Studying Human Brain Inflammation in Leptomeningeal and Choroid Plexus Explant Cultures.

The meninges (dura, pia and arachnoid) are critical membranes encasing and protecting the brain within the skull. The leptomeninges, which comprise the arachnoid and pia, have many functions beyond brain protection including roles in neurogenesis, fibrotic scar formation and brain inflammation. Similarly, the choroid plexus plays important roles in normal brain function but is also involved in brain inflammation. We have begun studying the role of human leptomeninges and choroid plexus in brain inflammation and leptomeninges in fibrotic scar formation, using human brain derived explant cultures. To study the composition of the cells generated in these explants we undertook immunocytochemical characterisation. Cells, mainly pericytes and meningeal macrophages, emerge from leptomeningeal explants (LME's) and respond to inflammatory mediators by producing inflammatory molecules. LME-derived cells also respond to mechanical injury and cytokines, providing an in vitro human brain model of fibrotic scar formation. Choroid plexus explants (CPE's) generate epithelial cells, pericytes and microglia/macrophages. CPE-derived cells also respond to inflammatory mediators. LME and CPE explants survive and generate cells for many months in vitro and provide a remarkable opportunity to study basic mechanisms of human brain inflammation and fibrosis and to test human-active anti-inflammatory and anti-scarring treatments.

A role for human brain pericytes in neuroinflammation.

BACKGROUND: Brain inflammation plays a key role in neurological disease. Although much research has been conducted investigating inflammatory events in animal models, potential differences in human brain versus rodent models makes it imperative that we also study these phenomena in human cells and tissue. METHODS: Primary human brain cell cultures were generated from biopsy tissue of patients undergoing surgery for drug-resistant epilepsy. Cells were treated with pro-inflammatory compounds IFNγ, TNFα, IL-1ß, and LPS, and chemokines IP-10 and MCP-1 were measured by immunocytochemistry, western blot, and qRT-PCR. Microarray analysis was also performed on late passage cultures treated with vehicle or IFNγ and IL-1ß. RESULTS: Early passage human brain cell cultures were a mixture of microglia, astrocytes, fibroblasts and pericytes. Later passage cultures contained proliferating fibroblasts and pericytes only. Under basal culture conditions all cell types showed cytoplasmic NFκB indicating that they were in a non-activated state. Expression of IP-10 and MCP-1 were significantly increased in response to pro-inflammatory stimuli. The two chemokines were expressed in mixed cultures as well as cultures of fibroblasts and pericytes only. The expression of IP-10 and MCP-1 were regulated at the mRNA and protein level, and both were secreted into cell culture media. NFκB nuclear translocation was also detected in response to pro-inflammatory cues (except IFNγ) in all cell types. Microarray analysis of brain pericytes also revealed widespread changes in gene expression in response to the combination of IFNγ and IL-1ß treatment including interleukins, chemokines, cellular adhesion molecules and much more. CONCLUSIONS: Adult human brain cells are sensitive to cytokine challenge. As expected 'classical' brain immune cells, such as microglia and astrocytes, responded to cytokine challenge but of even more interest, brain pericytes also responded to such challenge with a rich repertoire of gene expression. Immune activation of brain pericytes may play an important role in communicating inflammatory signals to and within the brain interior and may also be involved in blood brain barrier (BBB) disruption . Targeting brain pericytes, as well as microglia and astrocytes, may provide novel opportunities for reducing brain inflammation and maintaining BBB function and brain homeostasis in human brain disease.

A rapid and sensitive method to detect siRNA-mediated mRNA cleavage in vivo using 5' RACE and a molecular beacon probe.

Specific detection of mRNA cleavage by 5'RACE is the only method to confirm the knockdown of mRNA by RNA interference, but is rarely reported for in vivo studies. We have combined 5'-RNA-linker-mediated RACE (5'-RLM-RACE) with real-time PCR using a molecular beacon to develop a rapid and specific method termed MBRACE, which we have used to detect small-interfering RNA (siRNA)-induced cleavage of ApoB, RRM1 and YBX1 transcripts in vitro, and ApoB in vivo. When RNA from siRNA-transfected cells was used for 5'-RLM-RACE and a cleavage site-specific molecular beacon probe was included in subsequent real-time PCR analysis, the specific mRNA cleavage product was detected. Detection of siRNA-mediated cleavage was also observed when RNA from mouse liver following administration of ApoB-specific siRNA was analysed, even in cases where ApoB knockdown measured by real-time PCR was <10%. With its sensitivity and specificity, this variation on the 5'RACE method should prove a useful tool to detect mRNA cleavage and corroborate knockdown studies following siRNA use in vivo.

