Isolation and Comprehensive Analysis of Cochlear Tissue-Derived Small Extracellular Vesicles

Abstract

Small extracellular vesicles (sEVs) act as a critical mediator in intercellular communication. Compared to sEVs derived from in vitro sources, tissue-derived sEVs can reflect the in vivo signals released from specific tissues more accurately. Currently, studies on the role of sEVs in the cochlea have relied on studying sEVs from in vitro sources. This study evaluates three cochlear tissue digestion and cochlear tissue-derived sEV (CDsEV) isolation methods, and first proposes that the optimal approach for isolating CDsEVs using collagenase D and DNase І combined with sucrose density gradient centrifugation. Furthermore, it comprehensively investigates CDsEV contents and cell origins. Small RNA sequencing and proteomics are performed to analyze the miRNAs and proteins of CDsEVs. The miRNAs and proteins of CDsEVs are crucial for maintaining normal auditory function. Among them, FGFR1 in CDsEVs may mediate the survival of cochlear hair cells via sEVs. Finally, the joint analysis of single CDsEV sequencing and single-cell RNA sequencing data is utilized to trace cellular origins of CDsEVs. The results show that different types of cochlear cells secrete different amounts of CDsEVs, with Kölliker's organ cells and supporting cells secrete the most. The findings are expected to enhance the understanding of CDsEVs in the cochlea.

Keywords: FGFR1; cochlea; hair cells; miRNA; small extracellular vesicle; supporting cells.

Figure 1

Flow chart of CDsEV isolation. ① The cochleae were dissected from the neonatal mice, separated into the basilar membrance (BM), modiolus (M), and spiral ligament (SL), and cut into small pieces. ② The pieces were digested with collagenase D combined with DNase I, with collagenase III alone, or with papain alone. Subsequently, an appropriate tissue digestion enzyme was selected. ③ The digested samples were pretreated to remove cell debris and large vesicles. ④ CDsEVs were isolated by UC, SDGU, or SEC.

Figure 2

Comparison of the digestive efficiency of different enzymes on cochlear tissue for CDsEV isolation. A–E) The expression of the sEV marker proteins CD63, Flotillin‐1, and Tsg101 and the negative markers Calnexin and Rab7 were detected in CDsEVs by WB (A) after digestion with 0.5 U collagenase D combined with 40 U DNase Ι (D&I), 75 U collagenase III (III), or 20 U papain (P). Cochlear tissue total lysate (TL) was used as the positive control. The protein quantification (B), the particle numbers (C), the ratio of particle number to protein concentration (D), and the concentration and size distribution (E) of CDsEVs were detected and calculated. F–O) The protein quantification (F, K), the particle numbers (G, L), the ratio of particle number to protein concentration (H, M), the concentration and size distribution (I, N), and the EV marker protein analysis (J, O) of CDsEVs were detected and calculated after the digestion of cochlear tissue using different concentrations of collagenase D combined with 40 U DNase I (F‐J) or different concentrations of collagenase III (K‐O). P–T) The protein quantification (P), the particle numbers (Q), the ratio of particle number to protein concentration (R), and the transmission electron microscopy (TEM) morphology (S, T) of CDsEVs were compared between two enzymes. Scale bar, 200 nm. *p < 0.05. n = 3.

Figure 3

Comparison of the efficiency of different CDsEV isolation methods. A–D) The WB analysis of the sEV‐positive markers CD63, Tsg101, and CD9 and the negative markers Calnexin and Rab7 (A), the protein quantification (B), the particle numbers (C), and the concentration and size distribution (D) of F1–F4 obtained by SDGU. E,F) The concentration and size distribution (E) and TEM visualization (F) of the F2 and F3 mix obtained by SDGU. G–J) The WB analysis of sEV‐positive and negative markers proteins (G), the protein quantification (H), the particle numbers (I), and the concentration and size distribution (J) of F1–F6 obtained by SEC. K,L) The concentration and size distribution (K) and TEM visualization (L) of the F1, F2, and F3 mix obtained by SEC. M–O) The comparison of protein quantification (M), the particle numbers (N), and the ratio of particle number to protein concentration (O) of CDsEVs for the UC, SDGU, and SEC groups. Scale bar, 200 nm. *p < 0.05, **p < 0.01, n = 3.

