A key role of PIEZO2 mechanosensitive ion channel in adipose sensory innervation

Abstract

Compared to the well-established functions of sympathetic innervation, the role of sensory afferents in adipose tissues remains less understood. Recent work revealed the anatomical and physiological significance of adipose sensory innervation; however, its molecular underpinning remains unclear. Here, using organ-targeted single-cell RNA sequencing, we identified the mechanoreceptor PIEZO2 as one of the most prevalent receptors in fat-innervating dorsal root ganglia (DRG) neurons. We found that selective PIEZO2 deletion in fat-innervating neurons phenocopied the molecular alternations in adipose tissue caused by DRG ablation. Conversely, a gain-of-function PIEZO2 mutant shifted the adipose phenotypes in the opposite direction. These results indicate that PIEZO2 plays a major role in the sensory regulation of adipose tissues. This discovery opens new avenues for exploring mechanosensation in organs not traditionally considered mechanically active, such as the adipose tissues, and therefore sheds light on the broader significance of mechanosensation in regulating organ function and homeostasis.

Figure 1.. PIEZO2 is expressed in fat-innervating neurons (see also S1)

A. Schematics of fat-targeted single-cell RNA sequencing. CTB-488 was injected in iWAT, CTB-647 was injected in eWAT, and DRGs from T10-L6 were dissected, dissociated, FACS sorted, and loaded for 10x Genomics scRNA-seq. B. UMAP projection of iWAT-DRGs, eWAT-DRGs, and unlabeled-DRGs. C. Expression of Piezo2 and Mrgprd in DRG samples. D. Quantification of Piezo2, Trpv1, and Trpa1 expression percentage in iWAT-DRGs. E. Bar plot showing the percentage of indicated genes expressed in iWAT-DRGs and color-coded by the enrichment ratio (defined by the ratio between the percentage in iWAT-DRGs and unlabeled-DRGs). F. Schematics of retrograde viral labeling from iWAT. G. Representative optical section images of YFP staining and Piezo2 HCR in whole-mount DRG. H. Quantification of Piezo2 percentage in YFP-positive DRGs, N =5. I. Schematics of anterograde viral labeling in PIEZO2-Cre. AAV-DIO-mScarlet was injected into T13/L1 DRGs of PIEZO2-Cre. J. Representative images of cleared iWAT tissues after anterograde labeling and zoomed-in views of regions around lymph node. K. Zoomed in view of PIEZO2+ sensory nerves in iWAT, showing vessel wrapping and parenchymal innervation morphologies. Scale bar: 30 um in G, K, 500 um in J.

Figure 2.. PIEZO2 deletion mimics DRG ablation-induced gene expression changes in fat (see also S2)

A. Schematics of unilateral fat-DRG PIEZO2 deletion. Retrograde virus expression Cre-YFP or YFP were injected into iWAT of PIEZO2^fl/fl mice. B. Representative optical section view of Piezo2 HCR after PIEZO2^KO. Scale bar: 30 um. C. Quantification of Piezo2 knock-out efficiency in iWAT-DRGs. N = 4. Statistics determined by two-tailed paired t-test. D. iWAT fat mass changes after PIEZO2^KO. N = 12. Statistics determined by ratio paired t-tests. E. Relative expression level of indicated genes (normalized to Cre- side) in iWAT. N = 12. Statistics determined by multiple paired tests controlling false discovery rate (FDR) by the original FDR method (Benjamini and Hochberg). F. Fold change correlation between PIEZO2^KO and DRG-ablation (reproduced from Wang et al.) for DRG-ablation signature genes (defined by p_adj < 0.01 induced by DRG-ablation). N=6 for PIEZO2^KO, N=5 for DRG-ablation. Nonparametric Spearman correlation and two-tailed p-value were calculated.

Figure 3.. PIEZO2 gain-of-function (GOF) partially reverses molecular changes (see also S3)

A. Schematics of unilateral fat-DRG PIEZO2^E2799del GOF. Retrograde virus expression Cre-YFP or YFP were injected into iWAT of PIEZO2^E2799del fl/fl mice. B. Volcano plot showing the transcriptional changes in iWAT after PIEZO2^E2799del GOF. C. Schematics of unilateral fat-DRG PIEZO2^E2799del GOF and high-fat diet (HFD) treatment. D. Relative expression level of indicated genes (normalized to Chow condition Cre- side) in iWAT under Chow and HFD conditions. N = 9 for chow and N = 8 for HFD. Statistics determined by multiple paired tests controlling false discovery rate (FDR) by the original FDR method (Benjamini and Hochberg). E. Fold change correlation GOF-HFD and LOF-HFD for DRG-ablation signature genes. N=6 for each group. Nonparametric Spearman correlation and two-tailed p-value were calculated. F. Heatmap showing the normalized expression level in indicated conditions (DRG-ablation, LOF-HFD, GOF-HFD) for top changed genes induced by DRG-ablation. G. Principal component analysis of all gene fold changes in all indicated conditions. H. Heatmap showing gene fold change for individual genes involved in indicated pathways.