A new study suggests that aging or Parkinson’s-triggering mutations create inflammation in peripheral tissues, and then circulating extracellular vesicles spread it to the brain, which might contribute to the disease [1].
Brain on fire
Nature made the brain remarkably well-protected, including from the elements by the skull and from pathogens by the blood-brain barrier (BBB). The brain was also long considered to be immune-privileged – that is, limiting local immune responses to reduce swelling and damage from inflammation; another example is the eye.
However, more recent studies have shown that some brain cells, particularly microglia, can act as immune cells, and brain inflammation exists and probably drives brain aging [2]. What has been largely unknown is whether systemic inflammation, such as inflammaging, affects brain inflammation, and if yes, what pathways are involved?

Read More
One well-characterized molecular engine of sterile inflammation is the cGAS-STING pathway, which senses DNA in the cell’s cytosol. Normally, it detects foreign DNA, such as from viruses and bacteria. The problem is, in aging and senescence, a cell’s own DNA – from a damaged nucleus or leaky mitochondria – can spill into the cytosol. cGAS can’t distinguish self from non-self and fires anyway, producing a smoldering, chronic interferon type I (IFN-I) response [3].
Exporting inflammation
A new study by an international collective of scientists, published in Cell Reports, asks whether cGAS-STING-driven systemic inflammation can be a factor in Parkinson’s disease, and what mechanisms might be involved. Leucine-rich repeat kinase 2 (LRRK2 in humans, Lrrk2 in mice) regulates the cell’s degradation and recycling machinery (endolysosomal system). The most common genetic cause of Parkinson’s is the G2019S mutation in the associated gene, which increases the enzyme’s activity. The authors refer to this gain-of-function mutation as LRRK2GoF.
The authors’ central hypothesis is that LRRK2GoF accelerates aging by degrading endolysosomal function, which causes self-DNA to accumulate in the cytosol, instead of being promptly recycled, and to be exported in DNA-carrying extracellular vesicles (EVs). Those EVs activate cGAS-STING not just in the original cell but also in distant cells – including, eventually, the brain.
First, the researchers took plasma and cerebrospinal fluid (CSF) samples from young healthy donors, aged healthy donors, and Parkinson’s patients. The latter showed elevated systemic IFN-I activity but not NF-κB activity, which belongs to a more general inflammation pathway.
At the cellular level, blood monocytes from Parkinson’s patients showed elevated IFNB1 transcripts, but an LRRK2 inhibitor normalized that IFNB1 elevation back to healthy levels, pointing to LRRK2’s causal involvement. However, the sample sizes in this and several other experiments were small (n=3-4).
To test causality, the team moved to a G2019S knock-in mouse model. Compared to controls, these Lrrk2GoF mice had a markedly elevated, age-dependent IFN-I signature across plasma, monocytes, bone marrow, and spleen. RNA sequencing of spleen showed elevated inflammatory and senescence-associated gene expression.
The researchers then asked whether the brain also becomes inflamed and whether behavior is affected. In aged Lrrk2GoF mice, neurons and microglia showed increased IFN-I and interferon-stimulated genes (ISGs) as well as a pronounced age-dependent motor decline. Importantly, the mutant mice also had increased BBB permeability and smaller brains. A leaky BBB might explain how peripheral inflammatory signals reach the brain.
Using RNA sequencing at different time points, the team found that the peripheral IFN-I signature was already present at 3 months, whereas the brain IFN-I signature and locomotor decline did not appear until 12 months. The author’s interpretation, crucial for the entire paper, is that inflammation starts peripherally and reaches the brain later.
The tiny Trojan horses
Mechanistically, IFN-I has triggers other than cGAS/STING. However, the researchers determined that eliminating those other triggers had no effect, while deleting STING completely reversed the elevated IFN-I response to wild-type levels in both splenocytes and microglia, reflecting changes in both a peripheral tissue and the brain. STING deletion also reduced microglial inflammation markers and protected the mice from the motor decline.
Lrrk2GoF mice lost 51% of dopaminergic neurons with age, a major hallmark of Parkinson’s, versus about 30% in wild-type mice, and STING deletion prevented that loss. Changes in other neuronal populations were inconsistent.
Moving back in vitro to hunt for additional mechanistic insights, the authors showed that Lrrk2GoF mouse fibroblasts reached senescence earlier than their wild-type counterparts and had elevated IFN-I activity, which could be normalized by inhibiting Lrrk2. Intriguingly, in transwell co-culturing (which blocks direct cell contact but allows diffusible/vesicular signals), senescent fibroblasts evoked a STING-dependent IFN-I response in physically separated macrophages.
Defective endolysosomal clearance is known to cause increased EV secretion, which is exactly what the researchers found in both Lrrk2GoF and naturally aged fibroblasts. Again, this effect was abrogated by blocking Lrrk2. EVs taken from Lrrk2GoF fibroblasts contained more genomic and mitochondrial DNA and induced STING-dependent IFN-I response in recipient macrophages.
The team then confirmed the EV mechanism in vivo and in humans. In Lrrk2GoF mice, EV accumulation in plasma appeared early, but in the cerebrospinal fluid (CSF), it started much later, consistent with the “peripheral inflammation slowly causes brain inflammation” hypothesis. In humans, Parkinson’s patient macrophages showed reduced endolysosomal degradation. Plasma and CSF from Parkinson’s patients had more DNA-containing EVs, and EVs derived from these patients induced STING-dependent IFN-I, demonstrating a clean mouse-to-human bridge.
Literature
[1] Öberg, M., Myers, C., Saffarzadeh, N., Maric, I., Murillo-León, M., Strömberg, A., … & Härtlova, A. (2026). STING-dependent peripheral inflammaging drives neurodegeneration via extracellular vesicles. Cell Reports, 45(7).
[2] Yin, F., Sancheti, H., Patil, I., & Cadenas, E. (2016). Energy metabolism and inflammation in brain aging and Alzheimer’s disease. Free Radical Biology and Medicine, 100, 108-122.
[3] Gulen, M. F., Samson, N., Keller, A., Schwabenland, M., Liu, C., Glück, S., … & Ablasser, A. (2023). cGAS–STING drives ageing-related inflammation and neurodegeneration. Nature, 620(7973), 374-380.
View the article at lifespan.io














