(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Elevated vesicular Zn2+ in dorsal root ganglion neurons expressing the transporter TMEM163 causes age-associated itchy skin in mice [1]

['Fang Tong', 'Shanghai Stomatological Hospital', 'School Of Stomatology', 'State Key Laboratory Of Medical Neurobiology', 'Moe Frontiers Center For Brain Science', 'Institutes Of Brain Science', 'Institutes Of Biomedical Sciences', 'Fudan University', 'Shanghai', 'Shuai Liu']

Date: 2024-12

The prevalent itching condition associated with aging, historically referred to as senile pruritus, diminishes quality of life. Despite its impact, effective treatments remain elusive, largely due to an incomplete understanding of its pathological cause. In this study, we reveal a subset of dorsal root ganglion neurons enriched with Zn 2+ that express the vesicular Zn 2+ transporter TMEM163. These neurons form direct synapses with and modulate the activity of spinal NPY + inhibitory interneurons. In aged mice, both the expression of TMEM163 and the concentration of vesicular Zn 2+ within the central terminals of TMEM163 + primary afferents show marked elevation. Importantly, the excessive release of vesicular Zn 2+ significantly dampens the activity of NPY + neurons, triggering the disinhibition of itch-transmitting neural circuits and resulting in chronic itch. Intriguingly, chelating Zn 2+ within the spinal dorsal horn effectively relieves itch in aged mice. Our study thus unveils a novel molecular mechanism underlying senile pruritus.

In this study, we found that there is one subpopulation of DRG neurons that express the vesicular Zn 2+ transporter TMEM163. These neurons can form monosynaptic connections with spinal NPY + -derived inhibitory INs and modulate their neuronal activity by releasing Zn 2+ . During aging, both the expression of TMEM163 and the concentration of vesicular Zn 2+ stored in the central terminals of TMEM163 + DRG neurons significantly increased. Stimulation increased Zn 2+ release into the synaptic cleft, inhibiting the activity of NPY + INs, leading to disinhibition of itch-transmitting neural circuitry and chronic itch. Targeting primary sensory neuron-derived Zn 2+ in the spinal dorsal horn effectively alleviates chronic itch in aged mice.

Zinc (Zn), the second most abundant trace element in the body, serves dual roles as a structural component of proteins and a signaling molecule regulating various vital processes [ 8 ]. The majority of Zn in neurons exists as free Zn 2+ stored in synaptic vesicles [ 8 ]. Vesicular Zn 2+ is released into the synaptic cleft together with glutamate during neuronal firing, modulating synaptic plasticity and neural transmission [ 9 , 10 ]. Mice lacking vesicular Zn 2+ exhibit deficits in various processes, including context recognition, learned fear, extinction, spatial working memory, and cognition [ 9 , 11 – 14 ]. Two distinct classes of Zn 2+ transporters exist: Zrt- and Irt-related proteins (ZIPs), which facilitate the movement of Zn 2+ from extracellular or intracellular organelle compartments into the cytosol, and zinc transporters (ZnTs), which mediate Zn 2+ transport in the opposite direction [ 8 ]. To date, a total of 14 ZIPs and 11 ZnTs have been identified. Among them, TMEM163 (also known as SV31 or ZnT11) was initially characterized as a synaptic vesicle membrane protein and is primarily localized to the vesicular membrane, where it plays a crucial role in the transportation of Zn 2+ from the cytosol into synaptic vesicles [ 15 – 17 ]. NMDA receptors (NMDARs) are one of the key targets of Zn 2+ [ 18 – 20 ]; Zn 2+ binds to NR2A when it is present at low levels (nM) [ 18 ] and to NR2B when it is present at high levels (μm) to inhibit NMDARs [ 21 ]. Mice harboring mutations in the Zn 2+ -binding sites of NR2A were found to exhibit increased heat hyperalgesia and mechanical allodynia in both inflammatory pain and neuropathic pain mouse models. Moreover, the analgesic effect of Zn 2+ is abolished in these mice, indicating that the Zn 2+ and NR2A interaction is essential for the analgesic properties of Zn 2+ [ 22 ]. AMPA receptors (AMPARs) play a critical role in fast excitatory glutamatergic neurotransmission, which can also be inhibited by Zn 2+ released from synaptic vesicles [ 23 ]. Zn 2+ released from dorsal cochlear nucleus (DCN), boutons synapsing onto CA1 neurons, or mossy fibers synapsing onto CA3 neurons inhibits AMPARs [ 24 – 27 ].

Itch represents a distressing sensation and emotional experience that provokes scratching or a strong desire to scratch. Intractable itch, a prevalent skin-related complaint among elderly individuals, significantly decreases their overall quality of life. In the spinal dorsal horn, several subpopulations of excitatory interneurons including gastrin-releasing peptide receptor (GRPR) + interneurons [ 1 , 2 ], Urocortin 3 (UNC3) + interneurons (INs) [ 3 ], Neuropeptide Y receptor type 1 (NPY1R) + interneurons [ 4 ], and Tachykinin 2 (TAC2) + interneurons [ 5 ] have been identified as critical regulator of itch transmission. Recent studies have uncovered a distinct subpopulation of neuropeptide Y (NPY)-positive INs within the spinal cord, which receive input from low-threshold mechanoreceptors (LTMRs) in hairy skin and tightly gate the transmission of itch, and the dysfunction of LTMRs-NPY + INs can cause the pathogenesis of skin lesion and chronic itch [ 3 , 6 , 7 ]. However, the identity of the LTMRs connected to NPY + INs and the molecular mechanisms underlying the loss of inhibitory function in NPY + neurons during chronic itch remains elusive.

