The disconnection of BLA-LHA system that selectively abolished CS-potentiated feeding implicates the LHA as a final common pathway for the influence of learned potentiation on food consumption. However, in addition to direct projections from BLA to the LHA, other indirect pathways in the forebrain could also be involved. The BLA participates in a forebrain network with many indirect pathways for gaining access to the hypothalamus, including projections from BLA to medial prefrontal cortex (mPFC) and the nucleus accumbens (ACB), along with projections to the central nucleus of the amygdala (CEA), each of which, in turn, innervates the LHA [49
]. To study these routes we used functional anatomical methods to further delineate critical components of a forebrain-LHA network engaged in CS-potentiated feeding.
We employed a novel approach for functional mapping of activated circuitry in the CS-potentiation paradigm. Retrograde tract-tracing was used in conjunction with methods to identify activated neurons. The effector immediate early genes (IEGs), Arc
(also known as Arg3.1
, which encodes activity-regulated cytoskeleton-associated protein), and Homer 1a (H1a)
, were employed as markers of activated neurons [50
mRNAs are expressed in the same neurons with temporally offset appearance/disappearance, and as such can be employed to detect neuronal populations activated by two temporally distinct experiences in a single brain [50
]. This approach allowed us to view the components of the BLA-LHA circuitry that are selectively activated in food consumption tests with CS presentations that stimulate eating. Thus, we assessed sated rats for food consumption in the presence of a cue that was previously paired with food (CS+), or in the presence of another cue that was never paired with food (CS−), in two consecutive tests temporally arranged for activation of Arc
and Homer 1a
and examined the selective induction of these IEGs in BLA, CEA, mPFC, and ACB neurons that project to LHA, as identified with the retrograde tracer (FluoroGold).
Here we found that LHA projection neurons, as defined with retrograde tracer, localized in the basomedial and adjacent basolateral nuclei within the BLA and in the mPFC were activated selectively by a cue that stimulates eating in sated rats (, pathways indicated in red). In both regions, a significantly larger percentage of the neurons projecting to the LHA show IEG induction in response to food consumption tests with CS+ compared to tests with CS−. Thus, these findings demonstrate that both direct BLA projections and a prefrontal projection system to the LHA are activated during CS-driven eating.
Figure 1 This diagram summarizes recent findings from neural systems analysis of the behavioral model, conditioned potentiation of feeding, that relies on learned cues to override metabolic signals. In that model, a cue previously paired with food when an animal (more ...)
Interestingly, the two other brain regions examined in that study that project to the LHA, the ACB and CEA, did not contain neurons that were selectively activated in the setting of potentiated feeding. That result suggests that anatomical routes to LHA via the ACB and CEA are not critical for learning-dependent modulation of feeding in our paradigm. This interpretation is further corroborated by results from lesion studies. In addition to the earlier described result, namely that bilateral neurotoxic CEA lesions spare potentiated feeding [24
], we also found no impairment after disconnection of the BLA and ACB (using a preparation conceptually similar to the BLA-LHA disconnection involving lesions placed contralaterally in BLA and ACB) [10
]. At the same time, it is important to note that a variety of evidence has implicated those regions of the forebrain in other aspects of motivational control in feeding [33
]. Thus, the ACB and CEA might interact with the BLA-LHA in settings dissociable from cue-driven eating. Within that context, the BLA-LHA system is needed in ACB-dependent, μ-opioid induced consumption of fat [52
]. Notably the CEA, including its projection to the LHA, which is not critical for cue-‘enhanced’ consumption, may be needed to modulate feeding in aversive settings [53
]. Thus, subsystems within a larger network appear to be recruited by different processes, including a more general motivation to eat, motivation for highly palatable foods, stress-regulated eating, or selective cue-driven consumption. Of course, it will be of great interest to determine whether these different subsystems interact with common or different metabolic regulators within the LHA and other components of the feeding system. Clearly, more work is needed to fully delineate the exact circuitry and mechanisms used to modulate feeding in many different settings.