Blood-oxygenation-level-dependent functional magnetic resonance imaging (BOLD fMRI) is widely used in to study human brain function; however the cellular and molecular mechanisms underlying the BOLD signal remain poorly understood. The BOLD signal is highly complex as it represents disproportionate interactions of cerebral blood flow (CBF), cerebral blood volume (CBV), and cerebral metabolic rate of oxygen (CMRO2) during neuronal activation. On the cellular level, while lactate generated from the astrocytes is used to sustain neuronal activity, astrocytic signaling also releases vasoactive compounds, indicating that BOLD could reflect a combined response of both neurons and astrocytes. Dissecting the fractional contribution of neurons, astrocytes, their crosstalk, and specific molecular signaling cascades to BOLD, CBF, CBV, and CMRO2 is crucial to more accurately model and interpret BOLD data. Unlike neurons, astrocytes lack the appropriate ion channels to propagate action potentials but rather mediate their activity predominantly through G-protein-coupled receptors (GPCRs). Substantial pharmacological evidence has suggested that astrocytic GPCRs are key molecular players in their control of CBF through their binding of various paracrine compounds released by neurons. Interestingly, some studies have questioned this conclusion, demonstrating that activation of astrocytic Gq-GPCRs are not critical for CBF modulation. Further, it remains unclear how other GPCR subfamilies (i.e., Gs and Gi) affect BOLD. These controversies and missing data prompted us to systematically investigate the following questions for the first time: 1) whether selective activation of astrocytic Gq-, Gs-, or Gi-GPCR signaling pathways modulate hemodynamic or BOLD responses in vivo, 2) can neurons or astrocytes independently elicit hemodynamic and BOLD responses without the involvement of the other, and 3) what molecular mechanisms contribute to the BOLD signal disruption in disease states where astrogliosis and neuronal remodeling occur. We will employ cutting-edge chemogenetic tools, a.k.a. Designer Receptors Exclusively Activated by Designer Drugs (DREADDs), to selectively modulate Gq-, Gs- and Gi-signaling cascades in neurons and astrocytes. We will also utilize multimodal fMRI tools that allow measurement of BOLD, CBV, CBF, and CMRO2 changes in a single setting. Additionally, we will perform immunohistochemistry in all subjects, allowing within-subject comparison of the number or ratio of activated/suppressed cells and the observed hemodynamic responses.
In Aim 1, we propose to use DREADDs to directly activate the signaling of each of the main astrocytic GPCR subfamily during fMRI, allowing precise interrogation of the astrocytic signaling pathways that contribute to changes in BOLD.
In Aim 2 a, we will employ a novel means to concomitantly suppress astrocytic cyclic-adenosine-monophosphate-related activity using Gi-DREADD during neuronal activation. Conceptually, this will ?remove? the astrocytes during fMRI mapping of neuronal activation.
In Aim 2 b, we will silence neurons using Gi-DREADD while exclusively activating Gq- and Gs-DREADDs in astrocytes. This will ensure the exclusion of potential paracrine factors released from neurons that could directly modulate vascular tone.
In Aim 3, we will employ an endotoxin-induced model of chronic neuroinflammation using lipopolysaccharide (LPS), thus creating well-characterized region and time-specific pathological profiles. We will scan these animals identically as described in Aim 2, but under two stages of neuroinflammation: 1) the acute phase (3 days after LPS exposure) which consists of peak presence of astrogliosis with very minimal neuronal remodeling, and 2) the chronic phase (90 days after LPS exposure) which consists of moderate to mild astrogliosis with substantial neuronal remodeling. We anticipate that our results will reveal the respective roles of neurons, astrocytes, and specific GPCR signaling cascades in the generation of BOLD. We also expect our study to shed considerable light on the mechanisms by which the BOLD signal can be disrupted in disease states involving neuroinflammation. Lastly, we will perform BOLD modeling with the unique datasets to be generated in this study, with the ultimate hope of building a more solid foundation for human brain mapping.

Public Health Relevance

This project aims to investigate whether and how neurons and astrocytes mediate blood-oxygenation- level-dependent (BOLD) functional magnetic resonance imaging (fMRI) signal, a technique that has been widely used to probe brain functional activity and connectivity in humans. A novel chemogenetic approach, namely, Designer Receptors Exclusively Activated by Designer Drugs (DREADDs), will be used to selectively manipulate neuronal and astrocytic activity during fMRI. Further, we will investigate the mechanism(s) by which the BOLD signal is altered in neuroinflammation. The novel findings from this project will provide the missing link between macroscopic changes in BOLD with cellular-level neural signaling, develop a more complete model of BOLD, and ultimately build a solid foundation for human brain mapping.

National Institute of Health (NIH)
National Institute of Mental Health (NIMH)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZMH1)
Program Officer
Kim, Douglas Sun-IL
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of North Carolina Chapel Hill
Schools of Medicine
Chapel Hill
United States
Zip Code
Broadwater, Margaret A; Lee, Sung-Ho; Yu, Yang et al. (2018) Adolescent alcohol exposure decreases frontostriatal resting-state functional connectivity in adulthood. Addict Biol 23:810-823
Kao, Yu-Chieh Jill; Oyarzabal, Esteban A; Zhang, Hua et al. (2017) Role of Genetic Variation in Collateral Circulation in the Evolution of Acute Stroke: A Multimodal Magnetic Resonance Imaging Study. Stroke 48:754-761
Van Den Berge, Nathalie; Albaugh, Daniel L; Salzwedel, Andrew et al. (2017) Functional circuit mapping of striatal output nuclei using simultaneous deep brain stimulation and fMRI. Neuroimage 146:1050-1061