Targeted inCITE-Seq Analysis Identifies the Loss of Nuclear TDP-43 in Endothelium as a Mediator of Blood Brain Barrier Signaling Pathway Dysfunction in Neurodegeneration.

Despite the importance of the endothelium in the regulation of the blood brain barrier (BBB) in aging and neurodegenerative disease, difficulties in extracting endothelial cell (EC) nuclei have limited analysis of these cells. In addition, nearly all Amyotrophic Lateral Sclerosis (ALS) and Frontotemporal Degeneration (FTD), and a large portion of Alzheimer's Disease (AD) exhibit neuronal TDP-43 aggregation, leading to loss of nuclear function, but whether TDP-43 is similarly altered in human BBB ECs is unknown. Here we utilize a novel technique for the enrichment of endothelial and microglial nuclei from human cortical brain tissues, combined with inCITE-seq, to analyze nuclear proteins and RNA transcripts in a large cohort of healthy and diseased donors. Our findings reveal a unique transcriptional signature in nearly half of the capillary endothelial cells across neurodegenerative states, characterized by reduced levels of nuclear ß-Catenin and canonical downstream genes, and an increase in TNF/NF-kB target genes. We demonstrate that this does not correlate with increased nuclear p65/NF-kB, but rather a specific loss of nuclear TDP-43 in these disease associated ECs. Comparative analysis in animal models with targeted disruption of TDP-43 shows that this is sufficient to drive these transcriptional alterations. This work reveals that TDP-43 is a critical governor of the transcriptional output from nuclear p65/NF-kB, which has paradoxical roles in barrier maintenance and also barrier compromising inflammatory responses, and suggests that disease specific loss in ECs contributes to BBB defects observed in the progression of AD, ALS and FTD.

Loss of Endothelial TDP-43 Leads to Blood Brain Barrier Defects in Mouse Models of Amyotrophic Lateral Sclerosis and Frontotemporal Dementia.

Loss of nuclear TDP-43 occurs in a wide range of neurodegenerative diseases, and specific mutations in the TARDBP gene that encodes the protein are linked to familial Frontal Temporal Lobar Dementia (FTD), and Amyotrophic Lateral Sclerosis (ALS). Although the focus has been on neuronal cell dysfunction caused by TDP-43 variants, TARDBP mRNA transcripts are expressed at similar levels in brain endothelial cells (ECs). Since increased permeability across the blood brain barrier (BBB) precedes cognitive decline, we postulated that altered functions of TDP-43 in ECs contributes to BBB dysfunction in neurodegenerative disease. To test this hypothesis, we examined EC function and BBB properties in mice with either knock-in mutations found in ALS/FTLD patients (TARDBPG348C and GRNR493X) or EC-specific deletion of TDP-43 throughout the endothelium (Cdh5(PAC)CreERT2; Tardbpff) or restricted to brain endothelium (Slco1c1(BAC)CreERT2; Tardbpff). We found that TARDBPG348C mice exhibited increased permeability to 3kDa Texas Red dextran and NHS-biotin, relative to their littermate controls, which could be recapitulated in cultured brain ECs from these mice. Nuclear levels of TDP-43 were reduced in vitro and in vivo in ECs from TARDBPG348C mice. This coincided with a reduction in junctional proteins VE-cadherin, claudin-5 and ZO-1 in isolated ECs, supporting a cell autonomous effect on barrier function through a loss of nuclear TDP-43. We further examined two models of Tardbp deletion in ECs, and found that the loss of TDP-43 throughout the endothelium led to systemic endothelial activation and permeability. Deletion specifically within the brain endothelium acutely increased BBB permeability, and eventually led to hallmarks of FTD, including fibrin deposition, microglial and astrocyte activation, and behavioral defects. Together, these data show that TDP-43 dysfunction specifically within brain ECs would contribute to the BBB defects observed early in the progression of ALS/FTLD.

Splice factor polypyrimidine tract-binding protein 1 (Ptbp1) primes endothelial inflammation in atherogenic disturbed flow conditions.

