MaRGA Meeting, 2004: Abstracts

11: Neuroscience / Communication / Behavior

Functional Magnetic Resonance Imaging in Conscious Marmosets:
Methods and Applications in Neuroscience Research

C. Ferris

Affiliation: Center for Comparative Neuroimaging, University of Massachusetts Medical School, Worcester, MA

Functional magnetic resonance imaging (fMRI) at high magnetic field strengths (4.7T - 11.7T) has far superior spatial and temporal resolution than any other non-invasive imaging technique. Functional MRI using the BOLD (blood oxygenation-level-dependent) technique measures changes in blood flow to areas of increased synaptic and neuronal activity. Mobil protons associated with hydrogen atoms in water are the primary source of MR signal. The level of paramagnetic deoxygenated hemoglobin in the blood vessels alters the magnetic-susceptibility of protons aligned in the magnetic field of the spectrometer. Enhanced neuronal activity is accompanied by an increase in metabolism concomitant with changes in cerebral blood flow and volume to the area of activation. The local blood flow to this area exceeds oxygen uptake, lowering the level of deoxygenated hemoglobin and increasing the T2*-relaxation time and MRI signal intensity. Fully conscious animals are not used routinely in fMRI studies because of technical problems associated with motion artifacts. Any minor head movement distorts the image and may also create a change in signal intensity that can be mistaken for stimulus-associated changes in brain activity. In addition to head movement, motion outside the field of view can also obscure or mimic the signal from neuronal activation. Therefore most fMRI studies use general anesthetics to immobilize the animal. However, anesthetics preclude the study of brain activity involving cognition and emotion. Furthermore, anesthetics depress neuronal activity reducing BOLD signal. To circumvent these problems technology was developed to image fully conscious marmoset monkeys.

Methods

A restrainer was developed consisting of head and body holder with built-in radiofrequency electronics. Prior to imaging animals are habituated to the restraint stress and imaging procedure in a simulated environment. During MR sessions, animals are first lightly sedated with ketamine plus medetomidine (Domitor), and placed in the MR head and body holder. Once securely restrained, anesthesia is reversed with atipamezole (Antiseden). Prior to imaging animals are tested for their sensitivity and recovery to this anesthetic procedure. Animals recovering from anesthesia are judged to be fully conscious when they can both locomote normally and attend to cognitive tasks in their environment. This recovery time from anesthesia is incorporated into the imaging study. High-resolution anatomical data sets of ca. 12 min in duration are acquired using a multi-slice spinecho (RARE) pulse sequence at the beginning and end of each imaging session. In plane spatial resolution is ca 120 µm x 120 µm with a slice thickness of 1 mm. Functional images are acquired at 15 sec intervals using a gradient-echo (FLASH) sequence weighted for T2*. BOLD signal changes between control and stimulation periods are analyzed and significant differences presented as activational maps and time-course plots. Subtraction of the first and last anatomical data sets and observation of functional time-series helps detect motion artifact.

Applications in Neuroscience Research

Emotional States
Fear, anger, hunger, and sexual arousal are examples of emotion states and are fertile areas of investigation using fMRI. A library of vocalizations, smells and visual images with proven ethological significance in the animal's natural habitat and in the semi-natural environment of the laboratory setting can be collected and used to communicate with the animal in the magnet. For example, presentation of the odor of a novel reproductively receptive female marmoset to a male marmoset in the magnet will elicit changes in brain activity related to sexual arousal.

Brain/Environment Interactions in Development
There are myriad examples in animal studies showing early emotional or environmental insult can affect brain development with long-term neurobiological and behavioral consequences. Insights into the etiology of mental illness may be gleaned by longitudinal studies on marmosets examining the interaction between a vulnerable gene pool and a stressful environment at critical times in development. Since fMRI is non-invasive and can be used to study the same animal over the course of its life it is possible to observe developmental changes in neuroanatomy, brain activity and brain chemistry (spectroscopy).

Drugs Effects on Brain Activity
Studying changes in brain activity in response to acute and prolonged exposure to psychotherapeutics and drugs of addiction are two other obvious applications of fMRI. For example, many psychotropic drugs cause a prompt increase in brain levels of neurotransmitters. Nonetheless, patients require weeks of treatment before reporting an improvement in their condition. This would suggest drug efficacy for the treatment of mental illness is due to secondary changes in the neurochemical signals and pathways that are slowly affected by the continuous exposure to the psychotropic agent. Functional MRI would help to resolve this mechanism of action. In the case of drugs of addiction, rhesus monkeys can be trained to self-administer cocaine, withdrawn from the drug and later reinstated in response to conditioned cues. It is feasible to image these different phases of cocaine addiction.

Testing Cognitive Performance
Since animals will readily respond to peripheral stimulation while in the magnet fMRI they may be used in studies of classical conditioning. For example, foot shock can be used as an unconditioned response in associative learning paradigms. When coupled with a conditioned stimulus like scent or light it can be used in learning studies examining discrimination and perception. Operant conditioning would be more difficult because a behavioral action, e.g. bar pressing eliciting rewarding or punishing stimuli, would be necessary. However, recent imaging studies on conscious rhesus monkeys show this is feasible and opens the area of cognitive neuroscience to investigation with fMRI in marmosets.

Acknowledgement: This work was funded by grants from the National Institute of Mental Health MH58700, MH59501