Regional differences in itch transmission

To the Editor

The nervous system allocates disproportional neuronal resources to sensory information critical for an organism’s survival and reproduction. The most striking feature of the somatosensory homunculus is the enlarged area representing the distal limbs, especially the hairless glabrous skin regions such as the palms of the hands and soles of the feet, to facilitate environmental exploration. Heightened tactile and pain sensitivity in glabrous skin has been well established (Mancini et al. 2014). However, whether regional differences exist in itch signal processing is not clear. Interestingly, cholestatic pruritus, itching due to reduced or blocked bile flow, is particularly intense in glabrous skin, suggesting sharpened itch sensitivity to endogenous pruritic stimuli in the glabrous skin (Beuers et al. 2014). Our recent study demonstrated that MrgprC11⁺ DRG sensory neurons mediate both hairy and glabrous skin itch (Steele et al. 2021), providing a cellular target to investigate regional differences in itch. This study revealed heightened itch sensitivity in the glabrous skin, supported by the unique axonal branching pattern of itch-sensing neurons.

We generated MrgprC11^ChR2 mice and employed an optogenetic approach to test itch sensitivity in different skin locations (Figure 1 A–D). Light stimulation of MrgprC11⁺ nerves in trunk hairy skin evoked significant scratching, whereas activation of MrgprC11⁺ nerves in plantar glabrous skin induced strong biting, both of which demonstrate itch sensations. Interestingly, a much lower laser power is required to evoke glabrous skin biting, suggesting heightened itch processing for glabrous skin (Figure 1B). To rule out the possibility that different expression levels of ChR2 contribute to different light threshold, MrgprC11⁺ skin-innervating neurons were retrogradely labeled and collected for real time PCR. Comparable expression levels of ChR2 were observed in trunk skin and glabrous skin-innervating neurons (Figure 1E). Consistently, we found that plantar glabrous skin is much more sensitive to Bam8–22, a MrgprC11 agonist, than other skin locations. Although 10 μM of Bam8–22 is enough to evoke robust biting behavior towards glabrous skin, at least 1.5mM and 0.3mM of Bam8–22 is required to induce scratching behavior towards thoracic back skin and the nape of the neck skin respectively (Figure 1F–G).

Figure 1.

Higher itch sensitivity of glabrous skin. (A) Generation of MrgprC11^ChR2 mice. DRG or hindpaw plantar skin sections showing the expression of ChR2(H134R)-EYFP in MrgprC11⁺ neurons and epidermal skin nerves. Epi: epidermis. Der: dermis. Scale bar: 50 μm. (B) Percentage of responses induced by different intensities of 473nm blue light in MrgprC11^ChR2 mice. A much lower laser power is required to evoke glabrous skin biting. (C) Littermate control R26^ChR2 mice did not show any nocifensive behaviors after blue light application. (D) Pictures showing blue light application to the trunk hairy skin (top) or glabrous skin (bottom) of MrgprC11^ChR2 mice. (E) Real Time PCR analysis showing the expression levels of ChR2 are comparable in glabrous skin-innervating and trunk hairy skin-innervating MrgprC11⁺ neurons. (F-H) Concentrations of Bam8–22, a MrgprC11 agonist, required to induce itch behaviors in glabrous skin (F), thoracic trunk skin (G), and nape of the neck skin (H). Data are represented as mean ± SEM. Welch’s t-test for B and E. One way ANOVA for F-H. * p<0.05, ** p<0.01.

To gain mechanistic understanding of the cellular basis of regional differences in itch transmission we performed morphological analysis of MrgprC11⁺ nerves. We first quantified PGP9.5⁺ and MrgprC11⁺ intraepidermal nerve density (IEND) in different skin locations in MrgprC11^tdTomato mice. Both PGP9.5⁺ and MrgprC11⁺ IENDs in paw skin, including hairy and glabrous sides, are significantly lower compared to those in thoracic trunk skin (Figure 2A–B). This suggests that heightened itch sensitivity is not attributable to an increased density of peripheral nerves.

Figure 2.

Regionally distinct morphological organization of itch-sensing arbors in both skin and spinal cord. (A-B) PGP9.5⁺ (A) and MrgprC11⁺ (B) intraepidermal nerve density (IEND) in different body locations. (C) Whole-mount PLAP histochemistry of a lumbar DRG from MrgprC11^IAP mice without tamoxifen treatment. (D) MrgprC11⁺ arbors in the trunk skin and hindpaw plantar glabrous skin. (E) MrgprC11⁺ arbor area in different body locations. (F) MrgprB4⁺ arbor area in different body locations. (G) Spinal cord segments from AAV2/1-CMV-FLEX-PLAP injected MrgprC11^CreER mice showing sparsely labeled MrgprC11⁺ central arbors. Skin locations injected were indicated. Two or three injections were performed in each skin location to achieve labeling of at least one DRG neuron. More than one central arbor were observed in some spinal cords. Hindpaw glabrous skin-innervating neurons exhibit round arbors in the medial lumbar spinal region. Long arbors were observed when AAV2/1-CMV-FLEX-PLAP was injected into the thoracic back skin or proximal hindlimb skin. (H-I) Quantification of the Height/Width ratio (H) and arbor area (I). Data are represented as mean ± SEM. Welch’s t-test for F. One way ANOVA for other panels. * p<0.05. ** p<0.01. *** p<0.005. **** p<0.001. ns: not significant. A: Anterior, P: Posterior. Scale bar in D and G: 500 μm.

