lead to chronic nosebleeds, GI bleeding, etc

-Directly binds to COL1A1, which is in the underexpressed gene list :

-VWF binds to the LPR1 receptor, which is the receptor for APO-E, the top causative genetic factor for alzheimers, and also associated with CVD

- Macrophages use LPR1 to clear VWF, as well as lipoproteins. Knocking it out of macrophages leads to elevated VWF and lipoproteins and

-From wikipedia :

“In thrombotic thrombocytopenic purpura (TTP) and hemolytic uremic syndrome (HUS), ADAMTS13 either is deficient or has been inhibited by antibodies directed at the enzyme. This leads to decreased breakdown of the ultra-large multimers of vWF and microangiopathic hemolytic anemia with deposition of fibrin and platelets in small vessels, and capillary necrosis. In TTP, the organ most obviously affected is the brain; in HUS, the kidney.[12] Higher levels of vWF are more common among people that have had ischemic stroke (from blood-clotting) for the first time.[13] Occurrence is not affected by ADAMTS13, and the only significant genetic factor is the person’s blood group. High plasma vWF levels were found to be an independent predictor of major bleeding in anticoagulated atrial fibrillation patients.[14]”

So, if the enzyme that chews VWF is broken, VWF is elevated and fibrin and platelets are deposited in small vessels? Sounds like more of a fibrosis story ….

DERL1 : Derlin-1

“Derlin-1 is part of a complex (that includes VIMP, SEL1, HRD1, and HERP) that mediates endoplasmic-reticulum-associated degradation (ERAD) that detects misfolded proteins in the endoplasmic reticulum and targets them for destruction.”



ANIA4 : Renamed Dclk1 : doublecortin-like kinase 1

-A marker for Tuft cells, which are cilliated cells lining the GI and respiratory tracts with special chemosensory receptors. They only show up when parasitic infections or allergans are present, and specific cytokines are released that cause them to develop. This leads to a type-2 immunity. It’s mostly overexpressed in the CNS in this dataset, but slightly overexpressed in lung, which is interesting because they are responsible for helper-T cell mediated asthma reactions in mice when given IL33, also present in this dataset and associated with pulmonary fibrosis


- This gene functions as a high-affinity receptor for C-reactive protein (CRP) in leukocytes.

- This gene binds to serum amyloid P component, and other proteins of the pentraxin family. SAP is 14% of amyloid deposits in many tissues. SAP is part of the pentraxin gene family, and has 51% homology with C-reactive protein :

-Interestingly, the SAP localizes to atherosclerotic lesions, and binds directly to HDL and vLDL lipoproteins. “These results suggest that the function of SAP is related to its capability to interact with lipoproteins and this may have important implications in atherosclerosis and in amyloidosis.”



GNS : N-acetylglucosamine-6-sulfatase (NAG!)

-”It is involved in the catabolism of heparin, heparan sulphate, and keratan sulphate.[5]

-Deficiency of this enzyme results in a lysosomal storage disorder (Sanfilippo D syndrome)

-Note, Lysozyme above also degrades NAG, but in bacterial cell walls

-In recent years, heparin has become very relevant in the pathogenesis of alzheimer’s and possibly other neurodegenerative diseases

-Heparan binds and activates BACE1, an enzyme that processes beta-amyloid :

-Heparan Sulfate Proteoglycans are involved in the trans-synaptic transport of tau aggregates in the spread of alzheimers :

-”Neuronal heparan sulfates promote amyloid pathology by modulating brain amyloid-β clearance and aggregation in Alzheimer’s disease.”

-Warning, the biology of Heparin Sulfates is a DEEP rabbit hole … :

IL33 : Interluken 33

-”Cytokine that binds to and signals through the IL1RL1/ST2 receptor which in turn activates NF-kappa-B and MAPK signaling pathways in target cells (PubMed:16286016). Involved in the maturation of Th2 cells inducing the secretion of T-helper type 2-associated cytokines. Also involved in activation of mast cells, basophils, eosinophils and natural killer cells. Acts as a chemoattractant for Th2 cells, and may function as an “alarmin”, that amplifies immune responses during tissue injury”

-Nuclear IL-33 regulates soluble ST2 receptor and IL-6 expression in primary human arterial endothelial cells and is decreased in idiopathic pulmonary arterial hypertension :

-Pulmonary arterial hypertension caused by excess collagen deposition :

MPEG1 : Macrophage-expressed gene 1 protein

-Pore-forming protein that attacks invading pathogens

-Specifically activated by LPS and TNFa signaling :

-Along with Lgals3, this gene is overexpressed in muscular dystrophy mice:

NPC2 : NPC intracellular cholesterol transporter 2

- This appears to transport cholesterol out of the late endosomes, after cholesterol-containing lipoproteins are internalized. The fact that it is upregulated with age may suggest an enhanced reliance on cholesterol from external sources by many tissues.