NF-κB-mediated endothelial activation drives leukocyte recruitment and atherosclerosis, in part through adhesion molecules Icam1 and Vcam1. The endothelium is primed for cytokine activation of NF-κB by exposure to low and disturbed blood flow (LDF)but the molecular underpinnings are not fully understood. In an experimental in vivo model of LDF, platelets were required for the increased expression of several RNA-binding splice factors, including polypyrimidine tract binding protein (Ptbp1). This was coordinated with changes in RNA splicing in the NF-κB pathway in primed cells, leading us to examine splice factors as mediators of priming. Using Icam1 and Vcam1 induction by tumor necrosis factor (TNF)-α stimulation as a readout, we performed a CRISPR Cas9 knockout screen and identified a requirement for Ptbp1 in priming. Deletion of Ptbp1 had no effect on cell growth or response to apoptotic stimuli, but reversed LDF splicing patterns and inhibited NF-κB nuclear translocation and transcriptional activation of downstream targets, including Icam1 and Vcam1. In human coronary arteries, elevated PTBP1 correlates with expression of TNF pathway genes and plaque. In vivo, endothelial-specific deletion of Ptbp1 reduced Icam1 expression and myeloid cell infiltration at regions of LDF in atherosclerotic mice, limiting atherosclerosis. This may be mediated, in part, by allowing inclusion of a conserved alternative exon in Ripk1 leading to a reduction in Ripk1 protein. Our data show that Ptbp1, which is induced in a subset of the endothelium by platelet recruitment at regions of LDF, is required for priming of the endothelium for subsequent NF-κB activation, myeloid cell recruitment and atherosclerosis.

A method for rapid flow-cytometric isolation of endothelial nuclei and RNA from archived frozen brain tissue.

Endothelial cells are important contributors to brain development, physiology, and disease. Although RNA sequencing has contributed to the understanding of brain endothelial cell diversity, bulk analysis and single-cell approaches have relied on fresh tissue digestion protocols for the isolation of single endothelial cells and flow cytometry-based sorting on surface markers or transgene expression. These approaches are limited in the analysis of the endothelium in human brain tissues, where fresh samples are difficult to obtain. Here, we developed an approach to examine endothelial RNA expression by using an endothelial-specific marker to isolate nuclei from abundant archived frozen brain tissues. We show that this approach rapidly and reliably extracts endothelial nuclei from frozen mouse brain samples, and importantly, from archived frozen human brain tissues. Furthermore, isolated RNA transcript levels are closely correlated with expression in whole cells from tissue digestion protocols and are enriched in endothelial markers and depleted of markers of other brain cell types. As high-quality RNA transcripts could be obtained from as few as 100 nuclei in archived frozen human brain tissues, we predict that this approach should be useful for both bulk analysis of endothelial RNA transcripts in human brain tissues as well as single-cell analysis of endothelial sub-populations.

Identification of splice regulators of fibronectin-EIIIA and EIIIB by direct measurement of exon usage in a flow-cytometry based CRISPR screen.

The extracellular matrix protein fibronectin (FN) is alternatively spliced in a variety of inflammatory conditions, resulting in increased inclusion of alternative exons EIIIA and EIIIB. Inclusion of these exons affects fibril formation, fibrosis, and inflammation. To define upstream regulators of alternative splicing in FN, we have developed an in vitro flow-cytometry based assay, using RNA-binding probes to determine alternative exon inclusion level in aortic endothelial cells. This approach allows us to detect exon inclusion in the primary transcripts themselves, rather than in surrogate splicing reporters. We validated this assay in cells with and without FN-EIIIA and -EIIIB expression. In a small-scale CRISPR KO screen of candidate regulatory splice factors, we successfully detected known regulators of EIIIA and EIIIB splicing, and detected several novel regulators. Finally, we show the potential in this approach to broadly interrogate upstream signaling pathways in aortic endothelial cells with a genome-wide CRISPR-KO screen, implicating the TNFalpha and RIG-I-like signaling pathways and genes involved in the regulation of fibrotic responses. Thus, we provide a novel means to screen the regulation of splicing of endogenous transcripts, and predict novel pathways in the regulation of FN-EIIIA inclusion.

