Alzheimer's Disease (AD) is characterized by the accumulation of extracellular amyloid-β (Aβ) as well as CNS and systemic inflammation. Microglia, the myeloid cells resident in the CNS, use microRNAs to rapidly respond to inflammatory signals. MicroRNAs (miRNAs) modulate inflammatory responses in microglia, and miRNA profiles are altered in Alzheimer's disease (AD) patients. Expression of the pro-inflammatory miRNA, miR-155, is increased in the AD brain. However, the role of miR-155 in AD pathogenesis is not well-understood. We hypothesized that miR-155 participates in AD pathophysiology by regulating microglia internalization and degradation of Aβ. We used CX3CR1CreER/+ to drive-inducible, microglia-specific deletion of floxed miR-155 alleles in two AD mouse models. Microglia-specific inducible deletion of miR-155 in microglia increased anti-inflammatory gene expression while reducing insoluble Aβ1-42 and plaque area. Yet, microglia-specific miR-155 deletion led to early-onset hyperexcitability, recurring spontaneous seizures, and seizure-related mortality. The mechanism behind hyperexcitability involved microglia-mediated synaptic pruning as miR-155 deletion altered microglia internalization of synaptic material. These data identify miR-155 as a novel modulator of microglia Aβ internalization and synaptic pruning, influencing synaptic homeostasis in the setting of AD pathology.

Microglia maintain brain health and play important roles in disease and injury. Despite the known ability of microglia to proliferate, the precise nature of the population or populations capable of generating new microglia in the adult brain remains controversial. We identified Prominin-1 (Prom1; also known as CD133) as a putative cell surface marker of committed brain myeloid progenitor cells. We demonstrate that Prom1-expressing cells isolated from mixed cortical cultures will generate new microglia in vitro. To determine whether Prom1-expressing cells generate new microglia in vivo, we used tamoxifen inducible fate mapping in male and female mice. Induction of Cre recombinase activity at 10 weeks in Prom1-expressing cells leads to the expression of TdTomato in all Prom1-expressing progenitors and newly generated daughter cells. We observed a population of new TdTomato-expressing microglia at 6 months of age that increased in size at 9 months. When microglia proliferation was induced using a transient ischemia/reperfusion paradigm, little proliferation from the Prom1-expressing progenitors was observed with the majority of new microglia derived from Prom1-negative cells. Together, these findings reveal that Prom1-expressing myeloid progenitor cells contribute to the generation of new microglia both in vitro and in vivo. Furthermore, these findings demonstrate the existence of an undifferentiated myeloid progenitor population in the adult mouse brain that expresses Prom1. We conclude that Prom1-expressing myeloid progenitors contribute to new microglia genesis in the uninjured brain but not in response to ischemia/reperfusion. SIGNIFICANCE STATEMENT Microglia, the innate immune cells of the CNS, can divide to slowly generate new microglia throughout life. Newly generated microglia may influence inflammatory responses to injury or neurodegeneration. However, the origins of the new microglia in the brain have been controversial. Our research demonstrates that some newly born microglia in a healthy brain are derived from cells that express the stem cell marker Prominin-1. This is the first time Prominin-1 cells are shown to generate microglia.

Background The microvasculature (MV) of brains with Alzheimer’s disease neuropathologic change (ADNC) and cerebral amyloid angiopathy (CAA), in the absence of concurrent pathologies (e.g., infarctions, Lewy bodies), is incompletely understood. Objective To analyze microvascular density, diameter and extracellular matrix (ECM) content in association with ADNC and CAA. Methods We examined samples of cerebral cortex and isolated brain microvasculature (MV) from subjects with the National Institute on Aging-Alzheimer's Association (NIA-AA) designations of not-, intermediate-, or high ADNC and from subjects with no CAA and moderate-severe CAA. Cases for all groups were selected with no major (territorial) strokes, ≤ 1 microinfarct in screening sections, and no Lewy body pathology. MV density and diameter were measured from cortical brain sections. Levels of basement membrane (BM) ECM components, the protein product of TNF-stimulated gene-6 (TSG-6), and the ubiquitous glycosaminoglycan hyaluronan (HA) were assayed by western blots or HA ELISA of MV lysates. Results We found no significant changes in MV density or diameter among any of the groups. Levels of BM laminin and collagen IV (col IV) were lower in MV isolated from the high ADNC vs. not-ADNC groups. In contrast, BM laminin was significantly higher in MV from the moderate-severe CAA vs. the no CAA groups. TSG-6 and HA content were higher in the presence of both high ADNC and CAA, whereas levels of BM fibronectin and perlecan were similar among all groups. Conclusions Cortical MV density and diameter are not appreciably altered by ADNC or CAA. TSG-6 and HA are increased in both ADNC and CAA, with laminin and col IV decreased in the BM of high ADNC, but laminin increased in moderate-severe CAA. These results show that changes in the ECM occur in AD and CAA, but independently of one another, and likely reflect on the regional functioning of the brain microvasculature.

