1. To investigate the underlying neural mechanism of the resting-state BOLD signal using rodent models Despite the wide range of applications in systems neuroscience and its great clinical potential, the neurophysiological basis of the resting-state BOLD signal remains to be established. Converging studies support the view that local field potentials (LFP) in the high frequency range correlates with the evoked-BOLD signal (e.g. visual stimulation). It remains unclear if this relationship extends to rsfMRI. Numerous studies have explored this relationship in parallel groups but only a few have done so in a simultaneous manner, with contradicting results. In addition the relationships were only correlations and causal has yet to be explored. In the present study, we will use the rodent striatal network to differentiate the neurophysiological correlate of the resting-state BOLD signal by reversibly modulating the striatal LFP and measuring the resting-state BOLD response change. These findings will unequivocally establish the true LFP correlate of the resting-state BOLD signal. On our end, we have successfully developed a protocol which allows us to simultaneously record the resting electrical activity of the rodent brain during resting state imaging with minimal imaging artifacts. We have also begun to study the spatial and spectral correlates of the local field potential of the rodent brain and the resting-state BOLD signal. 2. Functional connectivity hubs in the awake marmoset monkey Increasing evidence indicates that like many complex systems, the network architecture of the primate brain may incorporate highly functionally connected nodes, or hubs; an idea that is consistent with small-world and scale-free network models of brain function. We recently demonstrated results from small New World monkeys that allows for the characterization of resting-state networks in the awake state (Belcher et al., Journal of Neuroscience 33: 16796-16804, 2013).
The aim of the current study was to employ our data-driven, novel algorithm to identify whether the conscious marmoset brain possesses local functional connectivity density (lFCD), or hubs of connectivity. Additionally, we investigated the hubs relative strength and variability to assess whether marmoset hubs show the characteristic dynamics seen in human functional hubs. Functional connectivity density mapping revealed prominent lFCD hubs in the visual cortex (V1, V2, V6(DM), A19M), medial parietal cortex (PGM), posterior (A23b/A31) and anterior (Cg/A24a) cingulate cortices and subcortical regions (Cd, Pu, LSI, Tha). In these regions, the lFCD was higher than the whole brain average signal. The probability of the lFCD decreased exponentially with the strength of the lFCD, such that there were few connectivity hubs and numerous weakly connected nodes. This lFCD distribution was highly significant across conditions and animals (P < 0.001, T-score > 3; t-test). Furthermore, the statistical significance of the lFCD was stronger for cortical regions than for subcortical regions as within-subject variability in the strength of the subcortical hubs was higher than for cortical hubs, reflecting the predominance of highly dynamic local connectivity in these cortical brain regions in the marmosets (relative variability of the lFCD=4921% for Cd and 2710% for V1; standard deviation-to-mean ratio; P < 0.006, paired t-test, df=9). The results provide weight to the rationale of using a model that characterizes more aptly the scale-free organization of the mammalian brain. 3. Manganese-Enhanced MRI Reflects Both Activity-Independent and Activity-Dependent Uptake within the Rat Habenulomesencephalic Pathway Manganese-enhanced magnetic resonance imaging (MEMRI) is a powerful technique for assessing the functional connectivity of neurons within the central nervous system. Despite the widely held proposition that MEMRI signal is dependent on neuronal activity, few studies have directly tested this implicit hypothesis. In the present series of experiments, MnCl2 was injected into the habenula of urethane-anesthetized rats alone or in combination with drugs known to alter neuronal activity by modulating specific voltage- and/or ligand-gated ion channels. Continuous quantitative T1 mapping was used to measure Mn2+ accumulation in the interpeduncular nucleus, a midline structure in which efferents from the medial habenula terminate. Microinjection of MnCl2 into the habenular complex using a protocol that maintained spontaneous neuronal activity resulted in a time-dependent increase in MEMRI signal intensity in the interpeduncular nucleus consistent with fast axonal transport of Mn2+ between these structures. Co-injection of the excitatory amino-acid agonist AMPA, increased the Mn2+-enhanced signal intensity within the interpeduncular nucleus. AMPA-induced increases in MEMRI signal were attenuated by co-injection of either the sodium channel blocker, TTX, or broad-spectrum Ca2+ channel blocker, Ni2+, and were occluded in the presence of both channel blockers. However, neither Ni2+ nor TTX, alone or in combination, attenuated the increase in signal intensity following injection of Mn2+ into the habenula. These results support the premise that changes in neuronal excitability are reflected by corresponding changes in MEMRI signal intensity. However, they also suggest that basal rates of Mn2+ uptake by neurons in the medial habenula may also occur via activity-independent mechanisms. (PLoS One. 2015;10(5):e0127773.) 4. A novel method to induce nicotine dependence by intermittent drug delivery using osmotic minipumps Osmotic pumps have been routinely used to automatically deliver various drugs and hormones systemically in preclinical neuroscience models, including rodent models of nicotine, cocaine and opiate dependence. However, this continuous method of drug administration fails to accurately mimic the typical pattern of nicotine intake in humans and also does not incorporate the rapid nicotinic receptor alterations that are known to occur in vivo. Intermittent drug delivery in rodents provides a more authentic model, but requires labor intensive intravenous surgery as well as instrumentation for and training with passive or self-administration procedures. In order to determine whether nicotine dependence can be induced by a simpler intermittent delivery system, rats were implanted (i.p.) with an osmotic pump attached to a Lynch coil, which was filled with alternating nicotine solution and mineral oil in volumes calculated to deliver a single nicotine injection every hr for 14 days. Two doses of nicotine (0.1 mg/kg/hr and 0.2 mg/kg/hr) and saline vehicle were administered (n=6 per group) using a model 2ML4 pump (Alzet). Animals were challenged with mecamylamine (1.5 mg/kg s.c.) and observed for somatic signs of nicotine withdrawal for 50 min (Malin et al., 1992) at 7, 14, and 21 days after pump implantation. The summation of these signs gave a withdrawal score for each rat, which was used to compare the degree of dependence between treatment groups and time points. A Group x Day ANOVA revealed a significant main effect of Group, whereby both low and moderate nicotine dose groups had significantly greater withdrawal scores than the saline group at 7, 14, and 21 days F (2, 15) = 8.87, P <0.05. Though somatic signs showed a trend for increased scores in a dose-dependent manner, the two nicotine groups did not significantly differ from one another. To our knowledge, this is the first study to demonstrate dose-dependent nicotine dependence following non-continuous delivery via an osmotic pump (under review).
Showing the most recent 10 out of 19 publications