Alternative Splicing of FN (Fibronectin) Regulates the Composition of the Arterial Wall Under Low Flow.

OBJECTIVE: Exposure of the arterial endothelium to low and disturbed flow is a risk factor for the erosion and rupture of atherosclerotic plaques and aneurysms. Circulating and locally produced proteins are known to contribute to an altered composition of the extracellular matrix at the site of lesions, and to contribute to inflammatory processes within the lesions. We have previously shown that alternative splicing of FN (fibronectin) protects against flow-induced hemorrhage. However, the impact of alternative splicing of FN on extracellular matrix composition remains unknown. Approach and Results: Here, we perform quantitative proteomic analysis of the matrisome of murine carotid arteries in mice deficient in the production of FN splice isoforms containing alternative exons EIIIA and EIIIB (FN-EIIIAB null) after exposure to low and disturbed flow in vivo. We also examine serum-derived and endothelial-cell contributions to the matrisome in a simplified in vitro system. We found flow-induced differences in the carotid artery matrisome that were impaired in FN-EIIIAB null mice. One of the most interesting differences was reduced recruitment of FBLN1 (fibulin-1), abundant in blood and not locally produced in the intima. This defect was validated in our in vitro assay, where FBLN1 recruitment from serum was impaired by the absence of these alternatively spliced segments. CONCLUSIONS: Our results reveal the extent of the dynamic alterations in the matrisome in the acute response to low and disturbed flow and show how changes in the splicing of FN, a common response in vascular inflammation and remodeling, can affect matrix composition.

Splenic function is not maintained long-term after partial splenectomy in children with sickle cell disease.

BACKGROUND: Partial splenectomy (PS) may allow preservation of splenic function in cases where splenectomy is indicated for hematologic diseases; however, the long-term outcomes are uncertain. We investigated the long-term outcomes of PS in patients with sickle cell disease (SCD). METHODS: A single-institution retrospective chart review was performed for children with SCD who underwent PS from 1997 to 2017. For comparison, we reviewed outcomes for patients who underwent PS for hereditary spherocytosis (HS). The primary endpoint was viability of the splenic remnant as inferred by the presence of remnant perfusion on ultrasound and/or liver spleen scan. RESULTS: Nine patients with SCD and 26 patients with HS underwent PS at a median age of 11 (IQR, 9-14) and 7.5 (IQR, 6-13) years, respectively. All underwent laparoscopic PS with three (7.9%) conversions to open. Two SCD patients were lost to long-term follow-up. The remaining seven SCD patients had initial postoperative splenic remnant perfusion demonstrated by ultrasonography. By 42 months postoperatively, however, none had a functioning splenic remnant. The median time to loss of splenic remnant was 12.6 (IQR 9.2-28.5) months. In contrast, all HS patients demonstrated robust splenic remnant blood flow with a median follow-up of 46 (IQR 37-82) months. CONCLUSION: No patient with SCD who underwent PS had viable splenic tissue for more than 42 months, likely due to continued autoinfarction typical of patients with this disease. Therefore, we believe that PS to preserve splenic function is not indicated in patients with SCD. LEVEL OF EVIDENCE: III.

Validating an opioid prescribing algorithm in post-operative pediatric surgical oncology patients.