In the aged mice, both the amplitude of AMPAR-mediated and NMDAR-mediated EPSCs of NPY + INs significantly reduced ( S8A–S8D Fig ). In dry skin model mice, the frequency of mEPSCs in NPY + INs were dramatically reduced, while the amplitude did not change ( S8E–S8I Fig ). Chelation of spinal Zn 2+ with ZX1 increased the amplitudes of both AMPAR-EPSCs and NMDAR-EPSCs ( S8J–S8M Fig ). These results are consistent with the notion that spinal Zn 2+ inhibits both AMPAR- and NMDAR-mediated EPSCs and modulates the neuronal activity of NPY + neurons via a postsynaptic mechanism. In sharp contrast, neither AMPAR-mediated EPSCs nor NMDAR-mediated EPSCs in Ucn3::Cre neurons were affected by chelation of endogenous Zn 2+ with ZX1, indicating that Ucn3::Cre neurons was not directly inhibited by vesicular Zn 2+ or that the inhibitory effect was not strong enough to affect the synaptic responses of Ucn3::Cre neurons ( S9 Fig ), which is also in line with the retrograde trans-monosynaptic tracing results ( S6E and S6F Fig ). Thus, these data provide evidence that Zn 2+ released from presynaptic vesicles exerts a modulatory effect on the neuronal activity of NPY + neurons through the inhibition of AMPARs and NMDARs.

Next, we investigated whether spinal Zn 2+ stored in the central terminals of Tmem163 + primary afferents modulate the neuronal activity of NPY + INs. Neither the amplitude nor the frequency of mEPSCs was affected by the chelation of Zn 2+ with TPEN (100 μm) or ZX1 (100 μm) ( S7A–S7K Fig ). Next, we sought to investigate the impact of spinal Zn 2+ chelation on the synaptic responses of NPY + inhibitory INs mediated by AMPA receptors (AMPARs) and NMDA receptors (NMDARs). The amplitude of AMPAR-mediated EPSCs (AMPAR-EPSCs) was significantly larger in the presence of ZX1 ( Fig 5D–5F ). We further examined focally evoked NMDAR-mediated EPSCs (NMDAR-EPSCs) in the presence of NBQX (10 μm), bicuculline (10 μm), and strychnine (2 μm) at a holding potential of −40 mV in these 2 groups. The amplitude of NMDAR-EPSCs was also significantly increased in the presence of ZX1 ( Fig 5G–5I ). To investigate the potential involvement of presynaptic mechanisms in the effects of spinal Zn 2+ on evoked excitatory postsynaptic currents (EPSCs), we employed paired-pulse ratio (PPR) analysis, a sensitive measure of presynaptic neurotransmitter release probability [ 46 ]. Chelation of spinal Zn 2+ using ZX1 did not induce any significant changes in the PPR of EPSCs in NPY + INs, indicating that presynaptic mechanisms do not contribute to the Zn 2+ -mediated attenuation of the EPSC amplitude ( S7L–S7N Fig ).

( A ) Schematic diagram illustrating the strategy for retrograde transsynaptic tracing of the upstream neurons of NPY::Cre INs or UCN3::Cre INs in the DRG by intraspinal injection of viruses. NPY + mice or UCN3::Cre mice received intraspinal injection of an AAV vector encoding Cre-dependent RVG (AAV2/5-hEF1a-DIO-RVG-WPRE-pA) and TVA fused with mCherry (AAV2/9-CAG-DIO-TVA-mCherry-WPRE-pA) into laminae III-V. After 3 weeks, EnvA-pseudotyped rabies virus (RV-ENVA-DG-EGFP) was injected for retrograde tracing. ( B ) Representative images showing infected presynaptic neurons (EGFP + ) of NPY + INs in the DRG and their coexpression of Tmem163 using ISH. Scale bar = 100 μm. ( C ) Quantitative analysis of the data in (B). ( D ) Representative eEPSCs in laminae III-V NPY + INs before and after ZX1 application (hold at −70 mV). ( E ) Time course of the change in eEPSC amplitude before and after ZX1 application; n = 9 cells. ( F ) Quantitative analysis of the eEPSC amplitude; paired t test; n = 9 cells/group. ( G ) Representative eEPSCs in laminae III-V NPY + INs before and after ZX1 application (hold at −40 mV). ( H ) Time course of the change in eEPSC amplitude before and after ZX1 application; n = 10 cells. ( I ) Quantitative analysis of the eEPSC amplitude; paired t test; n = 10 cells/group. All data are expressed as the mean ± SEM; *p < 0.05, **p < 0.01. The underlying data for Fig 5C, 5E, 5F, 5H, and 5I can be found in S1 Data . AAV, adeno-associated virus; DRG, dorsal root ganglia; IN, interneuron; ISH, in situ hybridization; SEM, standard error of the mean.

To unravel the specific involvement of vesicular Zn 2+ stored in the central terminals of TMEM163 + primary afferents in mediating aging-related itch, we investigated the postsynaptic neurons of these primary afferents within the spinal cord. Recognizing the critical role of inhibitory INs derived from the NPY + lineage in itch [ 3 ], we aimed to determine whether Tmem163 + primary afferents directly synapse with NPY + spinal inhibitory INs. In the spinal cord of NPY::cre;Ai3 mice, NPY + -derived inhibitory INs, distributed in laminae I-V, were labeled with EGFP ( S5A Fig ). Given that Slc17a7 (which encodes VGluT1) mostly colocalized with Tmem163 in the DRG (Figs 3D and S3C ), we assumed that the primary afferents of VGluT1 + neurons and Tmem163 + neurons in the spinal cord have the same projections and make the same synaptic connections. There were abundant VGluT1 + (Tmem163 + ) nerve terminals surrounding NPY + -derived INs in laminae III-V of the spinal cord ( S5B Fig ), indicating that there was a possible direct synaptic connection between VGluT1 + nerve terminals and NPY + -derived INs. To identify the presynaptic neurons in the DRG that formed direct synapses with NPY + INs, we performed retrograde trans-monosynaptic tracing by injecting an AAV vector encoding a Cre-dependent RVG (AAV2/5-hEF1a-DIO-RVG-WPRE-pA) and TVA fused with mCherry (AAV2/9-CAG-DIO-TVA-mCherry-WPRE-pA) into laminae III-V of NPY-Cre mice and then injected EnvA-pseudotyped rabies virus (RV-ENVA-DG-EGFP) into the mice 3 weeks later ( Fig 5A ). The results showed that 38.2% of mCherry + neurons expressing EGFP in the spinal dorsal horn ( S6A Fig ), and there were EGFP + neurons in the DRG, with 80.09 ± 2.209 (%) of these neurons expressing Tmem163, 81.90 ± 4.448 (%) of them expressing the A-fiber-generating DRG neuron marker NF200, only 2.50 ± 2.007 (%) of them expressing the Aδ-LTMR marker Ntrk2 (TrkB), and none of them expressing the C-LTMR marker tyrosine hydroxylase (TH) (Figs 5B , 5C , S6B , and S6C ). The same strategy was used to trace the presynaptic neurons of Ucn3::Cre INs in the DRG. Ucn3::Cre INs are mainly distributed in laminae II-III of spinal dorsal horn ( S6D Fig ). In contrast, only 27.74 ± 5.007 (%) of EGFP + DRG neurons traced in Ucn3::Cre mice coexpressed Tmem163 ( S6E and S6F Fig ). Collectively, these data strongly indicate that NPY + -derived INs receive monosynaptic inputs from Tmem163 + primary afferents, whereas Ucn3::Cre neurons exhibit minimal monosynaptic connections with Tmem163 + primary afferents.

