Hepes

Single‐cell RNA‐seq from human DRGs was performed as previously described (dx.doi.org/10.17504/protocol.io. 4r3I2qpr4I1y/v1; Hou et al. 2024 (link)). Human DRGs were surgically extracted from donors and placed in chilled, aerated artificial cerebrospinal fluid (aCSF). The aCSF solution contained 93 mM N‐methyl‐D‐glucamine (NMDG; Sigma‐Aldrich), HCl (12 N; Fisher), KCl (Sigma‐Aldrich), NaH₂PO₄ (Sigma‐Aldrich), NaHCO₃ (Sigma‐Aldrich), HEPES (Sigma‐Aldrich), D‐(+)‐glucose (Sigma‐Aldrich), L‐ascorbic acid (Sigma‐Aldrich), thiourea (Sigma‐Aldrich), Na+ pyruvate (Sigma‐Aldrich), MgSO₄ (2 M; Fisher), CaCl₂ dihydrate (Sigma‐Aldrich), and N‐acetylcysteine (Sigma‐Aldrich). The DRGs were transferred to a sterile petri dish on ice and trimmed to remove connective tissue and fat, using forceps and Bonn scissors. The dural coats (perineurium and epineurium) were carefully removed, isolating the ganglia bodies. These were further divided into approximately 1 mm thick sections using Bonn scissors. The tissue fragments were placed in 5 mL of a prewarmed enzyme mix containing Stemxyme 1, collagenase/neutral protease dispase, and deoxyribonuclease I (DNase I) in sterile filtered Hank's balanced salt solution (HBSS). The DRG–enzyme mixture was placed in a shaking water bath. It was gently triturated every 25 min using a sterile fire‐polished glass Pasteur pipette until the solution turned cloudy and the tissue chunks passed smoothly through the pipette without resistance. Following enzymatic digestion, the dissociated DRGs were passed through a 100 µm cell strainer to remove debris and achieve a uniform cell suspension. The DRG cells were further isolated by layering the cell suspension on a 10% bovine serum albumin (BSA) solution prepared in sterile HBSS. The BSA gradient was then centrifuged at 300 × g for 10 min, resulting in the isolation of DRG cells. The supernatant was discarded, and the cell pellet was immediately fixed using the 10× Chromium Fixed RNA Profiling kit. The cells were fixed for 17 h at 4°C, followed by incubation with the 10× Fixed RNA Feature Barcode kit for 16 h. The remainder of the library preparation was conducted according to the manufacturer's protocol. The samples were sequenced using a NextSeq2000 at the genome core facility at the University of Texas at Dallas. Sequencing data were processed and mapped to the human genome (GRCh38) using 10× Genomics Cell Ranger v7.
Data analysis for single‐cell RNA‐seq was conducted using the Seurat integration workflow (Stuart et al. 2019 (link)). To ensure data quality, only cells with less than 5% mitochondrial gene expression were included. The analysis workflow involved normalizing the data and selecting the top 2000 most variable features.
Following data scaling, standard clustering techniques using the Seurat pipeline (https://satijalab.org/seurat/get_started.html, Seurat (RRID:SCR_016341). The results were visualized using Uniform Manifold Approximation and Projection (UMAP) with a resolution parameter set to 1. UMAP a dimension reduction technique revealed presences of eleven distinct cluster populations in the human DRG. Next, we determined telocyte subsets from our single‐cell database that met the conditions CD34 > 1 and PDGFRA > 1. Cluster‐specific markers were identified using the Wilcoxon rank‐sum test implemented in Seurat and the telocyte marker gene list was ranked based on log2‐fold change (log2FC).
Using a curated ligand–receptor database (Wangzhou et al. 2021 (link)) and the Sensoryomic web tool (https://sensoryomics.shinyapps.io/Interactome/), we conducted an interactome analysis. The initial analysis aimed to identify the top potential interactions between telocyte ligand gene markers and receptors expressed in human DRG neurons, based on expression data (Bhuiyan et al. 2024 (link); Tavares‐Ferreira et al. 2022 (link)). Subsequently, we explored potential interactions between telocyte receptor genes and ligand expressed in human DRG neurons. Ranking were determined by the log2FC values and a p‐adjusted value of < 0.05 for telocyte‐specific genes. Additionally, ligand and receptor genes were annotated with the protein class of their gene products using the PANTHER database (Thomas et al. 2022 (link)). The analysis focused on telocyte‐receptor interactions with hDRG ligands and vice versa, identifying the top 50 receptor–ligand interactions.
The top 150 genes of each of the reclustered telocyte subclusters were analyzed by association with the Gene Ontology (GO) database (RRID:SCR_002811) and the Kyoto Encyclopedia of Genes and Genomes (KEGG; RRID:SCR_001120) using the online database STRING (http://STRING‐db.org, RRID:SCR_005223). The database was also used for protein–protein network analysis.

Poly(acrylamide)-g-(PMOXA, 1,6-hexanediamine, 3-aminopropyldimethylethoxysilane)
(4425:116.2:161.3 Mr; 0.20:0.40:0.40) PAcrAm-g-PMOXA
(NH₂,Si) and poly(acrylamide)-g-(PEG-N₃, 1,6-hexanediamine, 3-aminopropyldimethylethoxysilane) (3500:116.2:161.3
Mr; 0.15:0.425:0.425) PAcrAm-g-PEG-N3 (NH2,Si) were
synthesized and characterized by SuSoS, Switzerland. Each polymer
can be designed with multiple surface linkers; the linkers are specified
in parentheses. The terminal azide (N₃) group of the PAcrAm-g-PEG-N₃ polymer was modified with the bifunctional
DBCO-PEG₄-biotin (SigmaAldrich) at least 5 times excess
compared to the estimated number of N₃ groups. The PAcrAm-g-PEG-N₃ polymer (1 mg mL^–1) in HEPES (1 mM) (SigmaAldrich) buffer adjusted to pH 7.4 and reacted
overnight at room temperature with shaking in the dark. To remove
the excess DBCO modifier, the polymer was filtered five times through
an Amicon centrifugal filter with a 30 kDa cutoff. The success of
the modification was subsequently confirmed by proxy in an SPR experiment.