Transgenic Mouse Models for the Tracing of “Pain” Pathways

The traditional, textbook view of the “pain” pathway illustrates an unmyelinated primary afferent C-fiber, the nociceptor, contacting a second-order dorsal horn neuron at the origin of the spinothalamic and spinoreticular pathways. Although the ultimate cortical target of these different pathways is unclear, there is no question that a better understanding of the mechanisms through which noxious stimuli produce pain requires a better understanding of these circuits. The limitations of our knowledge, of course, go beyond the need to identify cortical targets. We recognize now that there are neurochemically and physiologically distinct populations of afferents, projection neurons, and diverse central nervous system (CNS) targets. In fact, even the classification of nociceptors into and categories is oversimplified (Snider and McMahon 1998). Thus, an array of transient receptor potential (TRP) channels, which respond to different temperatures, natural products, or environmental irritants, establishes subcategories of nociceptors, as do the various Na+ channel subtypes (Caterina and Julius 1999; McCleskey and Gold 1999; Cummins et al. 2007). Even the rather broad categorization of myelinated versus unmyelinated nociceptor is but a first approximation to the diversity of afferent fibers that transmit “pain” messages (Talavera et al. 2008). Furthermore, the spinal cord is also far more complicated and contains various classes of projection neurons (Todd 2002; Morris et al. 2004; Klop et al. 2005), which not only are differentially distributed in the gray matter (e.g., laminae I, V, VII, and X) but also differ in the selectivity of their responses to non-noxious and noxious stimuli, in their receptive field sizes, and in their central targets. What is still not clear, however, is the extent to which there are unique functional correlates of these neurochemically distinct populations of neurons along the pain pathway. For example, it is still not clear to what extent distinct classes of afferents differ in the types of pain provoked by their activation. Of particular interest is the differential contribution of neurons of laminae I and V to nociceptive processing. Some groups argue that only the lamina I neurons are essential for the highly selective discriminative aspect of the pain experience, and that the wide-dynamic-range neurons of lamina V are primarily contributors to sensorimotor integration (Craig 2004). Others argue for an essential contribution of the lamina V neurons (Price et al. 2003; Martin et al. 2004; Mazario and Basbaum 2007). Also unknown is the extent to which subpopulations of dorsal root and trigeminal ganglion (DRG and TG) neurons feed into these sensory-discriminative and limbic/emotional processing regions of the brain, especially in light of the relatively recent discovery of major spinohypothalamic (Burstein et al. 1987; Giesler et al. 1994) and spinoparabrachial-amygdala pathways (Bernard et al. 1989; Bernard and Besson 1990; Jasmin et al. 1997) in addition to the more traditional spinothalamic and spinoreticulothalamic systems. That neurochemically distinct populations of nociceptors indeed access different central circuits is illustrated by the demonstration that the major classes of nociceptors differ in their patterns of axon termination in the spinal cord dorsal horn. The peptide population terminates almost exclusively in the outer laminae of the superficial dorsal horn (laminae I and outer II), targeting projection neurons that transmit nociceptive messages to brainstem and/or thalamus; by contrast, the IB4 population primarily targets interneurons of the inner part of lamina II, a region just dorsal to a distinct subset of interneurons that synthesize the gamma isoform of protein kinase C (PKCγ) (Malmberg et al. 1997). Finally, myelinated neurons project primarily to deeper laminae (III-VII) of the spinal cord (and to a smaller extent to lamina I). These observations support the view that different classes of nociceptors indeed wire to different CNS circuits. Based on the remarkable electrophysiological specificity of afferents and their neurochemical distinctiveness, there is now a general consensus for specificity (i.e., labeled line) features to the afferent, at least with respect to response properties. But whether these afferents converge upon functionally distinct but related populations of projection neuron, resulting in functionally segregated ascending circuits, or whether there is convergence upon populations of projection neurons with common functional properties remains to be determined. Unfortunately, the information about these circuits is extremely limited. Not only is the identity of the neuron immediately postsynaptic to the different nociceptors inadequately specified, but the neurons and circuits that lie downstream of the first synapse in the dorsal horn are also largely uncharacterized. In some cases, the identity of postsynaptic neurons has been determined by electrophysiological analyses, and synapses have been characterized at the electron microscopic level (Westlund et al. 1992; Alvarez et al. 2004; Hwang et al. 2004; Shields et al. 2007; Neumann et al. 2008), but the sample from which the information is derived is extremely small. Studies that monitor Fos expression provide a much more extensive picture of populations of neurons activated by noxious stimuli (Menétrey et al. 1989; Abbadie et al. 1997; Neumann et al. 2008), but there is no information about the circuits that underlie Fos activation. Also unclear are the third-order neurons to which the laminae I and V neurons project. With some exceptions, the map of the intervening circuits is, in fact, largely unknown.