We next examined MrgprC11⁺ axonal arborizations in the skin and the spinal cord. CreERT2 in MrgprC11^CreERT2;R26^IAP mice generates background recombination in approximately 1% of DRG sensory neurons without tamoxifen treatment (Xing et al. 2021), allowing visualization of individual axonal arborizations in the trunk hairy skin (Figure 2C–D) and spinal cord Figure S1). CreERT2 background recombination does not label any glabrous skin-innervating neurons; therefore, skin arbors in the hindpaw glabrous skin were visualized after a low dose tamoxifen treatment (Figure 2D). MrgprC11⁺ skin arbors in different locations are all typical “free endings” featuring extensive axonal branching inside of the arbor. However, the arbors are larger in the paws and the largest ones are observed in hindpaw glabrous skin (Figure 2E). To understand if other subtypes of sensory neurons exhibit similar regional effects, we examined MrgprB4⁺ skin arbors using MrgprB4^PLAP mice. MrgprB4⁺ neurons are C-fiber touch sensors detecting stroking of hairy skin and do not innervate the glabrous skin (Liu et al. 2007; Vrontou et al. 2013). Interestingly, MrgprB4⁺ arbors exhibit the opposite trend with significantly smaller arbors in the paw skin (Figure 2F). In addition, previous study showed that MrgprD⁺ sensory neurons, nociceptors mediating pain sensation in glabrous skin, exhibit comparable skin arbor sizes in trunk and plantar glabrous skin (Olson et al. 2017). These results suggest that different types of skin sensory arbors have distinct regional effects.

We also found that glabrous skin-innervating MrgprC11⁺ neurons exhibit distinct central arbor morphology in the spinal cord. Whole mount PLAP histochemistry of the spinal cord isolated from MrgprC11^IAP mice without tamoxifen treatment allowed us to visualize itch-sensing central arbors across the somatotopic map (Figure S1). Central projections of sensory neurons are somatotopically organized along the rostrocaudal and mediolateral axis of the spinal cord. MrgprC11⁺ central axons normally turn rostrally after entering the spinal cord, terminate within the segment of entry, and arborize in the superficial dorsal horn (Figure S1A–D). Arbors in the spinal regions controlling the trunk (thoracic) and proximal limbs (lateral lumbar) show long morphology with rostrocaudal elongation and mediolateral compression, which is consistent with the previously examined C-type arbors (Sugiura et al. 1993). However, arbors showing round morphology with much lower Height/Width ratio are observed in regions controlling the distal parts of the body including head/face (upper cervical and medulla), distal limbs (medial cervical and lumbar), tail, and genital area (sacral) (Figure S1B–D). Arbors in the sacral region are notably larger, while those in the other segments show similar areas (Figure S1G). Since CreERT2 background recombination does not label glabrous skin-innervating neurons, round MrgprC11⁺ central arbors in the medial lumbar region likely represent neurons innervating the hairy side of the hindpaw and ankle area.

To directly correlate central arbor morphologies and the represented skin locations, we injected AAV2/1-CMV-FLEX-PLAP into the skin of MrgprC11^CreERT2 mice. MrgprC11⁺ central arbors were exclusively observed in spinal segments corresponding to injected skin locations (Figure 2G). Round arbors were observed in medial lumbar segments following injections into the hindpaw glabrous skin, while long arbors appeared in thoracic and lateral lumbar segments after injections into the thoracic and proximal hindlimb skin respectively (Figure 2G). Round and long arbors share a similar area (Figure 2H–I).

In summary, our results demonstrate regional differences in itch processing, with higher itch sensitivity in glabrous skin, aligning with the regionally distinct morphological organization of itch-sensing neurons. MrgprC11⁺ neurons exhibit low IEND and larger axonal arborization in glabrous skin, and regionally distinct central arbors in the spinal cord. Previous studies have linked regional-specific central arbor morphology to enhanced signal transmission of touch and pain in glabrous skin (Lehnert et al. 2021; Olson et al. 2017). Conversely, peripheral nerve organization contrasts with higher innervation density and smaller receptive field observed in touch sensors of distal limbs (Brown and Koerber 1978). Taken together, our findings suggest that region-specific morphological organization serves as a fundamental somatotopic mechanism to facilitate regional differences in sensory processing, although distinct organizations were employed by different sensory modalities.

Data availability statement

All data are included in the manuscript. No large datasets were generated or analyzed during this study.