Shades of Blue: how position gradually shapes structure and function in the locus coeruleus

In Chapter 9 of Principles of Neurobiology, we learned about neuromodulator systems and their roles in controlling behavior and sensation. One such neuromodulator, norepinephrine, is released from a small brainstem nucleus called the locus coeruleus, Latin for "blue spot," named for its appearance in unstained tissue. The LC is tiny, containing roughly 3,000 noradrenergic neurons in rodents and on the order of 30,000 in humans, yet it is the sole source of norepinephrine to the forebrain and projects throughout the central nervous system, from cortex to spinal cord. Through these projections, LC neurons shape attention, arousal, stress, and fear.

For decades, the LC was treated as a largely homogeneous structure acting as a brain-wide broadcast system, setting the level of arousal and flagging salient stimuli to all brain areas at once. Recent methodological advances, however, have allowed researchers to examine the LC in far greater detail, and new work suggests there may be meaningful specificity within it. In a recent paper, Su et al. tested this directly, asking whether LC norepinephrine (LC-NE) neurons are truly interchangeable or whether there may be distinguishable subgroups that do different things. If genes or molecules can separate functional classes of LC-NE neurons, that could point toward more precise therapies for the many disorders involving noradrenergic dysfunction.

To begin uncovering the details of LC-NE neurons, the researchers first asked whether these cells can be sorted by where they send their axons. To do this, they painstakingly traced the axons of individual LC-NE cells across the entire brain and built a brain-wide projection map. This map revealed that individual neurons do not project very broadly, as each individual cell reaches a limited set of nearby structures. This tracking also revealed that the deeper a neuron is located within LC, the further back in the brain it projects, with cortex-projecting neurons at the top, or dorsal part, of the LC and spinal-cord-projecting cells at the very bottom, most ventral location. They next asked whether gene expression organized the same way by sequencing RNA from the nuclei of thousands of individual LC-NE neurons and using spatial RNA mapping to link gene expression back to physical position. As with the projections, the authors found that gene expression varies in a smooth gradient along the dorsal-to-ventral axis of LC.

With multiple lines of evidence pointing to organization along the dorsal-to-ventral axis, the authors asked whether these genetic and anatomical variations also correspond to distinct roles in behavior. They trained mice on a task in which the animals licked for liquid rewards from either a left or right port, where the quality of each port changed over time. Mice learn to shift their licking toward the better port, and how quickly they shift serves as a readout of learning and flexibility. Recording neural activity, both by collecting spiking data from single neurons and by using calcium imaging to measure bulk activity of LC axons in the frontal cortex, Su et al. found that activity patterns also vary along the dorsal-to-ventral axis. Dorsal neurons, those projecting to cortex, were more active when mice switched their choice rather than repeating it, and their activity tracked reward prediction error, meaning it scaled with how much better or worse an outcome was than the animal expected. Ventral neurons behaved differently. They responded to the absence of reward regardless of what the animal had expected, signaling that something was missing rather than grading how good the outcome was.

The LC has been a fascinating challenge for researchers for decades, an enigmatic combination of apparent simplicity and real-life complexity. New techniques, like those used in the current study, give us a better view of LC function than ever, allowing researchers to link projection patterns, gene expression, cell body location, and function together to produce a more fulsome understanding of norepinephrine neurons. In this study, when Su et al. looked closely at LC-NE neurons, rather than finding discrete groups of cells doing distinctly different things, they found graded differences in anatomy, physiology, and gene expression. This moves our understanding forward while simultaneously raising a new set of questions. If organization is continuous, should every LC neuron be considered its own type? Where along these gradients do opposing functions give way to one another? Are those transition points fixed, or flexibly maintained by LC circuitry? These questions will certainly keep neuroscientists occupied for decades to come and will likely continue to clarify our understanding while revealing new layers of complexity.

Figure 1: LC norepinephrine neurons project broadly throughout the anterior-posterior axis of the brain, with each individual neuron innervating a constrined local area (left). Projections to anterior regions of the brain map to more dorsal spatial locations within lc, whereas more posterior-projecting neurons are localized more ventrally (middle). This spatial distribution also maps to gene expression patterns which show a graded distribution across the dorso-ventral axis of LC. Adapted from Su et al., 2026.

Reference:

Su, Z., Kosillo, P., Jung, K., Chen, S., Summers, M. T., Piet, A., Hou, H., Hagihara, K. M., Friedmann, D., Ho-Shing, O., Becker, M. I., Chartrand, T., Yin, X., Grotz, P., Hilton-VanOsdall, E., Lee, M., Javeri, R., Tuggle, S. L., Ouellette, N., . . . Cohen, J. Y. (2026). Topographic structure and function of locus coeruleus noradrenaline neurons. Nature. Advance online publication. https://doi.org/10.1038/s41586-026-11026-0