Computational model of brain endothelial cell signaling pathways predicts therapeutic targets for cerebral pathologies

https://doi.org/10.1016/j.yjmcc.2021.11.005Get rights and content

Highlights

  • Cerebrovascular endothelial cells play a central role in many CNS pathologies.

  • A computational model of cerebrovascular endothelial signaling was implemented.

  • Highly influential molecular signaling nodes were identified for brain endothelium.

  • Therapeutic targets were predicted for stroke, glioma, and Alzheimer's disease.

Abstract

Brain endothelial cells serve many critical homeostatic functions. In addition to sensing and regulating blood flow, they maintain blood-brain barrier function, including precise control of nutrient exchange and efflux of xenobiotics. Many signaling pathways in brain endothelial cells have been implicated in both health and disease; however, our understanding of how these signaling pathways functionally integrate is limited. A model capable of integrating these signaling pathways could both advance our understanding of brain endothelial cell signaling networks and potentially identify promising molecular targets for endothelial cell-based drug or gene therapies. To this end, we developed a large-scale computational model, wherein brain endothelial cell signaling pathways were reconstructed from the literature and converted into a network of logic-based differential equations. The model integrates 63 nodes (including proteins, mRNA, small molecules, and cell phenotypes) and 82 reactions connecting these nodes. Specifically, our model combines signaling pathways relating to VEGF-A, BDNF, NGF, and Wnt signaling, in addition to incorporating pathways relating to focused ultrasound as a therapeutic delivery tool. To validate the model, independently established relationships between selected inputs and outputs were simulated, with the model yielding correct predictions 73% of the time. We identified influential and sensitive nodes under different physiological or pathological contexts, including altered brain endothelial cell conditions during glioma, Alzheimer's disease, and ischemic stroke. Nodes with the greatest influence over combinations of desired model outputs were identified as potential druggable targets for these disease conditions. For example, the model predicts therapeutic benefits from inhibiting AKT, Hif-1α, or cathepsin D in the context of glioma – each of which are currently being studied in clinical or pre-clinical trials. Notably, the model also permits testing multiple combinations of node alterations for their effects on the network and the desired outputs (such as inhibiting AKT and overexpressing the P75 neurotrophin receptor simultaneously in the context of glioma), allowing for the prediction of optimal combination therapies. In all, our approach integrates results from over 100 past studies into a coherent and powerful model, capable of both revealing network interactions unapparent from studying any one pathway in isolation and predicting therapeutic targets for treating devastating brain pathologies.

Introduction

Many pathologies of the brain are characterized by both devastating daily effects for patients and daunting challenges and limitations to treatment for physicians [1]. These conditions, which include brain tumors, neurodegenerative diseases, and stroke, vary widely in their incidence, molecular mechanisms, and progression, and yet all remain inherently difficult to treat. At least some of the challenge of treatment rests in the nature of the brain structure itself, and more specifically, the blood-brain barrier (BBB). For many years, delivery of therapeutics to the brain required risky and invasive surgery to facilitate direct injections [[2], [3], [4], [5]], or brain-wide disruption of the BBB to allow passage of drugs [6,7]. More recently, focused ultrasound (FUS) in conjunction with intravenously-injected gas-filled microbubbles has been used to facilitate drug delivery to the brain in a spatially-targeted manner. This approach can be used to deliver genes or other therapeutics to a specific structure or region of the brain non-invasively [[8], [9], [10], [11]].

