Enzymatic remodeling of PNNs
Perineuronal nets (PNNs) and their specific components undergo constant remodeling once they are assembled. This process of remodeling PNNs and, more broadly, the ECM is largely performed by numerous metalloproteinases, enzymes that proteolytically cleave their components, most notably lecticans. In this way, metalloproteinases regulate PNN abundance by degrading specific components and, in some cases, lead to a complete disassembly of the PNN structure. The most studied metalloproteinase classes in relation to PNN biology are matrix metalloproteinases (MMPs) and disintegrases (ADAMTSs), which collectively are able to break down all elements of the ECM and PNNs.
These enzymes are produced by multiple cell types, including neurons and glia, although the specific cellular contributions have not been well studied. Because their activity is precisely controlled through differential expression, proteolytic activation and binding of their inhibitors, such as the tissue inhibitors of MMPs, they are able to spatially and temporally control the ECM and PNNs in response to experience and disease. MMPs belong to a family of zinc-binding endopeptidases and usually act in soluble forms located in the extracellular space. They are produced in an inactive zymogen form and require proteolytic activation.
There are 23 members of this family in humans, although only a select few have been researched in regards to PNN component degradation. MMPs are released in response to experiences that increase neural activity, such as learning and memory or exposure to an enriched environment and presumably “loosen” the ECM to promote plasticity. MMPs cleave numerous ECM constituents, resulting in a modulation of the perisynaptic ECM that consequently induces synaptic changes. Indeed, MMPs regulate both excitatory and inhibitory neuronal transmission in multiple ways. Reducing or inhibiting several MMPs (i.e., MMP 3, 7, and 9) also impacts cognitive function, causing impairments in fear and spatial memory.
While many MMP family members lack extensive study in the PNN context, research has demonstrated MMP9’s ability to influence PNN abundance. Genetic reduction or pharmacological inhibition of MMP9 increases PNNs, while higher levels of MMP9 correlate with lower PNNs. MMP9 has also been shown to be essential for PNN modulation driven by environmental enrichment. Stamenković et al. (2017) observed that while mice exhibit a decrease in PNNs after environmental enrichment, MMP9 KO mice show an increase in PNN levels. This means that MMPs are important regulators of PNN abundance as well as regulating PNN remodeling in response to more plastic states.
MMP dysregulation has been reported in many different diseases involving PNN abnormalities and likely contributes to disease pathology. It has been shown that MMP activity mediates PNN loss in a glioma-induced model of epilepsy and reducing this activity increases peritumoral PNNs. Moreover, MMP3 and MMP13 have been correlated to aggrecan and brevican PNN component degradation in HIV-infected brains, suggesting that they may play a role in the remodeling of these specific components following viral stimuli. MMPs have also been implicated in PNN degradation associated with behavioral and cognitive abnormalities. Knockout of MMP9 in Fmr1 KO mice prevents PNN degradation and ameliorates their behavioral abnormalities.
PNNs in Alzheimer development
One of the most devastating moments for family members of a patient with Alzheimer’s is when their loved one forgets who they are. New University of Virginia School of Medicine research may explain why that happens and could lead to a way to prevent it. UVA’s Harald Sontheimer, PhD, and graduate student Lata Chaunsali and their colleagues found that the failure to recognize family, friends and caregivers is caused by the breakdown of protective “nets”, the PNNs, that surround neurons in the brain. Preventing the loss of these nets in lab mice protected the mice from losing their memories of previous social interactions.
These net-like structures surround brain neurons and perform a critical barrier function that lets neurons communicate correctly. These communications allow the neurons to form and store new memories.Based on those results, Sontheimer and his collaborators suspected that disruptions of the nets could be a critical turning point in Alzheimer disease. Their latest work bears that out: scientists found that lab mice that had faulty nets lost their ability to remember other mice – their “social memory” – even as they could still form new memories of objects in their environment. This mirrors what is seen in people with Alzheimer’s, where social memory often fails before object memory.
Sontheimer and his team then used “MMP inhibitors” – a class of drugs already being investigated for their potential to treat cancer and arthritis – to see if they could prevent the loss of the perineuronal nets. Inhibition of MMP activity with GM6001 prevented PNN disruption and protects against social memory deficits in the 5XFAD dementia mouse model. In Alzheimer’s disease, people have trouble remembering their family and friends due to the loss of a memory known as social memory. Scientists found that the net-like coating PNNs protects these social memories. In their research with mice, when they kept these brain structures safe early in life, the mice suffering from this disease were better at remembering their social interactions.
These results were replicated by another science team at the Manipal School of Life Sciences, India, where the researchers employed an accelerated mouse model of aging to elucidate the causal link between ECM dynamics and recognition memory during aging. Aged mice exhibited impaired social and non-social recognition memory, accompanied by increased intensity of perineuronal nets in the hippocampal dorsal CA2 (dCA2). A reduction in the power of theta oscillations (3-7 Hz) in the dCA2 of aged mice was also observed. Notably, selective degradation of PNNs in the dCA2 using chondroitinase ABC (ChABC) rescued recognition memory deficits and restored theta oscillations.
Healthy aging alone can lead to cognitive decline, decreased brain size, protein aggregation, accumulation of senescent cells and neuroinflammation. Furthermore, age is the primary risk factor for several neurodegenerative disorders such as Parkinson and Alzheimer. Age-related neuroinflammation, as known as inflammaging, is thought to restrict brain plasticity. The team found also that aged striatal microglia displayed an activated morphology with larger cell bodies and reduced branching, as well as increased expression of markers for microgliosis, like TREM2 and CD68. The changes the scientists saw in the mice’s brains align with those seen in human patients with Alzheimer’s, suggesting that the targeting the nets in people could provide similar benefits.
- Edited by Dr. Gianfrancesco Cormaci, PhD, specialist in Clinical. Biochemistry.
Scientific references
Colon ZA et al. J Neuroinflammation. 2025 Nov 18; 22(1):274.
Chaunsali L et al. Alzheimers Dement. 2025 Oct; 21(10):e70813.
Mehak SF, Shivakumar AB et al. Aging Cell. 2025; 24(9):e70139.
Cheung SW et al. Neuropathol Appl Neurobiol. 2024; 50(3):e12982.