Figure 4

Small RNA sequencing analysis of CDsEVs. A,B) Cluster analysis (A) and small RNA type and percentage analysis (B) of CDsEV and cochlear tissue small RNA‐seq data. C) The Venn diagram of CDsEVs and cochlear tissue miRNA data. D) Analysis of the proportion of CDsEV miRNAs within the Vesiclepedia and Exocarta databases. E) The volcano map of differentially expressed miRNAs. FC >2, Q value < 0.05. F) Verification of differentially expressed miRNA by Q‐PCR. G–I) GO (G, H) and KEGG (I) analysis of target genes of enriched miRNAs in CDsEVs. The top cellular components (G) and biological processes (H) are shown. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, *****p < 0.00001, n = 3.

Figure 5

Proteomics analysis of CDsEVs. A,B) Cluster analysis (A) and Venn diagram (B) of the proteomics data for CDsEVs and cochlear tissue. C) Analysis of the proportion of proteins in CDsEVs within the Vesiclepedia and Exocarta databases. D) The volcano map of differentially expressed proteins. E) Verification of CDsEV‐enriched proteins by WB. F–H) GO (F, G) and KEGG (H) analysis of enriched proteins in CDsEVs. The top cellular components (F) and biological processes (G) are shown.

Figure 6

Expression and roles of CDsEV FGFR1. A–C) The expression of FGFR1 in the BM, M, and SL as measured by WB (A) and immunofluorescence staining in the OC (B) and SGN (C) in cochleae from P3 mice. Myosin7a (green) and SOX2 (blue) were used as HC and SC markers, respectively, in (B), and Tuj (green) was used as the SGN marker (C). Scale bar, 20 µm. D,E) Immunofluorescence staining (D) of BM explants treated with different concentrations (0, 15, 30, 50 µm) of the FGFR1 inhibitor PD166866 in vitro. Myosin7a (red) was used as the HC marker. Scale bar, 50 µm. Quantification of HC numbers per 100 µm of the apical, middle, and basal turns of BM explants (E). F) HEI‐OC1 cell viability with different concentrations of FGFR1 inhibitor PD166866 treatment. G) WB analysis of FGFR1‐overexpressing sEVs (FGFR1‐sEVs) obtained from the culture medium of 293T cells transfected with FGFR1‐EGFP‐FLAG plasmids. H) Immunofluorescence staining of EGFP and FLAG in HEI‐OC1 cells incubated with FGFR1‐sEVs. Scale bar, 20 µm. I) Cell viability of HEI‐OC1 cells treated with FGFR1 inhibitor (FI) and FGFR1‐sEVs. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, n = 3.

Figure 7

Single CDsEV transcriptomes and combined analysis with cochlear scRNA‐seq. A) Barcode rank plots showing the distribution of UMI counts associated with each barcode in the filtered matrix. B) The number of qualified CDsEVs, total detected genes, median number of genes, and median UMI counts in each CDsEV. C) UMAP visualization of the cluster profile for the single CDsEV‐seq data. D) The heat map showing the five genes that were highly expressed in each CDsEV cluster. E,F) UMAP visualization of the CDsEVs and cochlear cells (E) and cluster profile (F) for the integrated dataset of single CDsEV‐seq and cochlear scRNA‐seq. G,H) The bar chart shows the proportions (G) and count numbers (H) of the CDsEVs and cochlear cells in each cluster of the integrated dataset of single CDsEV‐seq and cochlear scRNA‐seq. Kölliker's organ cells, KO, Spiral ganglion neuron cells, SGN, Basal cells, BC, Supporting cells, SC, Hair cells, HC, Tympanic border cells, TBC.