To investigate the potential colocalization of TMEM163 and VGluT1 on presynaptic vesicles, we employed immunoelectron microscopy. Because specific antibodies against TMEM163 are not available, we utilized an adeno-associated virus (AAV) vector expressing TMEM163 fused with an EGFP reporter at its C-terminus (AAV-TMEM163-EGFP). Subsequently, we administered AAV-TMEM163-EGFP (titer: 1.045E+12 vg/ml, 10 μl) by i.t. injection to specifically infect large DRG neurons, whose central nerve branch fibers predominantly terminated in the deep laminae of the spinal dorsal horn ( Fig 4G ). Then, we labeled EGFP with 10 nm gold particles and VGluT1 with 6 nm gold particles separately. The results demonstrated the presence of TMEM163-EGFP and VGluT1 either in the same presynaptic vesicles or distributed across separate presynaptic vesicles (Figs 4H , S4A , and S4B ). Quantitative analysis showed that 20.24 ± 1.711 (%) 10 nm gold particles colocalized with 6 nm gold particles and 11.33 ± 1.624 (%) 10 nm gold particles colocalized with 10 nm gold particles in the same presynaptic vesicles ( S4C Fig ). These findings are consistent with the observed localization of TMEM163 and VGluT1 in ND7/23 cells and cultured DRG neurons ( Fig 4B–4F ). Collectively, these data strongly indicate that Zn 2+ and Glu may be incorporated into the same vesicles, cotransported to nerve terminals, and subsequently coreleased upon stimulation and depolarization of TMEM163-positive primary afferents.

( A ) ZnS AMG staining in the spinal cords of VGluT1::Cre;Ai14 mice. Scale bar = 100 μm. ( B ) Overexpression of mouse TMEM163-EGFP (green) and mouse VGluT1-mCherry (red) in ND7/23 cells. Scale bar = 10 μm. ( C, D ) TMEM163-EGFP (green) and VGluT1-mCherry (red) were overexpressed in cultured DRG neurons by electroporation. (C) VGluT1 (red) colocalized with TMEM163 (green) in the axons of cultured DRG neurons. (D) Live-cell imaging reveals the cotransport of TMEM163-EGFP (green) and VGluT1-mCherry (red) in the same vesicle. Scale bar = 5 μm. ( E, F ) Quantitative analysis of VGluT1 (red) cotransport with TMEM163 (green) in the axons of cultured DRG neurons. (E) The percentage of TMEM163-positive puncta containing VGluT1 and the percentage of VGluT1-positive puncta containing TMEM163, n = 11. (F) The percentage of VGluT1 and TMEM163 double-positive puncta exhibiting anterograde transport, retrograde transport, and no movement; n = 11. ( G ) Immunofluorescence staining of the spinal cord following intrathecal injection of AAV-TMEM163-EGFP into 6- to 8-week-old mice. Spinal cord tissue was harvested and subjected to immunostaining using primary antibodies against VGluT1 (red) and Nissl staining (purple). Scale bars = 100 μm (left), 20 μm (right). ( H ) Immunogold labeling of EGFP (10 nm) and VGluT1 (6 nm) in spinal cord slices from mice infected with AAV-TMEM163-EGFP. The dashed line circle indicates a vesicle containing both TMEM163-EGFP and VGluT1. The arrowhead indicates a TMEM163-EGFP-positive vesicle, and the arrow indicates a VGluT1-positive vesicle. Scale bars = 200 nm (left) and 50 nm (right). The underlying data for Fig 4E and 4F can be found in S1 Data . AAV, adeno-associated virus; DRG, dorsal root ganglia.

