Matrix metalloproteinase-3: a prerequisite for structural and functional plasticity in the visual cortex of the mouse

The brain is able to change its functional and anatomical organization in responseto environmental changes and this ability is termed neuroplasticity. The visualcortex, the area in the brain that receives visual input, is undoubtedly one of thebest and most studied areas of the brain for understanding cortical plasticity anddevelopment. However, the molecular mechanisms governing cortical plasticityare still elusive. Matrix metalloproteinases (MMPs) are Zn2+-dependentendopeptidases considered to be essential for normal brain development andneuroplasticity by modulating extracellular matrix proteins, receptors, adhesionmolecules, growth factors and cytoskeletal proteins. Specifically, MMP-3 hasrecently been implicated in synaptic plasticity, hippocampus-dependent learningand neuronal development and migration in the cerebellum. However, thefunction of this enzyme in the neocortex is understudied. Therefore, weexplored the phenotypical characteristics of the neuronal architecture andthe capacity for experience-dependent cortical plasticity in the visual cortexof adult MMP-3-deficient (MMP-3-/-) mice. Golgi Cox stainings revealed asignificant reduction in apical dendritic length, spine length and an increasednumber of apical obliques for layer V pyramidal neurons in the visual cortex ofadult MMP-3-/- mice compared to wildtype (WT) animals. To assess the effectof MMP-3 deficiency on cortical plasticity, we monocularly enucleated (ME)adult MMP-3-/- mice and analyzed the reactivation of the contralateral visualcortex seven weeks post-ME. In contrast to previous results in C57Bl/6J adultmice, activity remained confined to the binocular zone and did not expandinto the monocular regions indicative for an aberrant open-eye potentiation.Permanent hypoactivity in the monocular cortex lateral and medial to V1 also indicated a lack of cross-modal plasticity. These observations demonstratethat genetic inactivation of MMP-3 has profound effects on the structuralintegrity and plasticity response of the visual cortex of adult mice. To furtherassess the molecular changes governing the MMP-3-/- cortical phenotype, weperformed Western analysis on different neurofilament protein (NF) subunitsand collapsin response mediator proteins (CRMP). The former are markers fora healthy cytoskeleton and determine the shape and architecture of neuronswhereas CRMPs are mainly involved in regulating neurite outgrowth throughmicrotubule polymerization. A significant upregulation of both phosphorylatedand non-phosphorylated NF-high, phosphorylated NF-medium, NF-low andalfa-internexin was detected in the visual cortex of MMP-3-/- mice comparedwith WT. These results suggest that an altered stoichiometry of NF subunits isrelated to the truncated neuronal architecture observed in MMP-3-/- mice. Inaddition, the expression level of CRMP-5 was significantly elevated in MMP-3-/-samples whereas the levels of CRMP-1, -2, -3 and -4 did not differ compared withWT mice. Recent literature indicates that overexpression of CRMP-5 negativelyregulates dendritic outgrowth by reducing the number of mitochondria throughincreased autophagy and mitophagy. Therefore, we assesed the expression levelof a marker for autophagy, the microtubule-associated protein 1 light chain 3(LC3-II) and found a significant increase in MMP-3-/- samples. The expressionlevels of mitochondrial fusion protein mitofusin-2 (Mfn-2) and dynamin-relatedprotein 1 (Drp1), a mitochondrial fission protein, did not differ, indicating thatthe mitochondrial fusion and fission balance is not altered in the visual cortexof MMP-3-/- mice. Taken together, an altered NF composition, overexpressionof CRMP-5 and a possible upregulation of autophagy underlie the structuralphenotype in the visual cortex of MMP-3-/- mice. To assess the acute role of MMPs in ME-induced visual cortex plasticitywe performed Western analysis for MMP-3 to reveal fluctuations in MMP-3expression level associated with post-ME survival time in P45 and P120 mice.This approach revealed significant differences in proMMP-3 expression especiallyin P45 mice both in the medial monocular and binocular zone whereas in P120mice, enucleation did not induce large effects on proMMP-3 expression. Thissuggests that MMP-3 is potentially necessary to inhibit cross-modal plasticity as normally seen in P45 mice. However, no active form of MMP-3 was visibleusing Western analysis and therefore we tried to optimize a fluorescent activityassay to study MMP-3 proteolysis in visual cortex samples. However, whenperforming this assay on samples of MMP-3-/- mice, a fluorescent signal in thesame range as P45 and P120 ME C57Bl/6J samples was measured, rendering thistechnique unreliable. To extend the analysis of ME-induced cortical plasticityto other MMPs, we focused on the gelatinases (MMP-2 and MMP-9) becausetheir function in synaptic plasticity is well established. Using a combination ofWestern analysis and gelatin zymography we revealed no significant differences inMMP-2 expression in all conditions studied, suggesting that this enzyme does notplay a major role in ME-induced open-eye potentiation or cross-modal plasticity.No suitable antibodies for MMP-9 were available and gelatin zymography didnot reveal detectable levels of MMP-9 proteolytic activity, possibly due tosample preparation issues. To conclude, further research is needed to evaluatethe activity of different MMPs in the mouse visual cortex and this will providea basis for future experiments for pharmacological intervention to establish acausal relationship between MMP function and ME-induced cortical plasticity.nrpages: 186status: publishe