, 2013a and Polgár et al., 2013b). We previously found that ∼50% of inhibitory cells in laminae I-II express sst2A, and these can be further subdivided into subpopulations that contain galanin and/or nNOS (which constitute ∼60% of the sst2A-expressing cells and therefore approximately one-third of all the inhibitory neurons). The galanin cells coexpress PPD and are the major source of dynorphin in the superficial laminae (Bröhl et al., 2008 and Sardella et al., 2011).

In addition, we identified two other nonoverlapping groups among inhibitory interneurons that lack sst2A (NPY- and parvalbumin-expressing selleck inhibitor cells). There is already evidence that these neurochemical classes differ, both in their responses to noxious stimuli and in their postsynaptic targets (Todd, 2010, Hughes et al., 2012 and Polgár et al., 2013b). The present results provide further evidence in support of this classification scheme, since Bhlhb5−/− mice show a loss of inhibitory interneurons that is apparently restricted to neurochemically defined populations. We find that B5-I neurons correspond to two (mostly nonoverlapping) subpopulations—those that coexpress galanin and

dynorphin and those that express nNOS. The subpopulation of B5-I neurons that expresses galanin/dynorphin likely uses Dolutegravir GABA as its fast transmitter (Simmons et al., 1995), whereas TCL the B5-I neurons that express nNOS are thought to release GABA and glycine (Spike et al., 1993). Since relief of itch by counterstimuli begins almost instantaneously, we favor the idea that this component is mediated by fast-acting inhibitory transmitters. In contrast, dynorphin, which modulates neuronal

activity via G protein-coupled receptors, may underlie prolonged suppression of itch. A key finding from our study is that the loss of B5-I neurons (which results in an almost complete absence of dynorphin in the spinal cord) has a different phenotypic outcome than loss of dynorphin alone. Thus, Bhlhb5−/− mice show dramatically elevated itch, whereas PPD−/− mice display normal itch sensitivity. This distinction implies that an organism can compensate for the loss of dynorphin, but not for the loss of dynorphin-expressing neurons in the dorsal horn. We speculate that neuromodulatory mechanisms may be particularly amenable to homeostatic compensation ( Doi and Ramirez, 2010). In keeping with this idea, mice lacking either enkephalin or the mu opioid receptor have subtle pain phenotypes ( König et al., 1996 and Matthes et al., 1996), despite the fact that mu opioids are among the most effective analgesics. Adaptation also occurs in response to chronic opioid overexposure, as shown by the tolerance observed in humans and animal models following long-term treatment with opioid analgesics ( Morgan and Christie, 2011 and Williams et al., 2013).