- “Soluble NPC2 extracts cholesterol from endocytosed LDL and hands it over to the cholesterol-binding lumenal loop 1 of NPC1” :

ADIPOR2 : Adiponectin receptor 2

-This is neat. Adiponectin is a hormone secreted in fat cells that has gotten attention in obesity. Blood adiponectin levels are inversely correlated with BMI

-What’s interesting about adiponectin itself: the official name is “adiponectin, C1Q and collagen domain containing”

- That’s right. It has a collagen domain and a C1q domain….. Yet another connection to that axis ….

- The structure also resembles that of TNF-a :

- Adiponectin has been shown to be protective against atherosclerosis, and elevated leptin is bad :

-Hypothesis: maybe adiponectin receptors are increased when adiponectin itself is lowered, as in obesity



LITAF : lipopolysaccharide-induced TNF-α factor

-This is the transcription factor that expresses TNF-a when the cell is stimulated by LPS:

-Gee, I wonder if MPEG1 is activated by this transcription factor ….

SERPING1 — Complement C1 inhibitor

-Directly inhibits the complement system to prevent spontaneous activation! :





C4A : Complement component 4A

-Complement proteins C3 and C4 bind to collagen and elastin in the vascular wall: a potential role in vascular stiffness and atherosclerosis:

SPP1 : Also known as osteopontin

-Regulates bone ossification

-Has inflammatory roles: when phosphorylated it binds to integrin receptors on macrophages and polarizes them into the M1 proinflammatory state. A secret protein ACP5 (TRAP) has been observed in osteoclasts that dephosphorylates Osteopontin and returns macrophages to the M2 anti-inflammatory state.

-APOD binds to this gene and regulates its effect on cell migration.

-APOD and Osteopontin are highly coregulated in apoE knockout mice : “The top two induced genes were osteopontin (“OPN,” also called ”Spp1,” 416-fold increase) and apoD (80-fold increase); their expression levels in dKO mice increased with increasing age” —

-Protects in ischemia :

-Seems to be involved in inducing macrophage cleanup of beta-amyloid plaques :

-Since its primary role is in regulating bone deposition and ossification, and has secondary roles in immune signaling, then it should be of no surprise that it is a core mechanism of blood vessel calcification :

-This dataset doesn’t feature blood vessels, so I went looking for a gene expression study specific to atherosclerosis. Osteopontin is the most significantly upregulated gene :



EFEMP1 : EGF-containing fibulin-like extracellular matrix protein 1 (aka Fibulin 3 “FBLN3_HUMAN”)

-Has many EGF-like calcium binding domains

-Part of the Fibulin family : Fibulin 1 has 3 Anaphylatoxin domains, which are highly homologous to C3a, the product of the complement C3 gene when it is activated. These domains don’t show up in uniprot for EFEMP1, but there may still be an interplay here with the complement system.

-Fibulin-1 is present in atherosclerotic plaques and colocalizes with macrophages and foam cells :

LAPTM5 : lysosomal-associated protein transmembrane 5

-LAPTM5 Protein Is a Positive Regulator of Proinflammatory Signaling Pathways in Macrophages

“LAPTM5 facilitates activation of NF-κB and MAPK signaling and proinflammatory cytokine release mediated by cytokine and pattern recognition receptors.”







PTGES3 — Prostaglandin E synthase 3

-Binds to progesterone receptor and HSP90

-Prostaglandins are synthesized from arachidonic acid (AA), and several papers indicate that APOD’s job is the reduction of peroxidized AA as well as its enzymatic products.

-”Facilitates HIF alpha proteins hydroxylation via interaction with EGLN1/PHD2, leading to recruit EGLN1/PHD2 to the HSP90 pathway” …. See the section on collagen for a discussion on HIF!







