Soma size distinguishes projection neurons from neurokinin 1 receptor-expressing interneurons in lamina I of the rat lumbar spinal dorsal horn

Abstract

Lamina I of the spinal dorsal horn contains neurons that project to various brain regions, and approximately 80% of these projection cells express the neurokinin 1 receptor (NK1r), the main receptor for substance P. Two populations of NK1r-immunoreactive neurons have been identified in lamina I: small weakly immunoreactive cells and large cells with strong immunolabelling [Cheunsuang O and Morris R (2000) Neuroscience 97:335-345]. The main aim of this study was to test the hypothesis that the large cells are projection neurons and that the small cells are interneurons. Projection neurons were identified by injection of tracers into the caudal ventrolateral medulla and lateral parabrachial area, and this was combined with immunostaining for NK1r. We found a bimodal size distribution for NK1r-immunoreactive neurons. The small cells (with somatic cross-sectional areas <200 microm(2)) showed weak immunoreactivity, while immunostaining intensity was variable among the large cells. Virtually all (99%) of the immunoreactive cells with soma areas >200 microm(2) were retrogradely labelled, while only 10% of retrogradely labelled cells were smaller than this. Soma sizes of retrogradely labelled neurons that lacked NK1r did not differ from those of NK1r-expressing projection neurons. It has been suggested that a population of small pyramidal projection neurons that lack NK1r may correspond to cells activated by innocuous cooling, and we therefore assessed the morphology of retrogradely labelled cells that were not NK1r-immunoreactive. Fifteen percent of these were pyramidal, but these did not differ in size from pyramidal NK1r-immunoreactive projection neurons. These results confirm that large NK1r-immunoreactive lamina I neurons are projection cells, and suggest that the small cells are interneurons. Since almost all of the NK1r-immunoreactive cells with soma size >200 microm(2) were retrogradely labelled, cells of this type can be identified as projection cells in anatomical studies.

Fig. 1

Fluorogold and CTb injection sites in the three experiments. The drawings show the spread of tracer in each of these experiments. Each vertical column represents a single experiment, and the experiment number is shown at the bottom of the column. Numbers on the left give the approximate antero-posterior positions of the sections relative to the interaural plane. Drawings are based on those of Paxinos and Watson (2005). The upper four outlines in each column represent the Fluorogold injection (targeted on the LPb), while the lower five show the spread of CTb (injected into the CVLM). CnF, cuneiform nucleus; Cu, cuneate nucleus; IC, inferior colliculus; KF, Kölliker-Fuse nucleus; LPB, lateral parabrachial nucleus; LRt, lateral reticular nucleus; MPB, medial parabrachial nucleus; PAG, periaqueductal grey matter; py, pyramidal tract; scp, superior cerebellar peduncle; Sol, nucleus of the solitary tract; Sp5, spinal trigeminal nucleus.

Fig. 2

Examples of CTb and Fluorogold injection sites. (a): section (interaural ∼−5.3 mm) through the CVLM injection in experiment 2 which had been reacted with an immunoperoxidase method to reveal CTb. (b, c): fluorescent and brightfield photomicrographs of a section (interaural ∼−0.2 mm) through the Fluorogold injection in experiment 3. The spread of tracer is indicated by arrowheads. Scale bar=1 mm.

Fig. 3

Immunoreactivity for NK1r and the two retrograde tracers in a horizontal section from the L4 segment of experiment 2. (a) shows a field scanned to reveal NK1r-immunoreactivity (green), (b) has been scanned for CTb (red) and Fluorogold (FG, blue), while (c) is a merged image. The NK1r-immunoreactivity is associated with thin elongated profiles, which are dendrites, as well as with larger structures, which are cell bodies. This field contains the cell bodies of five NK1r-immunoreactive projection neurons (asterisks, numbered 1–5) that were retrogradely labelled from both CVLM and LPb, and therefore contain both tracers. In addition, several smaller NK1r-immunoreactive cell bodies are visible, and these are not retrogradely labelled. Six of these cells are indicated with arrowheads (numbered 6–11). Cells 1 and 5 were assigned a NK1r-immunoreactivity score of 3, cells 2–4 and 6–7 had a score of 2 and cells 8–11 a score of 1. The images are projections of two optical sections at 2 μm z-separation. Scale bar=20 μm.

Fig. 4

Examples of retrogradely labelled neurons in a horizontal section from L4 in experiment 1. (a) is a projected image through the cell bodies of several retrogradely labelled cells, with CTb shown in red and Fluorogold in blue. Most cells have taken up both tracers and therefore appear pink, while the cell numbered 1 is labelled only with Fluorogold, and the one marked with an asterisk is labelled only with CTb. The three numbered cells are pyramidal in shape and single optical sections through the cell bodies of each of these are shown in (b–g). From these images it can be seen that cells 1 and 2 are NK1r-immunoreactive, while cell 3 is not. Cell 1 was assigned a NK1r-immunoreactivity score of 3 and cell 2 was given a score of 4. The image in (a) is a projection of eight optical sections at 2 μm z-spacing. Scale bar=20 μm.

Fig. 5

Frequency histograms showing the soma cross-sectional areas for different groups of lamina I neuron: all of the NK1r-immunoreactive ones (All NK1r⁺, n=1341), those that were NK1r-immunoreactive but not retrogradely labelled (NK1r⁺ non-retrograde, n=900), those that were NK1r-immunoreactive and retrogradely labelled (NK1r⁺ retrograde, n=441) and those that were retrogradely labelled and not NK1r-immunoreactive (NK1r⁻ retrograde, n=101). In each case, the y-axis represents percentage. The dashed line corresponds to a cross-sectional area of 200 μm². The NK1r-immunoreactive neurons show a clear bimodal distribution, with the first and second peaks corresponding to the non-retrogradely labelled and the retrogradely labelled populations, respectively.

Fig. 6

Drawings of five of the 15 retrogradely labelled pyramidal cells that were not NK1r-immunoreactive. Cell 1 corresponds to cell 3 in Fig. 4. Three of these cells (1, 3 and 4) each give rise to three primary dendrites. Cell 2 has two primary dendrites originating from one pole of the soma, while cell 5 has an additional very fine dendrite originating from the soma. Scale bar=20 μm.

Fig. 7

Box and whisker plot of the soma sizes of retrogradely labelled pyramidal cells that were NK1r-immunoreactive (NK1r⁺, n=37) or non-immunoreactive (NK1r⁻, n=33). The boxes represent the median and interquartile range, while the upper and lower error bars show the 90th and 10th percentiles and the filled symbols are values outside these ranges. Data were taken from the present study and that of Al-Khater and Todd (2009). For further details, see text.