The cerebral microvasculature consists of a continuous, non-fenestrated layer of brain capillary endothelial cells. Brain endothelial cells are held together by tight junctions, which serve as a barrier to paracellular passive diffusion of materials from the systemic circulation into the surrounding brain tissue [12]. The brain endothelium also expresses drug efflux pumps, such as p-glycoprotein, to continuously remove potentially harmful materials from the central nervous system [13]. Thus, in essence, brain endothelial cells are the gatekeepers to the brain. As the gatekeepers, cerebral endothelial cells play a tremendous role in the state of the brain in both physiological and pathological conditions, by dictating the blood flow, oxygenation, nutrient transport, and drug concentrations in the brain parenchyma. Brain endothelial cells experience competing cues from growth factors, neurotrophic factors, and other cytokines and paracrine signaling molecules, and integrate these diverse signals to regulate behaviors such as angiogenesis, immune cell recruitment, and apoptosis. Therefore, appropriate therapeutic strategies to modulate cerebral endothelial cell signaling must function within this complex environment of diverse signaling cues present in the healthy or diseased brain. Designing such therapies relies on understanding how cells integrate these signals, yet to date no computational models of brain endothelial cell signaling exist.

In this study we develop the first computational model of the brain endothelial cell signaling network in order to identify disease context-dependent drivers of therapeutic patterns of molecule expression or phenotypes. The model incorporates the VEGF-A signaling pathway, the BDNF and NGF neurotrophin signaling pathways, and the Wnt signaling pathway. BDNF and NGF can act through either their unprocessed, pro-neurotrophin forms, which preferentially bind sortilin and the p75NTR receptor to activate a number of pro-apoptotic pathways, or can be processed by protein convertases such as furin or plasmin into their mature neurotrophin forms, which preferentially bind the Trk receptors and induce more pro-survival signaling [14]. VEGF-A signaling drives many pro-angiogenic processes [15], and Wnt signaling in the brain endothelium has been implicated in promoting survival and BBB integrity [16]. Additionally, our model features a FUS input node, based upon prior in vivo transcriptomic studies which demonstrated that FUS-mediated BBB opening results in upregulation of cathepsin D in cerebral endothelial cells [17]. The model's outputs were selected for their relevance to brain pathologies of interest. These include the proteins which constitute tight junctions, claudin-5 and occludin, and adherens junctions, VE cadherin, between endothelial cells. Outputs also include expression of the glucose transporter GLUT1 and the efflux transporter p-glycoprotein, as well as nitric oxide production (which helps regulate vasodilation and blood flow), and apoptosis or cell death.

After validating the model at baseline (“normal” physiological conditions) against the existing literature, we simulate the altered signaling network under three disease states (glioma, Alzheimer's disease, and ischemic stroke) and predict the most influential nodes in regulating potential therapeutic responses to those specific pathologies. This analysis reveals novel potential targets (or combinations of targets) for pharmaceutical intervention using FUS as a delivery mechanism. We submit that this model represents a valuable tool to develop strategic therapeutic approaches for FUS-mediated drug and gene delivery in a range of different disease contexts in the brain.

Section snippets

Validation of computational model of brain endothelial cell signaling

A predictive computational model of brain endothelial cell signaling (Fig. 1) was manually reconstructed from previous experimental studies from the literature. The detailed procedure for literature review and network reconstruction is provided in the Materials and Methods section. The list of molecular nodes implemented in the model is provided in Table S1, and full documentation of each reaction and the experimental evidence supporting that reaction is provided in Table S2. We first predicted

Discussion

Here we manually reconstructed a literature-based network of brain endothelial cell signaling that identifies the nodes regulating the expression of BBB tight junctional and adherens junctional proteins, GLUT1, and the P-glycoprotein efflux receptor, as well as nitric oxide release and apoptosis. This network was used to develop a logic-based predictive model of endothelial cell signaling (with or without FUS treatment), which validated at a rate of 73% in comparison to independent, published

Conclusions

We present here, to our knowledge, the first computational model of brain endothelial cell signaling. We have demonstrated its utility in modeling signaling pathways in brain endothelial cells in either normal physiology or a variety of different disease states. In each of these different states, combinations of therapeutically beneficial goals can be used to design treatment scores. Maximization of these treatment scores with either overexpression or inhibition of different nodes in the

Model construction

A brain endothelial cell signaling network was manually reconstructed from previous in vitro and in vivo experimental studies from the published literature. The network integrates four key inputs of interest, chosen due to their role in angiogenesis, cell survival, or neural stem cell recruitment. These include NGF (nerve growth factor), BDNF (brain-derived neurotrophic factor), VEGF-A (vascular endothelial growth factor-A), and Wnt. Of these, NGF and BDNF can act in either or both of two forms

Declaration of Competing Interest

None.