Analyzing pericytes under mild traumatic brain injury using 3D cultures and dielectric elastomer actuators.

Traumatic brain injury (TBI) is defined as brain damage due to an external force that negatively impacts brain function. Up to 90% of all TBI are considered in the mild severity range (mTBI) but there is still no therapeutic solution available. Therefore, further understanding of the mTBI pathology is required. To assist with this understanding, we developed a cell injury device (CID) based on a dielectric elastomer actuator (DEA), which is capable of modeling mTBI via injuring cultured cells with mechanical stretching. Our injury model is the first to use patient-derived brain pericyte cells, which are ubiquitous cells in the brain involved in injury response. Pericytes were cultured in our CIDs and mechanically strained up to 40%, and by at least 20%, prior to gene expression analysis. Our injury model is a platform capable of culturing and stretching primary human brain pericytes. The heterogeneous response in gene expression changes in our result may suggest that the genes implicated in pathological changes after mTBI could be a patient-dependent response, but requires further validation. The results of this study demonstrate that our CID is a suitable tool for simulating mTBI as an in vitro stretch injury model, that is sensitive enough to induce responses from primary human brain pericytes due to mechanical impacts.

Routine culture and study of adult human brain cells from neurosurgical specimens.

When modeling disease in the laboratory, it is important to use clinically relevant models. Patient-derived human brain cells grown in vitro to study and test potential treatments provide such a model. Here, we present simple, highly reproducible coordinated procedures that can be used to routinely culture most cell types found in the human brain from single neurosurgically excised brain specimens. The cell types that can be cultured include dissociated cultures of neurons, astrocytes, microglia, pericytes and brain endothelial and neural precursor cells, as well as explant cultures of the leptomeninges, cortical slice cultures and brain tumor cells. The initial setup of cultures takes ~2 h, and the cells are ready for further experiments within days to weeks. The resulting cells can be studied as purified or mixed population cultures, slice cultures and explant-derived cultures. This protocol therefore enables the investigation of human brain cells to facilitate translation of neuroscience research to the clinic.

Cardiac glycosides target barrier inflammation of the vasculature, meninges and choroid plexus.

Neuroinflammation is a key component of virtually all neurodegenerative diseases, preceding neuronal loss and associating directly with cognitive impairment. Neuroinflammatory signals can originate and be amplified at barrier tissues such as brain vasculature, surrounding meninges and the choroid plexus. We designed a high content screening system to target inflammation in human brain-derived cells of the blood-brain barrier (pericytes and endothelial cells) to identify inflammatory modifiers. Screening an FDA-approved drug library we identify digoxin and lanatoside C, members of the cardiac glycoside family, as inflammatory-modulating drugs that work in blood-brain barrier cells. An ex vivo assay of leptomeningeal and choroid plexus explants confirm that these drugs maintain their function in 3D cultures of brain border tissues. These results suggest that cardiac glycosides may be useful in targeting inflammation at border regions of the brain and offer new options for drug discovery approaches for neuroinflammatory driven degeneration.

Cultured pericytes from human brain show phenotypic and functional differences associated with differential CD90 expression.

The human brain is a highly vascular organ in which the blood-brain barrier (BBB) tightly regulates molecules entering the brain. Pericytes are an integral cell type of the BBB, regulating vascular integrity, neuroinflammation, angiogenesis and wound repair. Despite their importance, identifying pericytes amongst other perivascular cell types and deciphering their specific role in the neurovasculature remains a challenge. Using primary adult human brain cultures and fluorescent-activated cell sorting, we identified two CD73(+)CD45(-) mesenchymal populations that showed either high or low CD90 expression. CD90 is known to be present on neurons in the brain and peripheral blood vessels. We found in the human brain, that CD90 immunostaining localised to the neurovasculature and often associated with pericytes. In vitro, CD90(+) cells exhibited higher basal proliferation, lower expression of markers αSMA and CD140b, produced less extracellular matrix (ECM) proteins, and exhibited lesser pro-inflammatory responses when compared to the CD90(-) population. Thus, CD90 distinguishes two interrelated, yet functionally distinct pericyte populations in the adult human brain that may have discrete roles in neurovascular function, immune response and scar formation.

