My original research project supported by the Rappaport fellowship program proposed to discretely transplant GABAergic neurons into the amygdala to study fear and anxiety. The amygdala is perhaps the most important brain region in processing these strong emotions. When we are emotionally tense, anxious, frightened, or in a state of terror, the activity of the amygdala is increased. And when we are calm and content, the activity of the amygdala is decreased. Most psychiatric diseases have one thing in common, and that is a state of anxiety, fear, or frank horror, which may be ever-present in the patient. It is imperative then that we attempt to understand the role of the amygdala and its connections with other brain regions in the generation of fear.

One approach to learning about how the amygdala functions is to introduce specific cell types into its circuitry that alter or modulate its activity. GABAergic neurons are responsible for providing “inhibitory tone” in the brain. That is, the neurotransmitter, GABA, decreases the firing rate of neurons. Medicines that reduce anxiety, like the benzodiazepines (e.g., Valium, Serax, and Ativan), do so by enhancing the effects of GABA. Therefore, the original proposal was to attempt to decrease the activity of the amygdala in fearful states by engrafting neurons that produce the neurotransmitter, GABA.

The first challenge was to produce a suspension of cells that was rich in GABAergic neurons and then to surgically inject small volumes of this suspension into specific areas of the amygdala. With the assistance of my students and co-workers, we were indeed able to produce viable cell suspensions rich in GABAergic neurons by dissecting and dissociating tissue from embryonic animals from an area called the lateral ganglionic eminence (LGE). The LGE is where the GABAergic neurons of the brain are originally derived. After labeling a portion of the transplanted cells with red fluorescent microspheres (FLMs) to help identify them after transplantation, we were able to engraft these cells precisely into the lateral and basolateral subregions of the amygdala using the “micrografting” technique previously developed at the Massachusetts Institute of Technology.

It was then important to determine if transplanted cells could actually survive and interact within their new environment. Twelve weeks after transplantation, an immunoreaction against GABA was used to help identify GABA-producing cells. A large numbers of transplanted GABAergic neurons survived and integrated within the host amygdala.

Further proof of survival and integration was then achieved using electron microscopy (EM). Because the FLMs can be “photoreacted” to form an electron-dense product, transplanted cells containing FLMs can undergo elegant ultrastructural analysis with EM.

The ultimate test however, was to determine if these grafts had any effect on how the animals experienced fear and anxiety. Three models of fear and anxiety were used: the elevated plus maze, light-enhanced startle, and fear potentiated startle. These tests exploit the rat’s innate fear and avoidance of heights, open space, and light. The elevated plus maze is a model of anxiety based on animals’ uneasiness for walking onto the narrow open arms, as opposed to the enclosed arms, of a plus-shaped platform elevated 50 cm above the floor. Animals that have received anxiolytic drugs venture onto and spend more time on the open arms of the maze. Light-enhanced startle is another model of anxiety in which an animal’s startle response is augmented if the animal receives a startle-eliciting noise in an environment that is brightly lit as opposed to darkened. Fear-potentiated startle is a model in which an animal learns to fear a neutral, “conditioned” stimulus (such as a light or tone) after it is paired with an aversive stimulus (such as mild electrical shock). The level of fear in this model is quantified by the amount of the startle response after training. Anxiolytic drugs attenuate both fear-potentiated and light-enhanced startle.

Time spent on the open arms of the elevated plus maze was greater for the grafted animals than for control animals. Fear-potentiated startle, however, demonstrated that the grafted animals responded similarly to the control animals, suggesting that the amygdala circuitry remained intact after grafting, and animals retained the capacity to learn. Interestingly, grafted animals demonstrated a significantly less exaggerated startle response in the presence of light. These dramatic results strongly suggests that grafts may have the ability to attenuate anxiety, but they do not appear to interfere with the learning of threatening stimuli and responding appropriately, which might be expected if the grafts were in effect interfering with or destroying part of the amygdalar circuitry.

In conclusion, these exciting and compelling results have opened a new area of neuroscience investigation: the use of neural transplantation for the study and treatment of psychiatric disease. This approach has been studied for decades in the context of neurologic disorders, such as Parkinson’s disease, Huntington’s disease, stroke and epilepsy. The present studies demonstrate that such techniques and principles may indeed be applicable to debilitating diseases of perception and mood.