SGK1 : Serine/Threonine protein kinase, regulated by glucocorticoids, and activated by insulin axis and MTORC2

-Genomic upregulation corresponds with cell death in parkinsons :

-Phosphorylates TAU protein, causing it to aggregate in alzheimers :

- Note, this isn’t the only thing claimed to phosphorylate TAU (see S100A6, also in this list)

TXNIP : Thioredoxin binding protein

-Thioredoxin : ROS scavenger and part of ROS signaling, mice overexpressing this live 35% longer :

-Thioredoxin interacts with COL1A1, in our underexpressed gene list :

“These data showed that the reducing status of TRX active site cysteine residues is important for the TRX–CP-pro α 1(1) interaction, indicating that collagen biosynthesis is under the regulation of TRX-dependent redox control.” — See the discussion below on collagen for why ROS may be relevant in collagen production

-Thioredoxin can directly activate or inhibit the complement system :

-”Thioredoxin-interacting protein mediates high glucose-induced reactive oxygen species generation by mitochondria and the NADPH oxidase, Nox4, in mesangial cells.” :

-TXNIP turns on NOX4, which produces ROS during ischemic stress

-Mice with TXNIP knocked out had far less collagen accumulation




- Part of the MHCI complex, along with B2M above. This is a really atypical HLA gene, with much confusion on what it does.

- This paper however gives an excellent clue that ties into a lot of aging pathologies : Express and protect yourself: the potential role of HLA-G on muscle cells and in inflammatory myopathies : “Muscle is the site or the target of immunologic injury in several diseases. Whereas under physiologic conditions muscle fibers are negative for major histocompatibility complex(MHC) class I antigens, these are upregulated under pathologic conditions, thus rendering muscle a possible target for the recognition by cytotoxic CD8 T cells. Cultured muscle cells are capable of presenting antigens to CD4 and CD8 T cells, further indicating that muscle fibers in vivo are critically involved in the initiating or perpetuating steps of inflammatory responses. The finding that muscle fibers in autoimmune inflammatory myopathies in vivoand cultured muscle cells in vitro express the nonclassical major histocompatibility complex molecule HLA-G raises several hypothesis concerning its possible pathophysiologic role. We review present knowledge on the functional consequences of muscle-related HLA-G and provide concepts of its relevance under pathologic conditions. We further speculate on the potential therapeutic implications of HLA-G that relate to special approaches such as myoblast transplantation or strategies against inflammatory aggression in general.”


PCSK6 : Proprotein convertase subtilisin/kexin type 6

- “The protein encoded by this gene belongs to the subtilisin-like proprotein convertase family. The members of this family are proprotein convertases that process latent precursor proteins into their biologically active products. This encoded protein is a calcium-dependent serine endoprotease that can cleave precursor protein at their paired basic amino acid processing sites. Some of its substrates are — transforming growth factor beta related proteins, proalbumin, and Willebrand factor (VMF in the overexpressed genes list).”

- PCSK6 deficiency causes salt-sensitive hypertension in mice : PCSK6-mediated corin activation is essential for normal blood pressure


Genes underexpressed with age, ranked:

TFRC : Transferrin receptor, important for iron uptake

- Iron causes oxidative damage (via fenton reactions)

- This paper dicusssing iron’s connection with alzheimers, and its effects on cholesterol metabolism :

- A role for iron deficiency in dopaminergic neurodegeneration :

COLLAGEN (3 genes) :

A note on collagen … MUCH of the collagen protein is made of modified proline residues (hydroxyproline). The enzyme that catalyzes this reaction requires vitamin C (ascorbic acid), an O2, and Fe as a cofactor…… otherwise you get scurvy matey!

Required reading … I didn’t know much of this!

Notable highlights:

  • A highly reactive Fe(Iv)=O species is produced … iron enters this story yet again!
  • Oxygen is required, maybe hypoxia is bad
  • This article mentions that Hypoxia-Inducible-Factor (HIF) is targeted for degradation by similar hydroxylase enzymes … effectively a sensor to switch off that response pathway. Since ischemic stress appears to be a major theme in the cardiovascular disease, there is an interesting connection here

I just learned this, and this is crazy, but here’s some gene numbers …. So collagen maybe


-”This is the most abundant collagen of the human body. It is present in scar tissue, the end product when tissue heals by repair. It is found in tendons, skin, artery walls, cornea, the endomysium surrounding muscle fibers, fibrocartilage, and the organic part of bones and teeth.”

