A new study links STING-mediated signals from marrow fat cells to inflammation and declining NAD⁺ levels in bone-forming cells. Interrupting this pathway helped preserve aging bone [1].
Making old bones
Our bones live a complex life, as they are constantly remodeled: cells called osteoclasts break down existing bone, while osteoblasts build new bone. Osteoblasts arise from bone marrow mesenchymal stem cells (BMSCs), which can also produce fat cells. With aging, bone marrow accumulates more fat, becomes more inflammatory, and supports bone formation less effectively [2]. However, scientists do not fully understand how this process works yet.
A new study by Chinese researchers, published in Advanced Science, focused on macrophages, a type of immune cell. In particular, they investigated cGAS–STING, an immune signaling pathway activated by DNA appearing in the cell’s cytoplasm outside its usual compartments. This mechanism helps detect viral or bacterial DNA, but with age, it starts to misfire, triggered by stimuli such as leaked mitochondrial DNA [3]. The enzyme cGAS detects this misplaced DNA and produces the messenger molecule cGAMP, which activates STING and downstream inflammatory signaling.
A STING operation
The researchers asked whether STING’s net effect changes with age, what activates it in aging marrow, and how activated macrophages affect neighboring bone-forming cells. First, they compared bone marrow from mice aged 3, 10, 18, and 24 months and found progressively greater activation of the cGAS–STING pathway. This included more DNA in the cytoplasm, more cGAMP, and increased phosphorylation (a sign of activation) of STING and downstream signaling proteins.
They then compared different marrow cell populations. Macrophages had the highest baseline expression of the Sting gene, and their STING activation increased with age. This established macrophages as a plausible intervention target, but it was still unclear whether their STING activity was in itself harmful or a response to other damage.
The authors used a macrophage-specific viral vector to silence Sting from 12 to 18 months of age in mice. This significantly preserved bone density and structure. It also increased measures of new bone deposition and reduced marrow fat-cell numbers.
To understand what in aging marrow activates macrophage STING, the researchers investigated interactions between cells. Old macrophages impaired the ability of young BMSCs to develop into bone-forming cells, whereas old macrophages lacking STING were less harmful.
Next, the team examined potential upstream triggers. Exposure to bacterial lipopolysaccharide (LPS), a component of bacterial membranes, activated STING in macrophages, as expected. Old mice also had more endotoxin in blood and marrow, which can contribute to increased STING levels.
Inflammation and mitochondrial damage, combined
The authors then focused on marrow fat cells. Secretions from old marrow fat cells activated macrophage STING both in culture and when injected into young mice’s marrow. Co-culturing old fat cells with young macrophages produced a similar response.
44 secreted proteins elevated in bone marrow with age. The researchers singled out one of them: DPP4. Of all the tested marrow cell types, fat cells released substantially more soluble DPP4, and old fat cells released more than young ones.
Injecting DPP4 into young mice’s marrow for four weeks worsened bone density and architecture and activated macrophage STING. Conversely, reducing Dpp4 expression in marrow fat cells significantly improved one measure of bone structure health (spacing between trabeculae).
However, DPP4 alone did not significantly activate STING but, rather, strengthened its activation when combined with old marrow fluid. Thus, DPP4 appeared to contribute to an inflammatory environment rather than act as a sufficient trigger by itself.
The authors tested several possible receptors for DPP4 and focused on TLR4, a cell-surface receptor involved in innate immune responses. Blocking TLR4 reduced STING activation, and protein-interaction experiments supported direct binding between DPP4 and TLR4.
Importantly, old macrophages contained more mitochondrial DNA outside their mitochondria. Reducing mitochondrial DNA with low-dose ethidium bromide partially reduced STING activation. Old macrophages also showed more signs of mitochondrial stress.
The researchers suggest that external inflammatory signals, including DPP4-associated TLR4 signaling, “cooperate” with internal mitochondrial damage to sustain STING activity.
The NAD⁺ connection
The researchers also found that macrophage NAD⁺ levels declined with age, while NAD⁺-degrading activity increased. Experimentally activating the pathway lowered macrophage NAD⁺ but suppressing cGAS or STING counteracted these changes. Further experiments suggested that STING signaling increases production of the enzyme CD38, which degrades NAD⁺.
How could an enzyme in macrophages reduce NAD⁺ inside BMSCs? Further experiments suggested that CD38 that sits on the cell surface degrades extracellular NAD⁺ and its precursor NMN, potentially limiting the supply available to nearby cells.
Old BMSCs had impaired mitochondrial function and lower production of the cellular energy currency ATP. NMN supplementation partially rescued these metrics but to a reduced degree in old or artificially activated macrophages. Suppressing cGAS or STING in the macrophages seemed to work as well: macrophages from old mice with Sting or Cd38 knockout had better mitochondrial respiration.
Finally, the team tested a pharmacological intervention: STING inhibitor H-151 combined with NMN supplementation in mice from 12 to 18 months of age. The combination preserved bone density and structure, increased new bone deposition, reduced inflammatory markers, and improved NAD⁺ levels and mitochondrial function in BMSCs.
The researchers used alendronate, an established anti-osteoporosis drug, as a positive control. Improvements in bone structure in the alendronate group were largely similar to the H-151/NMN group.
Among the study’s limitations are its small size and combining two treatments in the intervention experiment. The latter makes it hard to disentangle the effects of STING inhibition from those of NMN supplementation. More research is needed to investigate these interesting findings further.
Literature
[1] Li, Z., Gao, K., Wang, M., Liang, S., Li, D., Tan, G., Xue, H., Liu, Y., Zhang, P., Xu, Z., Cheng, W., Wang, Y., & Li, N. (2026). Bone marrow macrophage STING signaling is a direct nexus linking marrow adipose expansion to NAD⁺ depletion and osteogenic decline in skeletal aging. Advanced Science.
[2] Aaron, N., Costa, S., Rosen, C. J., & Qiang, L. (2022). The implications of bone marrow adipose tissue on inflammaging. Frontiers in Endocrinology, 13, 853765.
[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