Next, we further investigated how vesicular Zn 2+ is transported from the DRG to the spinal dorsal horn by TMEM163 and released. A previous study showed that Zn 2+ is mostly stored in the same synaptic vesicles as Glu and released together with Glu into the synaptic cleft during neuronal firing, leading to a transient increase in the free Zn 2+ concentration in the synaptic cleft, which further modulates synaptic transmission in the hippocampus [ 8 , 14 , 43 , 44 ]. VGluTs localized on the membrane of synaptic vesicles play a critical role in incorporating Glu into synaptic vesicles. Because VGluT1 is selectively expressed in large DRG neurons [ 45 ], similar to the expression pattern of TMEM163, we examined the colocalization of TMEM163 and VGluT1 at the subcellular level. Data from VGluT1::cre; Ai14 mice showed that the central branches of VGluT1 + DRG neurons terminated in the deep laminae of the spinal dorsal horn, overlapping with the ZnS AMG signal ( Fig 4A ). Then, we induced overexpression of TMEM163-EGFP and VGluT1-mCherry in the ND7/23 cell line, a hybridized cell line consisting of mouse neuroblastoma and rat DRG neurons. TMEM163 colocalized with VGluT1 in vesicle-like punctate structures ( Fig 4B ). Then, we employed electroporation to induce overexpression of TMEM163-EGFP and VGluT1-mCherry in cultured DRG neurons. Remarkably, we observed distinct colocalization of TMEM163-EGFP and VGluT1-mCherry within the same vesicle-like puncta in the cell body, neurites, and nerve terminals ( Fig 4C and 4D ). Quantitative analysis showed that 42.10% of TMEM163-EGFP + vesicle-like puncta were vGluT1-mCherry-positive and 45.62% of vGluT1-mCherry + vesicle-like puncta were TMEM163-EGFP-positive ( Fig 4E ). Live cell imaging was also used to trace the cotransport of TMEM163-EGFP and vGluT1-mCherry. The results showed that 35.51% of double-positive vesicle-like puncta were anterogradely transported toward the nerve terminal, 16.19% of them were retrogradely transported toward the cell body, and 48.30% of them did not show any movement ( Fig 4D–4F ).

( A ) Relative mRNA levels of ZnT1-ZnT10 and TMEM163 in the mouse DRG; n = 3/group. The underlying primer data for Fig 3A can be found in S1 Table . ( B ) Profiling of the cell type expressing TMEM163 in the mouse DRG based on a previously published single-cell RNA-seq database (GSE59739). ( C ) Representative images of ISH combined with IHC using a probe targeting mouse Tmem163 (red) and a large DRG neuron marker (NF3200). Scale bar = 100 μm. ( D ) Profiling of the cell type expressing TMEM163 in the DRG. Immunofluorescence staining using markers for different DRG neuron subtypes (large-sized DRG neurons (NF200), peptidergic DRG neurons (SP and CGRP), nonpeptidergic DRG neurons (IB4), and C-LTMRs (TH)) was performed; n = 17 /group. ( E ) Relative mRNA level of Tmem163 in the DRG of naïve control mice and dry skin model mice; unpaired t test, n = 4 /group. ( F ) Relative mRNA level of Tmem163 in the mouse DRG at 2 months (2 m), 4 months (4 m), 6 months (6 m), 12 months (12 m), 18 months (18 m), and 24 months (24 m). Unpaired t test; n = 3–7/group. ( G ) Schematic diagram of the strategy employed to generate VGlut1::cre;Tmem163 f/f mice. ( H ) Genotyping of Tmem163 cKO mice using the primer listed in the S1 Table . The gel image shows the bands corresponding to WT mice (132 bp), Tmem163 f/f mice (202 bp), and Cre recombinase (100 bp). M indicates molecular weight marker bands for reference. ( I, J ) ZnS AMG staining in the DRG of WT mice and Tmem163 cKO mice. (I) Representative images of ZnS AMG staining in DRG sections. Scale bar = 20 μm. (J) Quantitative analysis of the mean intensity of ZnS AMG staining, Mann–Whitney test, n = 13–14/group. ( K ) Quantification of the [Zn 2+ ] in the CSF of WT mice and Tmem163 cKO mice; n = 3/group. ( L, M ) ZnS AMG staining in the spinal cords of WT mice and Tmem163 cKO mice. (L) Representative images of ZnS AMG staining in spinal cord sections. Scale bar = 50 μm. (M) Quantitative analysis of the mean intensity of ZnS AMG staining, Mann–Whitney test, n = 10/group. ( N ) Touch-evoked itch response elicited by application with different von Frey filaments in Tmem163 f/f mice and Tmem163 cKO mice; two-way ANOVA followed by Sidak’s multiple comparisons test; n = 6 /group. ( O ) Spontaneous itch response of Tmem163 f/f and Tmem163 cKO mice under dry skin conditions. Unpaired t test; n = 6/group. ( P ) Correlation analysis of Tmem163 mRNA level with spontaneous itch number of Tmem163 f/f mice and Tmem163 cKO mice. Pearson correlation analyze; n = 6/group. ( Q ) Touch-evoked itch response of Tmem163 f/f and Tmem163 cKO mice under dry skin conditions. Welch’s t test; n = 6/group. ( R ) Touch-evoked itch response of Tmem163 f/f and Tmem163 cKO mice after intrathecal injection of vehicle or ZnCl 2 . Two-way ANOVA followed by Sidak’s multiple comparisons test; n = 5–6/group. All data are expressed as the mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001. The underlying data for Fig 3A, 3D, 3E, 3F, 3J, 3K, 3M, 3N, 3O, 3P, 3Q, and 3R can be found in S1 Data . cKO, conditional knockout; CSF, cerebrospinal fluid; DRG, dorsal root ganglia; ISH, in situ hybridization; SEM, standard error of the mean; WT, wild-type.

