Neuronal inhibition and excitation, and the dichotomic control of brain hemodynamic and oxygen responses

Abstract

Brain's electrical activity correlates strongly to changes in cerebral blood flow (CBF) and the cerebral metabolic rate of oxygen (CMRO₂). Subthreshold synaptic processes correlate better than the spike rates of principal neurons to CBF, CMRO₂ and positive BOLD signals. Stimulation-induced rises in CMRO₂ are controlled by the ATP turnover, which depends on the energy used to fuel the Na,K-ATPase to reestablish ionic gradients, while stimulation-induced CBF responses to a large extent are controlled by mechanisms that depend on Ca²⁺ rises in neurons and astrocytes. This dichotomy of metabolic and vascular control explains the gap between the stimulation-induced rises in CMRO₂ and CBF, and in turn the BOLD signal. Activity-dependent rises in CBF and CMRO₂ vary within and between brain regions due to differences in ATP turnover and Ca²⁺-dependent mechanisms. Nerve cells produce and release vasodilators that evoke positive BOLD signals, while the mechanisms that control negative BOLD signals by activity-dependent vasoconstriction are less well understood. Activation of both excitatory and inhibitory neurons produces rises in CBF and positive BOLD signals, while negative BOLD signals under most conditions correlate to excitation of inhibitory interneurons, but there are important exceptions to that rule as described in this paper. Thus, variations in the balance between synaptic excitation and inhibition contribute dynamically to the control of metabolic and hemodynamic responses, and in turn the amplitude and polarity of the BOLD signal. Therefore, it is not possible based on a negative or positive BOLD signal alone to decide whether the underlying activity goes on in principal or inhibitory neurons.

Introduction

Brain function emerges from signaling in and between neurons and associated astrocytes, organized in large-scale synaptic and astrocytic networks, which causes fluctuations in the cerebral metabolic rate of oxygen (CMRO₂) and cerebral blood flow (CBF). These fluctuations induce changes in the blood content of deoxyhemoglobin which is the basis of the fMRI BOLD signal (Ogawa et al., 1990). Global CBF and CMRO₂ remain largely constant during normal mental operation (Sokoloff et al., 1955) (Fig. 1). Therefore, the stimulus-induced rises in local hemodynamic and metabolic signals relate to specific mechanisms that rapidly and flexibly reallocate CBF and in turn glucose and O₂ supply from less active to more active nerve cells. The net fMRI BOLD response, i.e. positive or negative deflection, depends on the gap between the stimulus-evoked increments in CBF and CMRO₂ (Raichle and Mintun, 2006). The physiological basis for this gap may be explained by differences in control mechanisms: activity-dependent rises in CMRO₂ correlate with and are caused by ATP turnover due to ion pumping (Erecinska and Silver, 1989, Mathiesen et al., 2011). In comparison, activity-dependent CBF responses are mainly evoked by signaling processes that trigger rises in cytosolic Ca²⁺ in neurons and astrocytes (Attwell et al., 2010, Chaigneau et al., 2007, Girouard et al., 2010, Lauritzen, 2005, Mathiesen et al., 2011, Takano et al., 2006), although Ca²⁺-independent processes also contribute (Caesar et al., 1999, Leithner et al., 2010).

Stimulation of somatosensory or other afferents activate both excitatory and inhibitory neurons in the cortex. It has been postulated that neuronal inhibition may raise CBF if there is either low local excitatory recurrence or if the region is not otherwise being driven by excitation. Conversely, with high recurrence or actively driven excitation, inhibition may lower the observed CBF or BOLD values (Tagamets and Horwitz, 2001). The hemodynamic signals are in other words, context-sensitive, i.e. to some extent unpredictable, since in response to the same input, a given network can produce different output patterns at different times, depending on the state of inhibition (Buzsaki et al., 2007), and by controlling the increases in intracellular Ca²⁺ evoked by excitation (Lauritzen, 2005, Nakamura et al., 2002).

This review is focused on 4 aspects of hemodynamic and metabolic signals. First, different models are discussed that we have used to study hemodynamic and metabolic signals in relation to functional activation, primarily the cerebellum and the primary somatosensory cortex. Second, we describe why variations in synaptic activity and not action potential generation correlate to and produce stimulus-induced CBF and CMRO₂ responses. Third, we describe how variations in the tonic level of GABA influence the stimulation-induced rises in CMRO₂ and CBF. Finally, we summarize studies from the primary somatosensory cortex which suggest that stimulation-induced CBF and CMRO₂ responses differ greatly for different synaptic inputs (Enager et al., 2009). In addition, the studies indicate that inhibitory interneurons produce rises in CBF responses and in turn positive BOLD signals under some conditions (Cauli et al., 2004, Enager et al., 2009, Kocharyan et al., 2008, Rancillac et al., 2006), while under other conditions synaptic inhibition correlates to decreases in CBF due to vasoconstriction and a negative BOLD signal (Devor et al., 2007, Devor et al., 2008, Kastrup et al., 2008, Stefanovic et al., 2004).