-Mutations lead to osteogenesis imperfecta, weak bones and irregular connective tissue


- “This is the collagen of granulation tissue, and is produced quickly by young fibroblasts before the tougher type I collagen is synthesized. Reticular fiber. Also found in artery walls, skin, intestines and the uterus”

-Mutations lead to Ehlers Danlos syndrome, can lead to bursting blood vessels


-”Basal lamina; eye lens. Also serves as part of the filtration system in capillaries and the glomeruli of nephron in the kidney.”

FABP3 : Heart type fatty acid binding protein

-Released by heart cells after an acute ischemic episode :

- It is a highly sensitive of heart attacks showing just hours after pain starts — perhaps this indicates that aging is associated with progressive hypoxia throughout the circulatory system …. and/or mitochondrial dysfunction since its role is the transport of long-chain fatty acids to the mitochondria

ATP5G3 : subunit of ATP-synthase — this subunit encodes part of the proton channel

-This paper discusses “Housekeeping Genes” (HKGs) that are typically used to normalize gene expression experiments.

“The expression pattern of 10 commonly used HKG (namely ATP synthase (ATP5G3), β-2-microglobulin (B2M), ACTB, β-glucuronidase (GUSB), ophilin A (PPIA), GAPDH, hypoxanthine ribosyltransferase (HPRT1), phosphoglycerokinase (PGK1), TATA-box-binding protein (TBP) and transferrin receptor (TFRC)) was investigated by qRT-PCR in different cell lines cultured in various hypoxic or aerobic conditions.”

Thus, since these are housekeeping genes used in normalization. It might be that ATPG3 and TFRC are false positives. On the other hand, the whole point of this paper is discussing how hypoxia modulates these pathways

“Altogether, these results confirm that HKGs used for normalising gene expression in qRT-PCR may vary. Moreover, they emphasise that HKGs may dramatically fluctuate according to hypoxic conditions.”!

NDUFB11 : NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 11

- Part of complex 1 in the electron transport chain. Favorite gene of Francesca and I. Complex 1 is the primary site of ROS generation, and is also dependent on Iron-Sulphur clusters for electron transport activity. Metformin is thought to act at this enzyme, possibly for its life-extending effects.

-Muscle and Kidney, in both humans and mice, appear to be the tissues where it is underexpressed in aging.

- “Kohda et al. (2016) found that knockdown of the Ndufb11 gene in Drosophila resulted in significantly reduced lifespan, decreased metabolic rate, loss of mitochondrial complex I assembly, and increased lactate and pyruvate.”

-Mutations in this gene cause complex 1 deficiency, which I have. Yay

CX3CL1 : Fractalkine precursor

-This ligand attracts T-cells and monocytes when solubilized. When expressed in the membrane of an epithelial cell, binds and activates leukocytes :

-Binds to the chemokine receptor CX3CR1. In the brain, this pathway regulates movement and activation of microglia and affects severity of ALS progression

CA4 : Carbonic anhydrase 4 — The wikipedia article is fascinating … suggests that lack of carbon dioxide transport out of the lungs may be affected by age

-Carbonic anhydrases (CAs) are a large family of zinc metalloenzymes that catalyze the reversible hydration of carbon dioxide.

-”CA IV has been identified in pulmonary epithelium of many mammalian species and may be uniquely adaptive for gas exchange necessary for the high metabolic requirements of mammals. A majority of the CO2 produced by metabolism is transported as bicarbonate (HCO3-). At the tissue capillary, CO2 diffuses from tissue to plasma. Other forms of carbonic anhydrase enzyme are not present in the plasma, restricting the equilibrium reaction of CO2+H2O = H2CO3 = H+ HCO3-. CO2 in the plasma diffuses into the Red Blood Cell. CA is present within the Red Blood Cell, facilitating the conversion of CO2 to HCO3-. HCO3- so produced is transferred by the HCO3-/Cl- “shuttle” from the interior of the Red Blood Cell to the plasma. HCO3- doers not diffuse across cell membranes and, in the absence of CA, stays as HCO3- and concentrates in plasma. Up to 80% of metabolically produced CO2 is transported in plasma in the form of HCO3-. Blood moves from the tissue capillary to the pulmonary capillary where CO2 is exchanged at the lung.”

