Chromogranin A is a prohormone, a large protein that cells snip into several smaller working peptides. One of those fragments, catestatin, runs low in the brains of people who died with three different dementias. A study published in Molecular Therapy ↗ reports that putting catestatin back reduced the signature protein damage of those diseases in cultured neurons and in mice.
The human tissue came first. Measuring autopsy brain, the group found reduced catestatin and a matching rise in pancreastatin, another fragment cut from the same parent protein, in the hippocampus and cortex of Alzheimer's brains, in the frontal cortex in corticobasal degeneration, and in the basal ganglia in progressive supranuclear palsy. All three are tauopathies, meaning the tau protein misfolds and clumps into tangles inside neurons.
Then came the supplementation experiments. In cortical neuron cultures and in organotypic slice cultures, which are thin living brain slices kept alive in a dish, adding catestatin lowered both the phosphorylation and the aggregation of tau, the two steps that turn a normal structural protein into a pathological tangle. In PS19 mice, a strain engineered to build human tau tangles as they age, catestatin injections decreased pathological tau, calmed gliosis, which is the inflammatory activation of the brain's support cells, and improved performance on memory tests. In a separate mouse model called 5xFAD, which builds amyloid plaques rather than tau, the same peptide cut plaque burden and neuroinflammation.
The mechanism the authors propose is indirect, and it is worth stating plainly. Catestatin lowered epinephrine, the hormone better known as adrenaline, in both mouse models, and it suppressed protein kinase A, an enzyme that runs hot under adrenergic stress signaling. The chain they draw runs like this. Too little catestatin lets adrenaline and protein kinase A stay high, and that overactive stress axis feeds tau pathology. Restore the peptide, quiet the axis, and the damage eases. It is a metabolic-stress explanation for a disease usually framed in terms of protein folding, which is part of why it is interesting.
Now the caveats, because they are large. This is cells and mice, not people. The human data are correlational, a snapshot of what autopsy brains looked like, and they cannot say whether low catestatin drives the disease or simply tracks it. The catestatin here is an endogenous peptide the body already makes, raised experimentally in animals, not a dosed drug with a safety file, a proven route across the blood-brain barrier, or a single human trial behind it. The work comes from one research group at the University of California, San Diego, and a version was posted as a preprint before peer review. Correlational human tissue paired with interventional mouse experiments is a legitimate and common way to build a mechanism, but it is a starting point, not a treatment.
What makes it worth logging is the bridge it draws. A peptide carved from a metabolic prohormone, acting through adrenaline and a stress-response enzyme, changing the amount of tau and amyloid a brain accumulates, is a different kind of target than the antibodies and aggregation blockers that dominate dementia research. The adrenergic angle is suggestive, since drugs that damp adrenaline signaling already exist in cardiology, but this paper does not test any of them against tau. It tests the missing peptide, and reports that the brain does better when the peptide is there.