To elucidate the molecular mechanism underlying the accumulation of spinal Zn 2+ in deep laminae of spinal dorsal horn and the selective regulatory effect of spinal Zn 2+ on aging-related itch, we conducted a comprehensive analysis of Zn 2+ transporters (ZnTs) responsible for transporting Zn 2+ from the cytosol into organelles [ 8 ] within the DRG. Our results revealed that the expression level of TMEM163 (ZnT11) in the DRG was markedly higher than that of other ZnTs ( Fig 3A ). Furthermore, we examined the expression profiles of ZnTs in a single-cell sequencing database [ 41 , 42 ], and the results showed that TMEM163, but not ZnT1-10, was selectively highly expressed in large-sized DRG neurons (neurofilament (NF)-containing DRG neurons) (Figs 3B and S3A ), which was consistent with the observed distribution of Zn 2+ in DRG ( Fig 1B and 1D ). Consistently, ISH combined with IHC showed that Tmem163 was mainly expressed in large-sized DRG neurons, with 79.24 ± 2.86% of Tmem163 + DRG neurons expressing NF200 and 87.53 ± 1.56% of Tmem163 + DRG neurons being positive for Slc17a7 (VGluT1) (Figs 3C , 3D , S3B , and S3C ). In mice with dry skin and aged mice, the expression of Tmem163 increased significantly ( Fig 3E and 3F ), indicating the potential role of this transporter in aging-related itch. To test the role of Tmem163 in aging-related Zn 2+ accumulation in the spinal dorsal horn and aging-related itch, we generated Tmem163 conditional knockout (cKO) mice in which Tmem163 was selectively deleted in VGluT1 + DRG neurons ( Fig 3G and 3H ). In the cKO mice, the intensity of ZnS AMG staining in the DRG was significantly reduced ( Fig 3I and 3J ). Furthermore, the concentration of Zn 2+ in the CSF was markedly decreased from 0.927 ± 0.177 ng/L in WT mice to 0.332 ± 0.035 ng/L in cKO mice ( Fig 3K ). We also examined the distribution of Zn 2+ in the deep laminae of the spinal dorsal horn in cKO mice and found a sharp reduction in the intensity of ZnS AMG staining ( Fig 3L and 3M ). These findings collectively indicate the critical role of TMEM163 in the accumulation of Zn 2+ in large-sized DRG neurons and the deep laminae of the spinal dorsal horn. We also assessed the itch behavior of the cKO mice under naïve conditions and in the context of dry skin. Touch-evoked itch was induced by stimulation of the ear region with graded von Frey filaments, and the results showed that itch evoked by the filaments with bending forces of 0.07 g and 0.16 g was significantly decreased in the cKO mice under naïve conditions ( Fig 3N ). Similarly, touch-evoked itch elicited by application of Von Frey filaments to the nape region was also reduced ( S3D Fig ). Under dry skin conditions, both spontaneous itch and touch-evoked itch were decreased dramatically in cKO mice (Figs 3O–3Q and S3E ). Importantly, the reduction in touch-evoked itch can be reversed by i.t. injection of ZnCl 2 under dry skin conditions ( Fig 3R ). Collectively, these findings provide compelling evidence that the Zn 2+ transporter TMEM163 mediates the accumulation of Zn 2+ in large-sized DRG neurons and the deep laminae of the spinal dorsal horn, thereby playing a vital role in the pathogenesis of aging-related itch.

Previous studies have highlighted the essential role of skin Zn 2+ in both acute itch and chronic itch [ 36 ]. We then investigated whether skin Zn 2+ is involved in aging-associated itch. Our results showed that chelating skin Zn 2+ via intradermal injection of ZX1 at the ear back significantly reduced both spontaneous and touch-evoked itch in aged mice ( S2E and S2F Fig ), suggesting that skin Zn 2+ plays a role in age-associated itch. I.t. injection is a well-established method for delivering drugs, siRNAs, or AAVs to DRG neurons and SC, as previously described [ 37 – 40 ]. Our findings demonstrated that I.t. injection of TPEN elicited an obvious tendency of reduction of [Zn 2+ ], although this difference was not statistically significant ( S2G and S2H Fig ), indicating that Zn 2+ in DRG neurons may be released into the skin and contribute to Zn 2+ levels. However, I.t. injection of ZnCl 2 had no effect on the [Zn 2+ ] in the skin ( S2I and S2J Fig ). These data suggest that I.t. injection of TPEN-elicited reduction of age-associated itch is partially due to the chelation of skin Zn 2+ .

(A, B) ZnS AMG staining in the spinal cords after I.T. injection ZX1. (A) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (B) Quantitative analysis of the mean ZnS AMG staining intensity in spinal cords after I.T. injection ZX1; unpaired t test, n = 26-21/group. (C, D) ZnS AMG staining in the spinal cords after I.T. injection TPEN. (C) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (D) Quantitative analysis of the mean ZnS AMG staining intensity in spinal cords after I.T. injection TPEN; unpaired t test, n = 21~19/group. ( E ) Spontaneous itch response of aged mice after intrathecal injection of the Zn 2+ chelators TPEN and ZX1. Two-way ANOVA followed by Sidak’s multiple comparisons test; n = 6 mice/group. ( F, G ) Touch-evoked itch response of aged mice elicited with a 0.07 g von Frey filament after intrathecal injection of vehicle, the Zn 2+ chelator TPEN (100 μm, 10 μl) (F) or ZX1 (100 μm, 10 μl) (G). Two-way ANOVA followed by Sidak’s multiple comparisons test; n = 5–6 mice/group. ( H ) Spontaneous itch response of dry skin model mice after intrathecal injection of vehicle or the Zn 2+ chelator TPEN. Two-way ANOVA followed by Sidak’s multiple comparisons test; n = 6 mice/group. ( I ) Touch-evoked itch response of dry skin model mice after intrathecal injection of vehicle or the Zn 2+ chelator TPEN. Two-way ANOVA followed by Sidak’s multiple comparisons test; n = 5–6 mice/group. ( J–O ) Effect of intrathecal injection of vehicle or the Zn 2+ chelator TPEN on acute chemical itch elicited by intradermal injection of 48/80 (2 μg/μl, 50 μl), histamine (10 μg/μl, 50 μl), CQ (4 μg/μl, 50 μl), 5-HT (6 μg/μl, 50 μl), β-alanine (6 μg/μl, 50 μl), and SLGRL (2 μg/μl, 50 μl). Unpaired t test; n = 5–6 mice/group. ( P, Q ) ZnS AMG staining in the spinal cords after I.T. injection Vehicle and ZnCl 2 . (P) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (Q) Quantitative analysis of the mean ZnS AMG staining intensity in spinal cords after I.T. injection Vehicle and ZnCl 2 ; unpaired t test, n = 19/group. ( R ) Spontaneous itch response of WT mice after intrathecal injection Vehicle and ZnCl 2 . Unpaired t test, n = 5 mice/group. ( S, T ) Itch response evoked by touch stimulation of the nape (S) and back of the ear (T) after intrathecal injection of 2 × NaCl and ZnCl 2 . Two-way ANOVA followed by Sidak’s multiple comparisons test, n = 6–12 mice/group. All data are expressed as the mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001. The underlying data for Fig 2B, 2D, 2E, 2F, 2G, 2H, 2I, 2J, 2K, 2L, 2M, 2N, 2O, 2Q, 2R, 2S, and 2T can be found in S1 Data . SEM, standard error of the mean.