-This paper suggests that carbonic anhydrase 2 limits longevity :

-A google search for “carbonic anhydrase longevity” turns up MANY hits — lots of popscience ….

CALB1 : calbindin 1

-Has 6 EF-hand motifs that bind calcium. This gene is vitamin-D responsive, and is considered to be primarily responsible for regulating calcium uptake, particularly in the intensine

-Calcium seems central to a few things in this dataset. S100A proteins also include EF-hands, and EFEMP1

GHITM : Growth hormone-inducible transmembrane protein

-Localizes to inner mitochondrial membrane and ER. Significantly upregulated by growth hormone. Appears to inhibit apoptosis. This gene directly regulates the release of cytochrome C from the matrix space, which is an important initiator of the apoptotic cascade.

-This gene didn’t catch my attention until I saw this twitter thread today :

Highlights :

- Short people live longer (jerks!). Height is negatively associated with longevity, and this might a lot of why okinawans live so long!

-There is a rare genetic disorder called Laron’s syndrome, which causes dwarfism. People with this disease have a 0% cancer rate …. They are also protected against insulin resistance. (suggests better mito, I think)

- So, one possibility is that with excess growth hormone, apoptosis is inhibited. Turn off GH, (or better yet, target this gene with a drug), and cells will be more likely to enter apoptosis


“This gene encodes a ubiquinone-binding protein of low molecular mass. This protein is a small core-associated protein and a subunit of ubiquinol-cytochrome c reductase complex III, which is part of the mitochondrial respiratory chain.”

- The fact that this is ubiquinone binding is interesting. The cofactor binding site is certainly required for electron transport!

C5ORF13: Renamed NREP :

- Also known as PTZ17, this gene is regulated by Metalothionen (an overexpressed gene in this dataset). This gene requires metalothionen expression to be expressed :

- Regulates smooth muscle development. When expressed in myofibroblasts, it suppresses TGF-beta signaling and collagen 1+3 expression. The authors conclude that this gene is protective of fibrosis — too bad it’s underexpressed! :

ACCS2: Cytoplasmic acetyl-coa synthetase

-Makes acetyl-coa from acetate, required for histone acetylation and energy metabolism. Activated by the SREBP axis, which are transcription factors that upregulate lipid and cholesterol synthesis

- If this is underexpressed, it could be that gene expression is hindered, but it has been shown that ACCS2 translocates to the nucleus in hypoxic tumors :

- This gene controls HIF-2 signaling in hypoxic cells :

- “Biosynthesis of cholesterol generally takes place in the endoplasmic reticulum of hepatic cells and begins with acetyl- CoA, which is mainly derived from an oxidation reaction in the mitochondria. However, acetyl-CoA can also be derived from the cytoplasmic oxidation of ethanol by acetyl-CoA synthetase.”

- This suggests that if mitochondria are compromised, and ACCS2 expression is down, the cell is reliant on cholesterol delivery from LDL particles!

- Cholesterol synthesis is activated by SREBP-2, whereas SREBP-1 is more for denovo lipogenesis and storage of fat droplets. Interestingly, this paper shows that ROS exposure to cells induces SREBP-2 and not SREBP-1 expression :

- SREBP-2 downregulates the APOE receptor LPR1, but directly upregulates the LDL Receptor (LDLR) with sterol regulatory elements (SREs) in the sequence :

- ACCS2 is a direct target of SREBP-2, implying that when the cell needs cholesterol, its supposed to upregulate cytosolic synthesis of acetyl-coa….

UQCRFS1 : Ubiquinol-cytochrome c reductase, Rieske iron-sulfur polypeptide 1

- This is another subunit of mitochondrial complex 3 (along with UQCRQ). This subunit binds an iron-sulphur cluster that is required for electron transport.

DIABLO : SMAC (second mitochondria-derived activator of caspases)

-A pro-apoptotic factor

Genetics of Cardiovascular disease

Since so many of these aging genes in the expression data seem to resemble pathways in atherosclerosis … lets try and find connections with GWAS.



infinome is a citizen science experiment to help people lose weight, feel better, and live longer.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store

infinome is a citizen science experiment to help people lose weight, feel better, and live longer.