Tissue microstructure modeling of diffusion MRI signal is an active research area striving to bridge the gap between macroscopic MRI resolution and cellular-level tissue architecture. Such modeling in neuronal tissue relies on a number of assumptions about the microstructural features of axonal fiber bundles, such as the axonal shape (e.g., perfect cylinders) and the fiber orientation dispersion. However, these assumptions have not yet been validated by sufficiently high-resolution 3-dimensional histology. Here, we reconstructed sequential scanning electron microscopy images in mouse brain corpus callosum, and introduced a random-walker (RaW)-based algorithm to rapidly segment individual intra-axonal spaces and myelin sheaths of myelinated axons. Confirmed by a segmentation based on human annotations initiated with conventional machine-learning-based carving, our semi-automatic algorithm is reliable and less time-consuming. Based on the segmentation, we calculated MRI-relevant estimates of size-related parameters (inner axonal diameter, its distribution, along-axon variation, and myelin g -ratio), and orientation-related parameters (fiber orientation distribution and its rotational invariants; dispersion angle). The reported dispersion angle is consistent with previous 2-dimensional histology studies and diffusion MRI measurements, while the reported diameter exceeds those in other mouse brain studies. Furthermore, we calculated how these quantities would evolve in actual diffusion MRI experiments as a function of diffusion time, thereby providing a coarse-graining window on the microstructure, and showed that the orientation-related metrics have negligible diffusion time-dependence over clinical and pre-clinical diffusion time ranges. However, the MRI-measured inner axonal diameters, dominated by the widest cross sections, effectively decrease with diffusion time by ~ 17% due to the coarse-graining over axonal caliber variations. Furthermore, our 3 d measurement showed that there is significant variation of the diameter along the axon. Hence, fiber orientation dispersion estimated from MRI should be relatively stable, while the “apparent” inner axonal diameters are sensitive to experimental settings, and cannot be modeled by perfectly cylindrical axons.

Little is known about the extracellular matrix (ECM) during progression of AD pathology. Brain ECM is abundant in hyaluronan (HA), a non-sulfated glycosaminoglycan synthesized by HA synthases (HAS) 1-3 in a high molecular weight (MW) form that is degraded into lower MW fragments. We hypothesized that pathologic severity of AD is associated with increases in HA and HA-associated ECM molecules. To test this hypothesis, we assessed HA accumulation and size; HA synthases (HAS) 1-3; and the HA-stabilizing hyaladherin, TSG-6 in parietal cortex samples from autopsied research subjects with not AD (CERAD = 0, Braak = 0- II, n = 12-21), intermediate AD (CERAD = 2, Braak = III-IV, n = 13-18), and high AD (CERAD = 3, Braak = V-VI, n = 32-40) neuropathologic change. By histochemistry, HA was associated with deposits of amyloid and tau, and was also found diffusely in brain parenchyma, with overall HA quantity (measured by ELSA) significantly greater in brains with high AD neuropathology. Mean HA MW was similar among the samples. HAS2 and TSG-6 mRNA expression, and TSG-6 protein levels were significantly increased in high AD and both molecules were present in vasculature, NeuN-positive neurons, and Iba1-positive microglia. These results did not change when accounting for gender, advanced age (≥ 90 years versus <90 years), or the clinical diagnosis of dementia. Collectively, our results indicate a positive correlation between HA accumulation and AD neuropathology, and suggest a possible role for HA synthesis and metabolism in AD progression.

To model the diffusion MRI signal in brain white matter, general assumptions have been made about the microstructural properties of axonal fiber bundles, such as the axonal shape and the fiber orientation dispersion. In particular, axons are modeled by perfectly circular cylinders with no diameter variation within each axon, and their directions obey a specific orientation distribution. However, these assumptions have not been validated by histology in 3-dimensional high-resolution neural tissue. Here, we reconstructed sequential scanning electron microscopy images in mouse brain corpus callosum, and introduced a semi-automatic random-walker (RaW) based algorithm to rapidly segment individual intra-axonal spaces and myelin sheaths of myelinated axons. Confirmed with a conventional machine-learning-based interactive segmentation method, our semi-automatic algorithm is reliable and less time-consuming. Based on the segmentation, we calculated histological estimates of size-related (e.g., inner axonal diameter, g-ratio) and orientation-related (e.g., Fiber orientation distribution and its rotational invariants, dispersion angle) quantities, and simulated how these quantities would be observed in actual diffusion MRI experiments by considering diffusion time-dependence. The reported dispersion angle is consistent with previous 2-dimensional histology studies and diffusion MRI measurements, though the reported diameter is larger than those in other mouse brain studies. Our results show that the orientation-related metrics have negligible diffusion time-dependence; however, inner axonal diameters demonstrate a non-trivial time-dependence at diffusion times typical for clinical and preclinical use. In other words, the fiber dispersion estimated by diffusion MRI modeling is relatively independent, while the "apparent" axonal size estimated by axonal diameter mapping potentially depends on experimental MRI settings.

The brain changes in volume and composition with normal aging. Cellular components of the brain are supported by an extracellular matrix (ECM) comprised largely of hyaluronan (HA) and HA-associated members of the lectican family of chondroitin sulfate proteoglycans (CSPGs). We examined regional differences in microvascular density, neuronal and glial markers, and accumulation of HA and CSPGs in mouse brains during normal aging. The cortex, hippocampus, dentate gyrus, and cerebellum of young (4 months), middle-aged (14 months), and aged (24-26 months) brains were analyzed. Microvascular density decreased in cerebral cortex and cerebellum with age. There were no detectable differences in neuronal density. There was an increase in astrocytes in the hippocampus with aging. HA accumulation was higher in aged brain relative to young brain in the cerebral cortex and cerebellum, but not in other regions examined. In contrast, CSPGs did not change with aging in any of the brain regions examined. HA and CSPGs colocalized with a subset of neuronal cell bodies and astrocytes, and at the microvasculature. Differences in accumulation of ECM in the aging brain, in the setting of decreased microvascular density and/or increased glial activation, might contribute to age-related regional differences in vulnerability to injury and ischemia.