Different distributions of calbindin and calretinin immunostaining across the medial and dorsal divisions of the mouse medial geniculate body

Abstract

We studied the distributions of calretinin and calbindin immunoreactivity in subdivisions of the mouse medial geniculate body and the adjacent paralaminar nuclei. We found that the vast majority of labeled cells in the dorsal division of the medial geniculate body were immunoreactive for calbindin-only, whereas most of the remaining labeled cells were double-labeled. Very few calretinin+ only cells were observed. By contrast, we observed significant proportions of calbindin+ only, calretinin+ only and double-labeled cells in the medial division of the medial geniculate body. Further, the distributions of calbindin-only, calretinin-only and double-labeled cells did not differ between the medial division of the medial geniculate body, the suprageniculate nucleus, the peripeduncular nucleus and the posterior intralaminar nucleus. We found essentially no somatic staining for either calbindin or calretinin in the ventral division of the medial geniculate body. These data suggest that there are distinct neurochemical differences between the two non-lemniscal auditory thalamic nuclei. In addition, these data extend previous observations that the medial division of the medial geniculate body shares many properties with the paralaminar group of nuclei.

Introduction

The presence or absence of several calcium-binding proteins, including calbindin (CB), parvalbumin (PV) and calretinin (CR), has been used to delineate different functional cell types in the neocortex, hippocampus, cerebellum and thalamus (Jones, 1998, Hof et al., 1999, Bastianelli, 2003, Jinno and Kosaka, 2006). The specific roles of these proteins in shaping neuronal activity have yet to be established, though it has been proposed that the differential total calcium-binding capacity and kinetics observed in these proteins can preferentially modulate specific types of calcium currents (Schwaller et al., 2002, Meuth et al., 2005).

Within the thalamus, the distribution of CB and PV are strikingly complementary, and these distributions have been used in the formulation of models of thalamic organization (Jones, 2001). For example, primary sensory thalamic nuclei (lateral geniculate nucleus, ventral posterior medial nucleus, ventral posterior lateral nucleus and the ventral division of the medial geniculate body, MGBv) demonstrate immunostaining for PV, with label found in both somata and in the neuropil. CB staining in these regions is weak or non-existent. Non-primary sensory nuclei, such as the lateral posterior–pulvinar complex, posterior medial nucleus and the dorsal and medial subdivisions of the medial geniculate body (MGBd and MGBm, respectively) show strong somatic immunoreactivity for CB and poor to non-existent PV immunoreactivity (Rausell et al., 1992, de Venecia et al., 1995, Morel et al., 1997, Jones, 1998, Cruikshank et al., 2001). More recently, CR-immunoreactivity was demonstrated in the thalamus in several species, and was shown to generally have a similar distribution as CB in most thalamic nuclei (Arai et al., 1994, Fortin et al., 1998, Hof et al., 1999, FitzGibbon et al., 2000, Münkle et al., 2000, González et al., 2002), though CR positivity appears to be particularly prominent in the intralaminar and midline groups of nuclei (Arai et al., 1994, Oda et al., 2004, Uroz et al., 2004). Therefore, it appears that both CB and CR may be markers for the non-primary sensory thalamic nuclei.

This differential distribution of thalamic CB/CR and PV corresponds to differences in the presumed roles of these nuclei. For example, neurons in non-primary sensory thalamic nuclei receive large-terminal afferents, in part, from cortical layer 5 and have been referred to as “higher-order” nuclei (Sherman and Guillery, 2002). It has been proposed that higher-order thalamic nuclei receive receptive field information from one cortical area and relay it to another (Guillery, 1995). In contrast to the higher-order nuclei, primary sensory nuclei receive receptive field information from the sensory periphery and relay this information to the cortex, and have been referred to as “first-order” nuclei. For further discussion of first and higher-order thalamic nuclei, see (Sherman and Guillery, 2005). Thus, it appears that in the sensory thalamic nuclei, CB/CR and PV positivity may correspond to higher-order and first-order thalamic nuclei, respectively.

Though CR and CB have been observed in similar groups of higher-order nuclei, it is not known if CB and CR colocalize to the same population of neurons, or if separate populations of CB- and CR-positive neurons exist. The answer to this question has potentially important implications on our understanding of the organization of higher-order thalamic nuclei, since there is evidence for connectional and functional heterogeneity within higher-order thalamic nuclei. For example, many higher-order thalamic nuclei receive large-terminal afferents from both cortical and subcortical structures, raising the possibility that higher-order circuits, driven by cortical inputs, may reside in the same nuclei of first-order circuits, driven by subcortical inputs. In addition, though the prototypical projection of thalamic principal neurons is to layers 4 and 6 of neocortex, many thalamic cells in higher-order nuclei project to layer 1 of neocortex (Rockland et al., 1999) or to other subcortical structures, such as the basal ganglia or amygdala (Harting et al., 2001, Cheatwood et al., 2003). This projection pattern is particularly prevalent among neurons in the intralaminar and the adjacent “paralaminar” nuclei (Herkenham, 1980), such as the suprageniculate (SG), posterior intralaminar nucleus (PIN) and the peripeduncular nucleus (PP, (Ryugo and Killackey, 1974, Ottersen and Ben-Ari, 1979, Clugnet et al., 1990)). The degree to which this heterogeneity may be reflected in different distributions of calcium-binding proteins has not yet been addressed.

In the current study, we take advantage of the tripartite organization of the auditory thalamus, which contains a PV-rich MGBv, and two CB/CR rich nuclei: the MGBd and MGBm. The MGBd receives large-terminal afferents from both the auditory cortex and external nuclei of the inferior colliculus and sends projections to layers 1, 4 and 6 of the secondary auditory cortical fields and layer 1 of the primary AC (Llano and Sherman, 2008). The MGBm receives input from the inferior colliculus, superior colliculus and spinal cord, and projects to layers 1 and 6 of cortex as well as to the basal ganglia and to the amygdala (Calford and Aitkin, 1983, Bordi and LeDoux, 1994, Linke, 1999). The MGBm also displays significant heterogeneity with respect to the intrinsic properties such that some neurons do not display bursting and there are subpopulations of neurons in the MGBm with large, reticular, non-bushy morphology (Winer and Morest, 1983, Smith et al., 2006). Given the functional and connectional heterogeneity in these nuclei, we hypothesized that distinct populations of CB-positive and CR-positive cells would be found in the MGBd and MGBm, and that the relative proportion of each cell type would differ.