The miR-183/96/182 cluster regulates sensory innervation, resident myeloid cells and functions of the cornea through cell type-specific target genes

Abstract

The conserved miR-183/96/182 cluster (miR-183C) is expressed in both corneal resident myeloid cells (CRMCs) and sensory nerves (CSN) and modulates corneal immune/inflammatory responses. To uncover cell type-specific roles of miR-183C in CRMC and CSN and their contributions to corneal physiology, myeloid-specific miR-183C conditional knockout (MS-CKO), and sensory nerve-specific CKO (SNS-CKO) mice were produced and characterized in comparison to the conventional miR-183C KO. Immunofluorescence and confocal microscopy of flatmount corneas, corneal sensitivity, and tear volume assays were performed in young adult naïve mice; 3' RNA sequencing (Seq) and proteomics in the trigeminal ganglion (TG), cornea and CRMCs. Our results showed that, similar to conventional KO mice, the numbers of CRMCs were increased in both MS-CKO and SNS-CKO vs age- and sex-matched WT control littermates, suggesting intrinsic and extrinsic regulations of miR-183C on CRMCs. The number of CRMCs was increased in male vs female MS-CKO mice, suggesting sex-dependent regulation of miR-183C on CRMCs. In the miR-183C KO and SNS-CKO, but not the MS-CKO mice, CSN density was decreased in the epithelial layer of the cornea, but not the stromal layer. Functionally, corneal sensitivity and basal tear volume were reduced in the KO and SNS-CKO, but not the MS-CKO mice. Tear volume in males is consistently higher than female WT mice. Bioinformatic analyses of the transcriptomes revealed a series of cell-type specific target genes of miR-183C in TG sensory neurons and CRMCs. Our data elucidate that miR-183C imposes intrinsic and extrinsic regulation on the establishment and function of CSN and CRMCs by cell-specific target genes. miR-183C modulates corneal sensitivity and tear production through its regulation of corneal sensory innervation.

Figure 1

Inactivation of miR-183C has no significant impact on the gross histological architecture of the cornea. (A) H&E staining of cross-sections of the corneas of MS-CKO, SNS-CKO and miR-183C conventional KO and age- and sex-matched WT control littermates. (B) Measurement of the thickness of the corneas. Epi epithelium, Endo endothelium.

Figure 2

Inactivation of miR-183C in either sensory neurons or myeloid cells results in increased number of corneal resident myeloid cells (CRMCs). Compressed confocal images of flatmount cornea of young adult MS-CKO (A) and SNS-CKO sand their age- and sex-matched WT control littermates (B). **p < 0.01.

Figure 3

Inactivation of miR-183C in sensory neurons, but not in the CRMCs, results in decreased sensory nerve density. Compressed confocal images of flatmount corneas of young adult MS-CKO (A) and SNS-CKO sand their age- and sex-matched WT control littermates (B). *p < 0.05; **p < 0.01. MFI mean fluorescence intensity.

Figure 4

Reduction of corneal sensory nerve density in the SNS-CKO is caused by decreased sensory nerve density in the epithelial layer but not the stromal layer. Compressed confocal images of corneal sensory nerves of the epithelial layers (Epi) and stromal layers (Str) of the whorl center areas (A) and peripheral regions (B) of flatmount corneas of young adult, male SNS-CKO sand their age- and sex-matched WT control littermates. *p < 0.05.

Figure 5

Inactivation of miR-183C in sensory nerves, but not in the myeloid cells, results in decreased corneal sensitivity to mechanical stimuli. Sensitivity test using a Cochet and Bonnet aesthesiometer in young adult (8–12 weeks old) miR-183C conventional KO (A), SNS-CKO (B) and MS-CKO mice and their age- and sex-matched WT control littermates (C). **p < 0.01; ***p < 0.001.

Figure 6

Inactivation of miR-183C in sensory nerves, but not in myeloid-cells, results in decreased basal tear volume. Phenol-red thread assays in young adult (8–12 weeks old) miR-183C conventional KO (A), SNS-CKO (B) and MS-CKO mice and their age- and sex-matched WT control littermates (C). *p < 0.05; **p < 0.01.

Figure 7

miR-183C target genes in TG sensory neurons are enriched in axon guidance (A) chemokine signaling pathways (B). Modified from www.genome.jp/kegg/pathway.html. Star-labeled molecules or complexes are or contain target genes of miR-183C.

Figure 8

miR-183C target genes in CRMCs are enriched in FcγR-mediated phagocytosis (A) and chemokine signaling pathways (B). Modified from www.genome.jp/kegg/pathway.html. Star-labeled molecules or complexes are or contain target genes of miR-183C.

Figure 9

Validation of miR-183C target genes by qRT-PCR, ELISA assay and proteomics. (A) Sequence alignment of miR-183C with its target site in in 3′ untranslated region (UTR) of human (upper) and mouse Cx3cl1 transcript (lower panel). (B) qRT-PCR; (C) ELISA assay on Cx3cl1 in the TG and cornea of miR-183C KO and WT control mice. (D,E) Proteomics study in the TG of KO and WT control mice. (D) Percentage of miR-183C target genes in upregulated proteins of the TG of KO vs WT control mice; (E) Venn diagram of miR-183C target genes detected by proteomics (Red) and 3′ RNA seq (Green). 12 target genes (yellow) were detected by both methods; 6 of which were further validated by qRT-PCR (F).