Acknowledgements

RJP was supported by National Institutes of Health grants R01EB030409, R01EB030744, R01NS111102, and R21NS118278. JJS was supported by National Institutes of Health grants R01 HL137755 and R01 HL137100.

References (63)

  • A.C. Zeigler

    Computational model predicts paracrine and intracellular drivers of fibroblast phenotype after myocardial infarction

    Matrix Biol.

    (2020)
  • K.A. Ryall

    Network reconstruction and systems analysis of cardiac myocyte hypertrophy signaling

    J. Biol. Chem.

    (2012)
  • V.L. Feigin

    Global, regional, and national burden of neurological disorders, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016

    Lancet Neurol.

    (2019)
  • A. Billiau

    Tissue distribution of human interferons after exogenous administration in rabbits, monkeys, and mice

    Arch. Virol.

    (1981)
  • M. Westphal

    A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma

    Neuro-Oncology

    (2003)
  • M.D. Prados

    A randomized, double-blind, placebo-controlled, phase 2 study of RMP-7 in combination with carboplatin administered intravenously for the treatment of recurrent malignant glioma

    Neuro-Oncology

    (2003)
  • C.M. Gorick

    Sonoselective transfection of cerebral vasculature without blood–brain barrier disruption

    Proc. Natl. Acad. Sci.

    (2020)
  • S.Y. Wu

    Characterizing focused-ultrasound mediated drug delivery to the heterogeneous primate brain in vivo with acoustic monitoring

    Sci. Rep.

    (2016)
  • M. Furuse

    Molecular basis of the core structure of tight junctions

    Cold Spring Harb. Perspect. Biol.

    (2010)
  • W. Löscher et al.

    Drug resistance in brain diseases and the role of drug efflux transporters

    Nat. Rev. Neurosci.

    (2005)
  • R. Lee et al.

    Regulation of cell survival by secreted proneurotrophins

    Science (80-)

    (2001)
  • Z.G. Zhang

    VEGF enhances angiogenesis and promotes blood-brain barrier leakage in the ischemic brain

    J. Clin. Invest.

    (2000)
  • S. Liebner

    Wnt/β-catenin signaling controls development of the blood - brain barrier

    J. Cell Biol.

    (2008)
  • A.S. Mathew et al.

    Single-cell mapping of focused ultrasound-transfected brain

    Gene Ther.

    (2021)
  • J.C. Lim

    Activation of β-catenin signalling by GSK-3 inhibition increases p-glycoprotein expression in brain endothelial cells

    J. Neurochem.

    (2008)
  • H. Seyedmirzaei et al.

    VEGF levels in patients with glioma: a systematic review and meta-analysis

    Rev. Neurosci.

    (2021)
  • J. Xiong

    ProBDNF and its receptors are upregulated in glioma and inhibit the growth of glioma cells in vitro

    Neuro-Oncology

    (2013)
  • B. Yang et al.

    Brain-derived neurotrophic factor propeptide inhibits proliferation and induces apoptosis in C6 glioma cells

    Neuroreport

    (2017)
  • M.C. Brown

    Regulatory effect of nerve growth factor in 9 1 integrin-dependent progression of glioblastoma

    Neuro-Oncology

    (2008)
  • M. Zuccarini

    The role of wnt signal in glioblastoma development and progression: a possible new pharmacological target for the therapy of this tumor

    Genes (Basel)

    (2018)
  • M.E. Fukuda

    Cathepsin D is a potential serum marker for poor prognosis in glioma patients

    Cancer Res.

    (2005)
  • Cited by (0)

    View full text