3D imaging and reconstruction of cortical perineuronal nets in focal cerebral ischemia and mild hypoperfusion

Perineuronal nets are predominantly expressed around the fast spiking parvalbumin expressing interneurons. In accordance with previous observations, the significant loss of cortical perineuronal nets (PNNs) was observed in the ipsilesional hemisphere following temporary occlusion of middle cerebral artery (MCAO). Importantly, the reduced expression of PNNs following MCAO and hypoperfusion was not associated with neuronal cell death in the ipsilesional cortex, but correlated with reactive gliosis in the ipsilesional striatum. However, the expression of PNNs was not completely abolished following ischemic stroke. Two major components of PNNs were immunolabeled using anti-aggrecan antibody (detecting the core protein of aggrecan, one of the most abundant proteoglycans within PNNs) and Wisteria floribunda agglutinin (WFA), binds to the disaccharide LacdiNAc chondroitin sulfate chains of CSPGs within the proteoglycans of PNNs. The ultrastructural analysis of PNN morphology was performed using superresolution structured illumination imaging, which revealed the subtle organization of these meshes. At all depths of the sample, imaging resolution comprised at least 200 nm in spatial directions and 400 nm in axial direction, as indicated by sub-resolution microsphere size analysis. To provide the quantitative analysis of the 3D organization of PNNs, a novel graph computation algorithm was developed, provided as MatLab code accompanying this article. The topological analysis of 3D meshes of single PNNs was conducted separately for aggrecan and WFA-labelled nets and the four key metrics were derived: i) percentage of nodes with non-zero degree; ii) maximal internode distance, at which 95% of the nodes have non-zero degree (this metric will be further abbreviated as R95); iii) average internode distance; iv) average degree of a node.

细胞周围网状结构（Perineuronal nets, PNNs）主要表达于快速放电的小白蛋白阳性中间神经元周围。与既往研究观察结果一致，短暂性大脑中动脉闭塞（middle cerebral artery occlusion, MCAO）后，病变同侧半球的皮层细胞周围网状结构出现显著丢失。值得关注的是，MCAO及脑低灌注后PNNs的表达下调，与病变同侧皮层的神经元死亡无显著关联，而是与病变同侧纹状体的反应性胶质增生呈显著相关。不过，缺血性脑卒中后PNNs的表达并未完全被抑制。本研究采用两种标记物对PNNs的两种主要组分进行免疫标记：其一为抗聚集蛋白聚糖抗体（anti-aggrecan antibody），用于检测聚集蛋白聚糖的核心蛋白——聚集蛋白聚糖是PNNs中丰度最高的蛋白聚糖之一；其二为紫藤凝集素（Wisteria floribunda agglutinin, WFA），其可特异性结合PNNs所含硫酸软骨素蛋白聚糖（CSPGs）的二糖LacdiNAc硫酸软骨素链。采用超分辨率结构照明显微成像技术对PNNs的形态开展超微结构分析，揭示了此类三维网状结构的精细组织模式。经亚分辨率微球尺寸分析验证，该成像系统在样本所有深度下，空间分辨率可达至少200 nm，轴向分辨率可达400 nm。为实现PNNs三维组织形式的定量分析，本研究开发了一种新型图计算算法，并以随文附带的MatLab代码形式提供。针对聚集蛋白聚糖标记与WFA标记的单个PNNs三维网状结构，我们分别开展了拓扑分析，并推导得到四项核心量化指标：① 非零度节点占比；② 95%节点具有非零度时的最大节点间距（该指标后续将简写为R95）；③ 平均节点间距；④ 节点平均度。