Graduation Year


Date of Submission


Document Type

Open Access Senior Thesis

Degree Name

Bachelor of Arts



Reader 1

Tony Yaksh

Reader 2

Cathy Reed

Rights Information

(c) 2020 Emily A Kussick


The present study examined morphological changes in the dorsal root ganglia (DRG) following an innate immune stimulus. The importance of the DRG has increasingly become recognized in pain processing as more than just the home of primary afferent cell bodies. All sensory information passes through the DRG via the primary afferents, and on to the spinal cord. The primary afferents synapse with second-order neurons in the spinal cord that ascend towards the brain, where they transmit the pain signal to the limbic forebrain and/or the somatosensory cortex for processing. The DRG is an interesting niche to study at as it lies outside the blood- brain-barrier (BBB) but projects to the CNS. Therefore, neurons in the DRG are exposed to large circulating molecules (such as LPS) that normally would not be able to act on second-order neurons in the spinal cord, due to the BBB, and thus can indirectly act on second-order neurons by sensitizing their primary afferent inputs, which increases primary afferent output into the spinal cord. Of particular interest, the DRG constitutively possesses a significant population of macrophages. Mice injected with lipopolysaccharide (LPS) will lead to an enhanced pain response that has been speculated to result from the activation of innate immune receptors (TLR4 expressed on the macrophage which leads to the release of active factors that sensitize the sensory neuron). To enable study of these DRG macrophages, we employed a novel 3D imaging technique to visualize morphological changes in whole cells, as opposed to previous imaging methods that only captured slices of cells, which greatly improved our ability to depict the dense and complex cellular environment of the DRG. A machine learning network was employed to train 3D stacks of DRG tissue images at different depths and intensities. After running thousands of replicates (epochs), through curated images, the machine learning program became very accurate and enabled us to characterize, in 3D reconstructions, the DRG macrophage population.

3D imaging of DRG macrophages 5

This approach enabled several observations. 1) Two distinct sub-populations of macrophages in the DRG were found: non-vascular and peri-vascular macrophages. 2) Spinal delivery of LPS which activated TLR4 receptors was found to increase the size of the individual macrophages, but in contrast to previous studies no the number. 3) This change in macrophage size vs. number was confirmed by the lack of change in the macrophage response in CXC3R (-) mutants which do not have circulating macrophages. The macrophage production of many spiny, amorphic processes were likely “double counted” in previous studies done on tissue slices (as opposed to the 3D, whole-cell modeling employed in these studies). By visualizing whole 3-D reconstructed cells, we were able to differentiate between increases in macrophage number versus increased macrophage volume that manifested. These techniques were validated by comparing our data analysis pipeline with commercially-available software, which yielded similar results. Further application, of our techniques and pipeline will enable the 3D imaging of different cell types and their interactions within the DRG.