Unimodal primary sensory cortices are directly connected by long-range horizontal projections in the rat sensory cortex.

Jimmy Stehberg,Phat T. Dang,Ron D. Frostig

doi:10.3389/fnana.2014.00093

Jimmy Stehberg, Phat T. Dang + Show 1 more

Open Access

https://doi.org/10.3389/fnana.2014.00093

Copy DOI

Abstract

Research based on functional imaging and neuronal recordings in the barrel cortex subdivision of primary somatosensory cortex (SI) of the adult rat has revealed novel aspects of structure-function relationships in this cortex. Specifically, it has demonstrated that single whisker stimulation evokes subthreshold neuronal activity that spreads symmetrically within gray matter from the appropriate barrel area, crosses cytoarchitectural borders of SI and reaches deeply into other unimodal primary cortices such as primary auditory (AI) and primary visual (VI). It was further demonstrated that this spread is supported by a spatially matching underlying diffuse network of border-crossing, long-range projections that could also reach deeply into AI and VI. Here we seek to determine whether such a network of border-crossing, long-range projections is unique to barrel cortex or characterizes also other primary, unimodal sensory cortices and therefore could directly connect them. Using anterograde (BDA) and retrograde (CTb) tract-tracing techniques, we demonstrate that such diffuse horizontal networks directly and mutually connect VI, AI and SI. These findings suggest that diffuse, border-crossing axonal projections connecting directly primary cortices are an important organizational motif common to all major primary sensory cortices in the rat. Potential implications of these findings for topics including cortical structure-function relationships, multisensory integration, functional imaging, and cortical parcellation are discussed.

Highlights

The classical description of the structural organization of the neocortex is based on the key concept of cortical tissue parcellation into different regions, areas, or subareas, where each such unit can be typically delineated using cytoarchitectonic or myeloarchitectonic histology
To increase our sampling probability, we opted for a combination of slices from layers 2 and 3 and in some cases from layer 5, in addition to the use of a variety of injection sizes (Table 1). While these steps have mitigated the under-sampling problem, one has to keep in mind that not all layer 2–3 sections were included in the analysis, layer 4 slices were excluded as they were used for CO analysis and only rarely were layer 5 sections included even though layer 5 slices always exhibited dense patterns of long-range projections within SI
The projections described here are only a fraction of the total border-crossing projections to be found in cortex

Summary

Introduction

The classical description of the structural organization of the neocortex (hereafter referred to as cortex) is based on the key concept of cortical tissue parcellation into different regions, areas, or subareas, where each such unit can be typically delineated using cytoarchitectonic or myeloarchitectonic histology. Parcellation by cytoarchitectonic- or myeloarchitectonic-based histology, especially in the extensively studied human cortex, has had a history of extreme variability in findings (Campbell: 14 areas, Broadmann: 44 areas, von Economo and Koskinas: 54 areas, Vogt and Vogt: >200 areas, Bailey and von Bonin: 8 areas, and Sarkissov and colleagues: 52 areas; reviewed by Nieuwenhuys et al, 2008; see Zilles and Palomero-Gallagher, 2001 and Van Essen et al, 2012) Despite such perplexing variability, the basic concept of parcellation is still considered fundamental for the description of cortical organization. For recent review on cortical parcellation in human, macaque and mouse see Van Essen (2013)

Methods

Results

Discussion

Conclusion