Next, we tested whether spinal Zn 2+ mediates aging-related itch, including spontaneous itch and mechanical itch hypersensitivity. Mechanical itch hypersensitivity, known as alloknesis, represents an abnormal sensory state in which mechanical stimuli, typically non-itch-inducing, induce intense itchiness, a phenomenon commonly observed in aged skin [ 32 , 33 ]. Aged mice developed both spontaneous itch and mechanical itch hypersensitivity ( S2A–S2D Fig ), which was consistent with previous studies [ 3 , 30 ]. Intrathecal (i.t.) injection of the specific Zn 2+ chelator TPEN (100 μm, 10 μl) or ZX1 (100 μm, 10 μl) [ 34 , 35 ] both can significantly reduce the intensity of ZnS AMG staining in laminae III-V of spinal dorsal horn ( Fig 2A–2D ). Importantly, these 2 chelators effectively alleviated both spontaneous itch and touch-evoked itch in aged mice in the first hour after injection ( Fig 2E–2G ). Consistently, both spontaneous itch and touch-evoked itch in the AEW treatment-induced dry skin model mice were effectively alleviated after chelation of spinal Zn 2+ through i.t. injection of TPEN ( Fig 2H and 2I ). However, neither histamine-dependent acute chemical itch (elicited by intradermal injection of compound 48/80 and histamine) nor histamine-independent acute chemical itch (elicited by intradermal injection of chloroquine, serotonin, β-alanine, and SLGRL) were affected by chelation of spinal Zn 2+ with i.t. injection of TPEN ( Fig 2J–2O ). Next, we investigated whether Zn 2+ can directly elicit itch. The results demonstrated that i.t. injection of ZnCl 2 increased the intensity of the Zn 2+ signal in the spinal dorsal horn ( Fig 2P and 2Q ). Behavioral tests showed that i.t. injection of ZnCl 2 was insufficient to elicit spontaneous itch but significantly potentiated touch-evoked itch stimulated by application of a 0.07 g filament to both the nape and back of the ear ( Fig 2R–2T ). Collectively, these data indicate that endogenous Zn 2+ in the spinal cord regulates aging-related itch.

( A ) Schematic diagram illustrating the projection of central branches arising from different DRG neuron subtypes within the spinal dorsal horn and the dorsal root ligation model. ( B , C ) ZnS AMG staining in the DRG in young mice and aged mice. (B) Representative images displaying ZnS AMG staining in DRG sections. Scale bar = 50 μm. (C) Quantitative analysis of the mean gray value of ZnS AMG . Welch’s t test, n = 6/group. ( D ) Size distribution of ZnS AMG -positive DRG neurons and total DRG neurons in naïve young mice. ( E , F ) ZnS AMG staining in the spinal cords of young mice and aged mice. (E) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (F) Quantitative analysis of the mean intensity of ZnS AMG staining; unpaired t test, n = 10/group. ( G ) The [Zn 2+ ] in the CSF of young mice and aged mice; unpaired t test, n = 8 mice/group. ( H ) The [Zn 2+ ] in the CSF of young patients and elderly patients; Welch’s t test, n = 5–6/group. (I–K) ZnS AMG staining in the spinal cords of dry skin mice. (I) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (J, K) Quantitative analysis of the mean intensity of ZnS AMG staining on the contralateral and ipsilateral sides; (J) lateral spinal cord, Mann–Whitney test, n = 11/group; (K) medial spinal cord, unpaired t test, n = 11/group. (L) The [Zn 2+ ] in the CSF of naïve mice and dry skin model mice; unpaired t test, n = 6 mice/group. ( M, N ) ZnS AMG staining in the spinal cords of dorsal root ligation model mice. (M) Representative image of ZnS AMG staining in the spinal cord. Scale bar = 50 μm. (N) Quantitative analysis of the mean ZnS AMG staining intensity in spinal cords of dorsal root ligation model mice; unpaired t test, n = 9/group. All data are expressed as the mean ± SEM. Significant differences were analyzed using unpaired t tests; *p < 0.05, **p < 0.01, and ***p < 0.001. The underlying data for Fig 1C, 1D, 1F, 1G, 1H, 1J, 1K, 1L, and 1N can be found in S1 Data . CSF, cerebrospinal fluid; DRG, dorsal root ganglia.

