Retrograde tracing of breast cancer-associated sensory neurons

Abstract

Breast cancer is one of the leading causes of mortality among women. The tumor microenvironment, consisting of host cells and extracellular matrix, has been increasingly studied for its interplay with cancer cells, and the resulting effect on tumor progression. While the breast is one of the most innervated organs in the body, the role of neurons, and specifically sensory neurons, has been understudied, mostly for technical reasons. One of the reasons is the anatomy of sensory neurons: sensory neuron somas are located in the spine, and their axons can extend longer than a meter across the body to provide innervation in the breast. Next, neurons are challenging to culture, and there are no cell lines adequately representing the diversity of sensory neurons. Finally, sensory neurons are responsible for transporting several different types of signals to the brain, and there are many different subtypes of sensory neurons. The subtypes of sensory neurons which innervate and interact with breast tumors are unknown. To establish the tools for labeling and subtyping neurons that interact with breast cancer cells, we utilized two retrograde tracers standards in neuroscience, wheat-germ agglutinin (WGA) and cholera toxin subunit B (CTB). In vitro, we employed primary sensory neurons isolated from mouse dorsal root ganglia, cultured in a custom-built microfluidic device DACIT, that mimics the anatomical compartmentalization of the sensory neurons soma and axons. In vivo, we utilized both syngeneic and transgenic mouse models of mammary carcinoma. We show that CTB and WGA trace different but overlapping sensory neuronal subpopulations: while WGA is more efficient in labeling CGRP+ neurons, CTB is superior in labeling the NF200+ neurons. Surprisingly, both tracers are also taken up by a significant population of breast cancer cells, both in vitro and in vivo. In summary, we have established methodologies for retrograde tracing of sensory neurons interacting with breast cancer cells. Our tools will be useful for future studies of breast tumor innervation, and development of therapies targeting breast cancer-associated neuron subpopulations of sensory neurons.