Correlation of Peripheral Innervation Density and Dorsal Horn Map Scale

Abstract

Dorsal horn map scale and peripheral innervation density were compared to test a hypothesized linear relationship. In anesthetized cats, low-threshold mechanoreceptive peripheral nerve innervation fields (IFs) were measured by outlining areas of skin from which action potentials could be elicited in cutaneous nerves. The same nerves were processed histologically and used to count myelinated axons. Innervation density for each nerve was calculated as number of axons divided by IF area. Single units were recorded throughout the hindlimb representation, in laminae III and IV. These data, combined with single-unit data from other animals and with cell counts in laminae III and IV, permitted estimation of numbers of cells whose receptive field centers fell in contiguous 1-cm bands from tips of toes to proximal thigh. A similar estimate was performed with the use of the nerve innervation data, so that peripheral innervation densities and map scales for the different 1-cm bands of skin could be compared. Correlation between the two was quite high (r = 0.8), and highly significant (P = 2.5 x 10(-7)). These results are consistent with a proposed developmental model in which map scale, peripheral innervation density, and reciprocal of dorsal horn cell receptive field size are mutually proportional, as a result of developmental mechanisms that produce constant divergence and convergence between primary afferent axons and dorsal horn cells.

Spatial convergence and divergence between cutaneous afferent axons and dorsal horn cells are not constant

“…The cell density distribution (Fig. 5) differed from the previous study (Wang et al, 1997). The main differences were higher densities at all locations, with less of a drop-off in the medial and lateral dorsal horn and in rostral segments.…”

“…Figure 4C illustrates the result obtained with a larger smoothing factor, ϭ 5.0 mm, used to eliminate discontinuities between innervation fields. Figure 4D represents results in the previous study (Wang et al, 1997). The combined data (Fig.…”

“…In an earlier study (Wang et al, 1997), we measured the innervation densities of different cutaneous nerves, by counting myelinated axons and dividing by the innervation areas of the nerves. In that study, we were able to achieve resolution sufficient to describe innervation density in 1-cm circumferential bands along the length of the leg but not sufficient for estimates of convergence and divergence.…”

“…Rostrocaudal (RC) and mediolateral (ML) recording locations were expressed in normalized form, as fractional segmental and mediolateral locations (e.g., RC ϭ L 7.24 , ML ϭ 0.36, for a site 0.24ϫ the distance from the rostral to the caudal end of L 7 , 0.36ϫ the distance from the medial to the lateral edge of the dorsal horn). The recording locations and RFs of the single units were added to our database of 551 cells (Koerber et al, 1993;Koerber and Brown, 1995;Wang et al, 1997;Brown et al, 1998) for purposes of modeling somatotopy. This was accomplished by recalculating the descriptive model of the lamina III-IV cells including the new cell data.…”

“…The RC, ML location of each identified cell was recorded. The cell locations were used to produce a cell density surface, by using exponential smoothing in the RC, ML plane (see Wang et al, 1997, for details of calculations). The cell densities were corrected for nucleoli on adjacent sections, multiplying by m/(m ϩ n), where m is the section thickness and n is the average diameter of nucleoli (Abercrombie, 1946).…”