PURPOSE: We developed an algorithm to decrease opioid prescriptions for pediatric oncology patients at discharge following surgery, based on a retrospective analysis to decrease variability and over-prescribing. The aim of this study was to prospectively test the algorithm. METHODS: Opioid-naïve patients undergoing surgery for tumor resection at a single institution were included. A prescribing algorithm was developed based on surgical approach, day of discharge, and inpatient opioid use. Prospectively collected data included outpatient opioid consumption and patient/family satisfaction. Total home dose prescribed was equal to that used in the 8 or 24 h, depending on length of stay and operative approach, prior to discharge, divided into 0.15 mg/kg doses. RESULTS: The algorithm was used in 121 patients and correctly predicted outpatient opioid requirements for 102 patients (84.3%). For 15 (12.4%) patients, the algorithm over-estimated opioid need by an average of 0.38 OME/kg. Four (3.3%) patients required additional opioids. Using this algorithm, we decreased overall opioid prescriptions from 6.17 to 0.21 OME/kg (p < 0.001), and all but one patient/family reported being satisfied with post-operative pain control. CONCLUSION: Using an algorithm based on inpatient opioid use, outpatient opioid needs can be accurately predicted, thereby reducing excess opioid prescriptions without detriment to patient satisfaction. TYPE OF STUDY: Prospective Quality Initiative Study. LEVEL OF EVIDENCE: Level III.

A Quality Improvement Intervention to Decrease Postoperative Opioid Prescriptions in Pediatric Oncology Patients.

PURPOSE: This quality improvement initiative aimed to minimize opioid prescribing after oncologic pediatric surgery. METHODS: Retrospective surgical data collected at a pediatric cancer hospital from July 2016 to June 2018 included hospitalization details, oral morphine equivalents prescribed, unplanned visits/calls because of pain, and parental/patient satisfaction with pain control. The quality improvement initiative promoted opioid prescription at discharge on the basis of prior inpatient requirements and education regarding nonopioid analgesia. Upon commencing this project in July 2018, we collected data prospectively. RESULTS: The retrospective and the prospective cohorts included 271 and 99 patients, respectively. Mean (SD) oral morphine equivalents (mg/kg) prescribed upon discharge was significantly reduced in the prospective (0.75±1.34) versus retrospective cohorts (5.48±6.94, P<0.001). The unplanned visits/calls regarding pain were 23 (retrospective, 8.5%) and 2 (prospective, 2.0%). In total, 44 patients (44.4%) received an opioid prescription at discharge in the prospective cohort, significantly fewer than retrospective cohort (251, 92.6%, P<0.001), and used a mean of 34.3 of 159.8 (21.5%) doses dispensed. Length of stay was comparable (P=0.88) between cohorts. Prospective satisfaction rate was 96.2%, leaving 3 patients (3.8%) not satisfied with their pain control regimen. CONCLUSIONS: Dramatic reduction of opioid prescriptions after oncologic surgery can be achieved without detriment to patient satisfaction or readmissions. LEVEL OF EVIDENCE: Level V.

Alternative RNA splicing in the endothelium mediated in part by Rbfox2 regulates the arterial response to low flow.

Low and disturbed blood flow drives the progression of arterial diseases including atherosclerosis and aneurysms. The endothelial response to flow and its interactions with recruited platelets and leukocytes determine disease progression. Here, we report widespread changes in alternative splicing of pre-mRNA in the flow-activated murine arterial endothelium in vivo. Alternative splicing was suppressed by depletion of platelets and macrophages recruited to the arterial endothelium under low and disturbed flow. Binding motifs for the Rbfox-family are enriched adjacent to many of the regulated exons. Endothelial deletion of Rbfox2, the only family member expressed in arterial endothelium, suppresses a subset of the changes in transcription and RNA splicing induced by low flow. Our data reveal an alternative splicing program activated by Rbfox2 in the endothelium on recruitment of platelets and macrophages and demonstrate its relevance in transcriptional responses during flow-driven vascular inflammation.

Effects of adenotonsillectomy on polysomnographic parameters in children with sickle cell disease.

Obstructive sleep apnea (OSA) in the pediatric sickle cell disease (SCD) population can promote nightly hemoglobin oxygen desaturation, which increases the risk of central nervous system insult and may impair cognitive function. Adenotonsillectomy can ameliorate OSA symptoms, but its effect in children with SCD has not been fully investigated. We reviewed the effects of adenotonsillectomy in thirteen children with SCD by comparison of pre and post-adenotonsillectomy polysomnography (PSG) parameters. Significant reduction in hemoglobin oxygen desaturation, decreased apnea-hypopnea index, and increased rapid eye movement sleep occurred after adenotonsillectomy. Adenotonsillectomy promotes improvement in sleep quality in children with SCD and PSG-confirmed OSA.