Assessing fibrinogen extravasation into Alzheimer's disease brain using high-content screening of brain tissue microarrays.

BACKGROUND: Tissue microarrays are commonly used to evaluate disease pathology however methods to automate and quantify pathological changes are limited. NEW METHOD: This article demonstrates the utility of the VSlide scanner (MetaSystems) for automated image acquisition from immunolabelled tissue microarray slides, and subsequent automated image analysis with MetaXpress (Molecular Devices) software to obtain objective, efficient and reproducible data from immunolabelled tissue microarray sections. RESULTS: Significant increases in fibrinogen immunolabelling were observed in 29 Alzheimer's disease cases compared to 28 control cases analysed from a single tissue microarray slide. Western blot analysis also demonstrated significant increases in fibrinogen immunolabelling in 6 Alzheimer's cases compared to 6 control cases. The observed changes were also validated with gold standard blinded manual H-scoring. COMPARISON WITH EXISTING METHOD: VSlide Metafer software offers a 'tissue microarray acquisition' plugin for easy mapping of tissue cores with their original position on the tissue microarray map. High resolution VSlide images are compatible with MetaXpress image analysis software. This article details the coupling of these two technologies to accurately and reproducibly analyse immunolabelled tissue microarrays within minutes, compared to the gold standard method of manual counting using H-scores which is significantly slower and prone to inter-observer variation. CONCLUSIONS: Here, we couple brain tissue microarray technology with high-content screening and automated image analysis as a powerful way to address bottle necks in data generation and improve throughput, as well as sensitivity to study biological/pathological changes in brain disease.

Adult human glia, pericytes and meningeal fibroblasts respond similarly to IFNy but not to TGFß1 or M-CSF.

The chemokine Interferon gamma-induced protein 10 (IP-10) and human leukocyte antigen (HLA) are widely used indicators of glial activation and neuroinflammation and are up-regulated in many brain disorders. These inflammatory mediators have been widely studied in rodent models of brain disorders, but less work has been undertaken using human brain cells. In this study we investigate the regulation of HLA and IP-10, as well as other cytokines and chemokines, in microglia, astrocytes, pericytes, and meningeal fibroblasts derived from biopsy and autopsy adult human brain, using immunocytochemistry and a Cytometric Bead Array. Interferonγ (IFNγ) increased microglial HLA expression, but contrary to data in rodents, the anti-inflammatory cytokine transforming growth factor ß1 (TGFß1) did not inhibit this increase in HLA, nor did TGFß1 affect basal microglial HLA expression or IFNγ-induced astrocytic HLA expression. In contrast, IFNγ-induced and basal microglial HLA expression, but not IFNγ-induced astrocytic HLA expression, were strongly inhibited by macrophage colony stimulating factor (M-CSF). IFNγ also strongly induced HLA expression in pericytes and meningeal fibroblasts, which do not basally express HLA, and this induction was completely blocked by TGFß1, but not affected by M-CSF. In contrast, TGFß1 did not block the IFNγ-induced increase in IP-10 in pericytes and meningeal fibroblasts. These results show that IFNγ, TGFß1 and M-CSF have species- and cell type-specific effects on human brain cells that may have implications for their roles in adult human brain inflammation.

A novel method of organotypic brain slice culture using connexin-specific antisense oligodeoxynucleotides to improve neuronal survival.

Organotypic slice cultures obtained from immature brain tissue represent a well-established model system for neuroscience research. Current culture methods, however, do not allow long-term culture of mature brain slices. Slice cultures from mature animals would provide an in vitro experimental environment suitable for investigation of neuropathologies, which in human, predominate in aged individuals. We hypothesized that damage, incurred by slicing of the brain, is propagated through intercellular connexin43 (Cx43) gap junction channels and that this damage is not easily repaired in mature central nervous system (CNS) tissue that lacks the pluripotency of immature tissue. We investigated the role of Cx43 gap junctions in long-term survival of mature brain tissue using antisense oligodeoxynucleotide (AsODN) technology. The application of Cx43 AsODN immediately after slicing of the mature brain led to a significant but transient knockdown of Cx43 protein. This treatment was associated with the long-term survival of hippocampal neurons with normal morphology within whole brain slices taken from 14 and 40-day-old adult rats.