The effect of a serotonine agonist on neural activity and BOLD activity in monkey primary visual cortex, a pharmacological fMRI (PhMRI) study

Functional magnetic resonance imaging (fMRI) offers great diagnostic potential for monitoring brain activity due to its non-invasiveness. However the neurophysiological basis of BOLD contrast mechanisms in fMRI is not fully understood. Pharmacological functional magnetic resonance imaging (PhMRI) is a promising new direction in biomedical imaging, which allows for monitoring drug related effects on brain processes. When using drugs with known pharmacodynamics (drug effects on the brain), PhMRI offers great possibilities to get a better understanding of the neuronal basis of the BOLD signal. It can provide the link between drug induced biomolecular changes and their corresponding BOLD response. To take full advantage of PhMRI we are developing an integrated software and hardware platform to record in real-time mode simultaneously neurophysiological and BOLD signals to follow drug induced changes in both signals. Real-time mode allows for controlling drug induced effects tightly and offers the possibility to online modify application parameters of the drug. We started to test pharmacological agents and investigated the effect of the neuromodulator BP554 in the primary visual cortex (V1), of anesthetized monkeys. BP554 is a 5-HT1A agonist acting primarily on the membrane of efferent neurons by potassium-induced hyperpolarization. Combined electrophysiology and (fMRI) experiments suggested that local field activity (LFP) is a better predictor of the BOLD signal than multi-unit activity (MUA). This is particularly true because BOLD responses remain undiminished in cases where spiking might be entirely absent despite clear, strong stimulus-induced modulation of the field potentials. To further test this hypothesis we induced the dissociation of MUA from LFP activity with injections of BP554 into primary visual cortex. Neuroimaging was performed in a 4.7 Tesla Scanner (Bruker, Germany). Recorded were spiking activity and local field potentials. V1 was stimulated by rotating polar checkerboard stimulus (blocks by 30 sec stimulus, 30 sec blank, 37 repetitions). 300 microm to the recording electrode we injected BP554 (100 microM solution). The infusion of BP554 in visual cortex reliably reduced MUA without affecting LFP and BOLD activity. This finding suggests that the efferents of a neuronal network pose little metabolic burden compared to the overall pre- and postsynaptic processing of incoming afferents. These results show how powerful PhMRI can be in approaching the still open issue of the coupling between neuronal activity and the BOLD signal, when appropriate hardware and software achievements are incorporated