Hydroxycarbamide alters erythroid gene expression in children with sickle cell anaemia.

Sickle cell anaemia (SCA) is a severe debilitating haematological disorder associated with a high degree of morbidity and mortality. The level of fetal haemoglobin (HbF) is well-recognized as a critical laboratory parameter: lower HbF is associated with a higher risk of vaso-occlusive complications, organ damage, and early death. Hydroxycarbamide treatment can induce HbF, improve laboratory parameters, and ameliorate clinical complications of SCA but its mechanisms of action remain incompletely defined and the HbF response is highly variable. To identify pathways of hydroxycarbamide activity, we performed microarray expression analyses of early reticulocyte RNA obtained from children with SCA enrolled in the HydroxyUrea Study of Long-term Effects (NCT00305175) and examined the effects of hydroxycarbamide exposure in vivo. Hydroxycarbamide affected a large number of erythroid genes, with significant decreases in the expression of genes involved in translation, ribosome assembly and chromosome organization, presumably reflecting the daily cytotoxic pulses of hydroxycarbamide. Hydroxycarbamide also affected expression of numerous genes associated with HbF including BCL11A, a key regulator of baseline HbF levels. Together, these data indicate that hydroxycarbamide treatment for SCA leads to substantial changes in erythroid gene expression, including BCL11A and other potential signalling pathways associated with HbF induction.

Pharmacokinetics, pharmacodynamics, and pharmacogenetics of hydroxyurea treatment for children with sickle cell anemia.

Hydroxyurea therapy has proven laboratory and clinical efficacies for children with sickle cell anemia (SCA). When administered at maximum tolerated dose (MTD), hydroxyurea increases fetal hemoglobin (HbF) to levels ranging from 10% to 40%. However, interpatient variability of percentage of HbF (%HbF) response is high, MTD itself is variable, and accurate predictors of hydroxyurea responses do not currently exist. HUSTLE (NCT00305175) was designed to provide first-dose pharmacokinetics (PK) data for children with SCA initiating hydroxyurea therapy, to investigate pharmacodynamics (PD) parameters, including HbF response and MTD after standardized dose escalation, and to evaluate pharmacogenetics influences on PK and PD parameters. For 87 children with first-dose PK studies, substantial interpatient variability was observed, plus a novel oral absorption phenotype (rapid or slow) that influenced serum hydroxyurea levels and total hydroxyurea exposure. PD responses in 174 subjects were robust and similar to previous cohorts; %HbF at MTD was best predicted by 5 variables, including baseline %HbF, whereas MTD was best predicted by 5 variables, including serum creatinine. Pharmacogenetics analysis showed single nucleotide polymorphisms influencing baseline %HbF, including 5 within BCL11A, but none influencing MTD %HbF or dose. Accurate prediction of hydroxyurea treatment responses for SCA remains a worthy but elusive goal.

Genetic predictors for stroke in children with sickle cell anemia.

Stroke is a devastating complication of sickle cell anemia (SCA), affecting 5% to 10% of patients before adulthood. Several candidate genetic polymorphisms have been proposed to affect stroke risk, but few have been validated, mainly because previous studies were hampered by relatively small sample sizes and the absence of additional patient cohorts for validation testing. To verify the accuracy of proposed genetic modifiers influencing stroke risk in SCA, we performed genotyping for 38 published single nucleotide polymorphisms (SNPs), as well as α-thalassemia, G6PD A(-) variant deficiency, and ß-globin haplotype in 2 cohorts of children with well-defined stroke phenotypes (130 stroke, 103 nonstroke). Five polymorphisms had significant influence (P < .05): SNPs in the ANXA2, TGFBR3, and TEK genes were associated with increased stroke risk, whereas α-thalassemia and a SNP in the ADCY9 gene were linked with decreased stroke risk. Further investigation at these genetic regions may help define mutations that confer stroke risk or protection in children with SCA.