Divergent innervation of the olfactory bulb by distinct raphe nuclei

Raphael Steinfeld, Jan T. Herb, Rolf Sprengel, Andreas T. Schaefer, Izumi FukunagaThe Journal of Comparative Neurology 2014 · DOI

TL;DR

This study reveals that two key brainstem nuclei, the dorsal raphe nucleus (DRN) and the median raphe nucleus (MRN), which both provide serotonin to the olfactory bulb (OB), have surprisingly distinct projection patterns. Contrary to previous assumptions, the DRN primarily innervates the granule cell layer (GCL) of the OB, while the MRN is the main source of serotonin in the crucial glomerular layer (GL), suggesting a more nuanced role for serotonin in olfactory processing.

What Problem Does This Paper Address?

The olfactory bulb (OB) is the first major processing center for smell, and its function is known to be modulated by serotonin, a neurotransmitter originating from the raphe nuclei in the brainstem. Serotonergic fibers are particularly dense in the glomerular layer (GL) of the OB, where they are thought to play a critical role in gating incoming olfactory signals. While it was known that both the dorsal raphe nucleus (DRN) and the median raphe nucleus (MRN) project to the OB, their specific laminar targets within the OB were not clearly distinguished. This ambiguity made it difficult to understand how serotonin from these distinct nuclei might differentially influence olfactory processing. This paper aims to resolve this by precisely mapping the projections of the DRN and MRN to the different layers of the OB.

Methods

The researchers employed a sophisticated anterograde tracing technique using recombinant adeno-associated viruses (rAAVs). They engineered rAAVs to express enhanced green fluorescent protein (EGFP) fused to synaptophysin, a protein found in presynaptic terminals. This allowed them to visualize the projection terminals of neurons from specific raphe nuclei.

  1. Targeted Viral Injections: Small volumes of rAAVs were stereotactically injected into either the DRN or the MRN of mice. This targeted approach ensured that the viral vectors primarily infected neurons within the chosen nucleus.
  2. Anterograde Tracing: Once injected, the rAAVs were transported down the axons of the infected neurons, leading to the expression of EGFP-synaptophysin at their presynaptic terminals throughout the OB.
  3. Immunohistochemistry: Brain tissue was then processed using immunohistochemistry to detect both EGFP (marking the viral-traced projections) and serotonin (confirming the serotonergic nature of the neurons).
  4. Confocal Microscopy and Image Analysis: High-resolution confocal microscopy was used to image the OB. Sophisticated image processing and quantitative analysis, including colocalization studies and signal density measurements across different OB layers (GL, EPL, GCL), were performed to precisely map the termination patterns of DRN and MRN projections.

Datasets

Not discussed in the paper. The study utilized in vivo experiments on mice and subsequent tissue analysis.

Key Insights and Ideas

  • Distinct Laminar Targeting: The most striking insight is that the DRN and MRN, despite both being serotonergic and projecting to the OB, target different layers. This suggests a functional division of labor.
  • MRN's Role in the GL: The MRN is identified as the primary source of serotonergic innervation to the glomerular layer (GL), the initial processing site for olfactory sensory input. This implicates the MRN in gating or modulating the very first stages of smell perception.
  • DRN's Role in the GCL: Conversely, the DRN's projections are predominantly found in the granule cell layer (GCL), a deeper layer involved in more complex signal processing and integration. This suggests the DRN's influence is on later stages of olfactory processing.
  • Challenging Existing Assumptions: The findings directly challenge the long-held assumption that the DRN is a major contributor to GL innervation. This highlights the importance of precise tracing techniques to refine our understanding of neural circuits.
  • Nucleus-Specific Neuromodulation: The study strongly supports the idea that distinct raphe nuclei can exert nucleus-specific neuromodulatory effects on different OB layers, potentially enabling fine-tuned control over olfactory processing.

Key Results

  • Serotonin in OB Layers: Serotonin immunofluorescence was densest in the GL (0.014 ± 0.003 µm²/100 µm²), with sparser labeling in the EPL and GCL.
  • DRN Projections: Injections into the DRN resulted in dense EGFP labeling primarily in the GCL (0.012 ± 0.004 µm²/100 µm²), with minimal labeling in the GL and EPL.
  • MRN Projections: Injections into the MRN led to dense EGFP labeling in the GL (0.013 ± 0.004 µm²/100 µm²), with less labeling in the EPL and GCL.
  • Colocalization: EGFP signals from MRN injections showed partial colocalization with serotonin in the GL, supporting the MRN as the source of GL innervation.
  • Correlation: A strong correlation (r = 0.84, P = 0.017) was found between the relative abundance of EGFP signals in the GL vs. GCL and the relative infection levels in the MRN vs. DRN, further solidifying the nucleus-specific projections.

Limitations and Future Work

  • Potential for Unintended Infection: While efforts were made to target specific nuclei, the authors acknowledge the possibility of some minor, unintended infection of nearby nuclei, particularly when injecting the MRN, which is located close to the DRN.
  • Ambiguity in DRN GCL Innervation Transmitter: While DRN projections were dense in the GCL, the colocalization with serotonin was not as clear as with MRN projections in the GL. The authors suggest that the DRN might use other neurotransmitters in the GCL, or that detecting sparse serotonin labeling in this layer is challenging.
  • Functional Implications: The paper establishes the anatomical pathways but calls for further investigation into the functional consequences of these distinct projections. How do MRN-mediated GL modulation and DRN-mediated GCL modulation differentially impact olfactory perception and behavior?
  • Other OB Layers: Innervation in the external plexiform layer (EPL) was low for both DRN and MRN injections, suggesting other sources or less significant serotonergic input to this layer.
  • Future Directions: The authors suggest that future work should explore how these distinct projection patterns translate to specific functional outcomes, potentially by manipulating the activity of the DRN and MRN independently and observing the effects on olfactory processing and behavior. They also propose investigating whether the DRN uses other neurotransmitters in the GCL.

Summary updated

Abstract

The raphe nuclei provide serotonergic innervation widely in the brain, thought to mediate a variety of neuromodulatory effects. The mammalian olfactory bulb (OB) is a prominent recipient of serotonergic fibers, particularly in the glomerular layer (GL), where they are thought to gate incoming signals from the olfactory nerve. The dorsal raphe nucleus (DRN) and the median raphe nucleus (MRN) are known to densely innervate the OB. The majority of such projections are thought to terminate in the GL, but this has not been explicitly tested. We sought to investigate this using recombinant adeno-associated viruses (rAAV)-mediated expression of green fluorescent protein (GFP)-synaptophysin targeted specifically to neurons of the DRN or the MRN. With DRN injections, labeled fibers were found mostly in the granule cell layer (GCL), not the GL. Conversely, dense labeling in the GL was observed with MRN injections, suggesting that the source of GL innervation is the MRN, not the DRN, as previously thought. The two raphe nuclei thus give dual innervation within the OB, with distinct innervation patterns.
BibTeX
@article{steinfeld2014divergent,
  author = {Raphael Steinfeld and Jan T. Herb and Rolf Sprengel and Andreas T. Schaefer and Izumi Fukunaga},
  title = {{Divergent innervation of the olfactory bulb by distinct raphe nuclei}},
  journal = {The Journal of Comparative Neurology},
  year = {2014},
  doi = {10.1002/cne.23713},
}
Figure