To investigate the role of endogenous Zn 2+ in itch sensation, we assessed its distribution in the DRG and spinal cord using zinc sulfide autometallography (ZnS AMG ) staining, which is used to specifically detect free Zn 2+ [ 28 , 29 ]. The results showed that there was a large amount of Zn 2+ selectively distributed in large-sized DRG neurons (cross-sectional area ≥600 μm 2 ), the central branches of which mainly terminate in laminae III-V of the spinal dorsal horn ( Fig 1A–1D ). Notably, there was a significant increase in the intensity of ZnS AMG staining specifically within large-sized DRG neurons in aged mice (≥18 months) ( Fig 1B and 1C ). In the spinal cord, Zn 2+ was mostly distributed in the deep dorsal horn from laminae III to laminae V ( Fig 1E ); this area overlapped with the area where the central branches of large-sized DRG neurons terminated. Consistently, the intensity of ZnS AMG staining in both laminae III-V of the spinal dorsal horn and the epidermis of nape region significantly increased in aged mice (Figs 1E and 1F and S1A and S1B ). We measured the concentration of Zn 2+ ([Zn 2+ ]) in the cerebrospinal fluid (CSF), and the results showed that the [Zn 2+ ] increased from 0.808 ± 0.054 ng/L in young mice (6 to 8 weeks) to 1.270 ± 0.080 ng/L in aged mice ( Fig 1G ). Consistently, the [Zn 2+ ] in the CSF of elderly people (≥54 years) was significantly higher than that in the CSF of young people (≤38 years) ( Fig 1H ). Dry skin is a common skin condition in older adults [ 30 , 31 ]. Thus, we evaluated the distribution of vesicular Zn 2+ in the spinal cord and the [Zn 2+ ] in the CSF in a well-established acetone–ether–water (AEW) model that recapitulates dry skin observed in elderly patients [ 30 ]. The intensity of ZnS AMG staining in the lateral part of dorsal horn, receiving sensory information from nape region, significantly increased in ipsilateral side, while that of medial part was not significantly changed ( Fig 1I–1K ). However, the [Zn 2+ ] in the CSF of dry skin model mice was comparable to that of naïve control mice ( Fig 1L ), possibly because the pathological changes in the nape are insufficient to elicit the change of [Zn 2+ ] in CSF. The selective distribution of Zn 2+ in large-sized DRG neurons and laminae III-V of the spinal dorsal horn indicated that Zn 2+ in the deep laminae of the dorsal horn may be transported from large-sized DRG neurons. To test this hypothesis, we performed a spinal dorsal root ligation assay ( Fig 1A ). Interestingly, our results demonstrated a substantial decrease in the intensity of ZnS AMG staining specifically in laminae III-V on the ipsilateral side compared to the contralateral side 3 days after ligation ( Fig 1M and 1N ). These findings strongly suggest that Zn 2+ in the deep laminae of the dorsal horn is indeed transported from DRG neurons and its level increased during aging and in dry skin.

Discussion

Pruritus, the sensation of itching, is a prevalent skin complaint among individuals aged 65 and above [47,48]. In elderly individuals, particularly in autumn and winter, xerosis and dry skin are leading causes of itchiness characterized by heightened mechanical sensitivity [3,48]. In this study, we found that there is one subset of DRG neurons that express the vesicular Zn2+ transporter TMEM163 and are enriched with Zn2+. These neurons establish direct synaptic connections with NPY+ inhibitory INs in laminae III-IV of the dorsal horn. Notably, during aging, the expression of TMEM163 significantly increases, resulting in the accumulation of higher levels of Zn2+ within large-sized DRG neurons and the central terminals of large-sized DRG neurons within the spinal cord. Upon stimulation, excessive Zn2+ is released into synaptic clefts in the spinal dorsal horn, where it modulates the neuronal activity of NPY::Cre INs by inhibiting NMDARs and AMPARs, leading to the disinhibition of mechanical itch-associated neural circuitry and aging-related itch.

Under physiological conditions, mechanical itch is tightly regulated by spinal inhibitory INs, resulting in minimal perception [5,6,30]. TLR5+ Aβ-low-threshold mechanoreceptors (LTMRs)/Ucn3+ spinal neurons and NPY1R+ spinal neurons have been found to selectively transmit mechanical itch signals, which are strictly gated by neuropeptide Y (NPY)::Cre-derived inhibitory INs under physiological conditions. Mice with selective ablation of or silencing of spinal NPY+ INs experience more severe mechanical itch hypersensitivity and develop obvious scratch-induced skin lesions [6]. In contrast, chemogenetic activation of spinal NPY+ INs not only attenuates acute mechanical itch responses to light punctate stimulation of the skin behind the ears but also suppresses pruritogen-induced itch, reduces acute nocifensive reflexes, and alleviates behavioral symptoms of associated with neuropathic and inflammatory pain [3,7]. Gentle touch applied to the skin can simultaneously activate both the neural circuitry involved in transmitting mechanical itch (e.g., TLR5+ LTMRs to Ucn3+/Tac2+ INs) and an inhibitory neural circuitry that effectively suppresses mechanical itch. NPY+-derived INs have been identified as key elements involved in gating mechanical itch in the spinal cord [3,6]. However, the specific LTMRs that form monosynaptic connections with NPY+-derived INs remain unknown. Electrophysiological experiments conducted by Bourane and colleagues reveal that NPY+ INs in the spinal dorsal horn receive monosynaptic inputs from Aβ, Aδ, and C-fiber generating DRG neurons. The proportions of these monosynaptic inputs are 10.8% from Aβ, 40.5% from Aδ, and 48.7% from C-fiber DRG neurons, respectively [6]. However, our retrograde trans-monosynaptic tracing experiments revealed that NPY+ INs primarily receive monosynaptic inputs from DRG neuron expressing TMEM163, while they make few direct synaptic connections with Aδ-LTMRs and C-LTMRs. In contrast, only 27.74 ± 5.007% of the monosynaptic inputs to Ucn3+ interneurons were from TMEM163+ primary afferents. The findings by Albisetti and colleagues demonstrate a significant tropism for DRG neurons when rabies viruses are injected into the dorsal horn, with non-peptidergic nociceptors and C-LTMRs showing resistance to infection [49]. This can explain why our transsynaptic retrograde-traced neurons are mostly TMEM163+. Here, we can only state that NPY+-derived INs receive monosynaptic inputs from Tmem163+ primary afferents, whereas Ucn3::Cre neurons exhibit minimal monosynaptic connections with Tmem163+ primary afferents. Another point that should be clarified is that the cells revealed by the reporter cross are not necessarily the same as those captured by intraspinal injection of virus in adults. In the Science paper by Bourane and colleagues, they crossed the NPY::Cre transgenic mice with R26LSL-tdTomato; (Ai14) reporter mice to label INs that express NPY in the spinal dorsal horn. ISH with a probe against mouse Npy showed that only 35% of tdTomato+ neurons expressed NPY at P30 [6]. Thus, they concluded that NPY::Cre; Ai14 mice capture 2 populations of NPY-expressing neurons: one that transiently expresses NPY during late embryonic/early neonatal development and another that shows persistent expression into adulthood.

Zinc has been implicated in somatosensation, including pain, itch, and acid sensation, in several studies [36,50–52]. Zinc selectively activates TRPA1 via intracellular cysteine and histidine residues in its NH2-terminus [50]. TRPA1 is a nonselective cation channel located on the plasma membrane known to act as a sensor of pain, itch, and environmental irritants and is responsible for various protective responses such as tearing, airway resistance, and coughing [53,54]. Intradermal injection of Zn2+ at concentrations above 0.3 mM induces pronounced scratching behavior in mouse models of neck and cheek itch, and this effect is dependent on TRPA1 activation [36]. Another crucial nonselective cation channel involved in histamine-dependent chemical itch and various forms of chronic itch is TRPV1, which can be inhibited by Zn2+ [55]. However, the expression of TRPA1 and TRPV1 is limited to small peptidergic or nonpeptidergic DRG neurons with central branches terminating in laminae I-II of the spinal dorsal horn. Thus, it is less likely that zinc stored in large-sized DRG neurons and their central terminals regulates aging-related itch through TRPA1 and TRPV1. Additionally, vesicular Zn2+ stored in the central terminals of TMEM163+ primary afferents is released into the synaptic cleft upon neuronal depolarization. ZIP transporters, which are responsible for zinc influx from the extracellular space into the cytosol, are also expressed in large-sized DRG neurons, suggesting that Zn2+ may be recycled from the synaptic cleft back into the cytosol [36,41,51]. Therefore, it is less plausible that Zn2+ in the synaptic clefts in laminae III-V diffuses into the central terminals area of small DRG neurons expressing TRPV1 and TRPA1 in laminae I-II to regulate itch via modulating the channel activity of TRPV1 and TRPA1.

Both AMPARs and NMDARs can be inhibited by free Zn2+ [18,35]. Chelation of spinal Zn2+ with ZX1 or TPEN had no effect on mEPSCs but significantly increased the amplitudes of AMPAR- and NMDAR-mediated evoked EPSCs in NPY+ INs. Consistently, light touch-evoked itch decreased after chelation of spinal Zn2+ under naïve conditions. The data obtained from our study indicate that the spontaneous release of Zn2+ from presynaptic vesicles is inadequate to inhibit AMPARs and NMDARs on NPY+ INs. However, electrical stimuli and light touch are capable of triggering enhanced Zn2+ release into the synaptic cleft. Increased Zn2+ release may inhibit NPY+ INs, leading to a reduction in mechanical itch sensation. In contrast to NPY+ INs-ablated mice, which exhibit spontaneous itch and pronounced touch-evoked itch [3,6], young mice or mice in a naïve state exhibit only mild touch-evoked itch and low levels of spontaneous itch. This observation may be attributed to the relatively low concentration of Zn2+ present in the central terminals of large-sized DRG neurons, suggesting that the release of vesicular Zn2+ during spontaneous or even light touch stimulation is inadequate to fully inhibit the neuronal activity of NPY+ INs. However, in the presence of dry skin or during the aging process, TMEM163 expression in large-sized DRG neurons is notably up-regulated, and this change is accompanied by a substantial increase in the concentration of Zn2+ within the central terminals of these neurons. It is important to note that we did not verify whether the NPY promoter-driven mCherry expression was specific to NPY+ neurons when recorded activity in NPY neurons in both young and aged mice, due to the lack of a reliable NPY antibody in our laboratory. Nathanson and colleagues previously demonstrated that the NPY promoter, when driving GFP expression, is largely restricted to GABAergic neurons. However, other subtypes of inhibitory interneurons, such as PV, VIP, and SST neurons, can also express GFP to some extent [56]. Therefore, we cannot entirely rule out the possibility that our recordings included activity from other inhibitory neuron types in addition to NPY+ neurons. Given these considerations, we concluded that neuronal transmission to inhibitory interneurons is diminished in the spinal cord of aged mice. Furthermore, the application of light touch or engagement in grooming behavior can elicit release of Zn2+ into the synaptic cleft, resulting in the significant inhibition of NPY+-derived INs activity and subsequently leading to mechanical itch hypersensitivity. This may also explain the scratch-itch cycle observed during aging and in dry skin. However, the acute chemical itch induced by pruritogens such as CQ and 48/80 is unaffected by spinal Zn2+ chelation, suggesting that key spinal neurons, including those in the Nppb/GRP/GRPR system, may receive minimal primary sensory input from TMEM163+ afferents. Moreover, NPY+-derived interneurons appear to selectively gate the Ucn3 pathway but not the chemical itch system in the spinal cord. A study demonstrated that the loss of cutaneous touch receptors, specifically Merkel cells, leads to increased alloknesis during aging and in xerosis, indicating that inputs from SAI Aβ-LTMRs may attenuate the gating activity of NPY+ interneurons [6,30]. Furthermore, several studies have reported the involvement of NPY+-derived INs in pain gating [5]. Therefore, it is worth further investigating whether TMEM163+ afferents play a role in the development of mechanical pain hypersensitivity under inflammatory or neuropathic conditions, as the underlying mechanism remains unclear and is clinically important.

In summary, we identified a subpopulation of DRG neurons that express the vesicular Zn2+ transporter TMEM163. These neurons establish direct monosynaptic connections with NPY+-derived inhibitory INs. Notably, TMEM163 expression significantly increases with aging and in xerosis conditions, resulting in enhanced Zn2+ loading into presynaptic vesicles. Upon light touch stimulation, excessive release of Zn2+ into the synaptic cleft disrupts the inhibitory effect exerted by NPY+-derived inhibitory INs on mechanical itch-transmitting neurons, thereby facilitating the generation of both aging-related itch and xerosis-related itch (S10 Fig). The findings of this study provide novel mechanistic insight into the pathogenesis of these types of itch and highlight a promising therapeutic target for their management.

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002888

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.