Functional and transcriptional adaptations of blood monocytes recruited to the cystic fibrosis microenvironment in vitro.
Description
Cystic fibrosis (CF) lung disease is dominated by the recruitment of myeloid cells (neutrophils and monocytes) from blood, which fail to clear the lung of colonizing microbes. In prior in vitro studies, we showed that blood neutrophils migrated through well-differentiated lung epithelium into CF airway fluid supernatant (ASN) mimic the dysfunction of CF airway neutrophils in vivo, including decreased bactericidal activity despite an increased metabolism. Here, we hypothesized that, similar to neutrophils, blood monocytes would also undergo significant adaptations upon recruitment to CFASN. To test this hypothesis, primary human blood monocytes were transmigrated in our in vitro model into ASN from healthy control (HC) or CF subjects to mimic in vivo recruitment to normal or CF airways, respectively. Surface phenotype, metabolic and bacterial killing activities, and transcriptomic profile by RNA sequencing (RNASeq), were quantified post-transmigration. Unlike neutrophils, monocytes were not metabolically activated nor showed broad differences in activation and scavenger receptor expression upon recruitment to CFASN compared to HCASN. However, monocytes recruited to CFASN showed decreased bactericidal activity. RNASeq analysis showed large effects of transmigration on monocyte RNA profile with differences between CFASN and HCASN conditions, including in immune and antiviral signaling. While monocytes undergo qualitatively different adaptations from those seen in neutrophils upon recruitment to the CF airway microenvironment, their bactericidal activity is also dysregulated, which could explain why they also fail to protect CF airways from infection.
Files
Steps to reproduce
RNA extraction and transcriptomic analysis: RNA was isolated utilizing the NucleoSpin RNA isolation kit (Takara Biosciences) and stored at -80°C until use. RNA quality was assessed using the Bioanalyzer (Agilent technologies), while RNA libraries were prepared following the Illumina True seq manufacturer’s protocol. Samples were subsequently run on Nextseq 550 sequencing system at 25 million single-ended reads per sample. Produced Fastq files from single end reads were aligned to the human reference genome (GRCh38.p13- Ensembl) using the alignment tool HISAT2 (version 2.1.0), using the default settings. Then, BAM files were sorted using SAMtools. Finally, to generate read counts expressed per gene, the tool FeatureCounts (1.5.2) was used. All processed counts were analyzed using DEseq2 to obtain differentially expressed genes (DEGs) between pre and post-transmigration. DEGs were defined as genes with fold changes > 2 folds, False discovery rate < 0.1, and p-value < 0.05. To understand the functions of the DEGs the Metacore server was used against pathway maps and pathway networks. To identify unique dysregulations on gene expression between treatments, all differentially expressed genes were intersected between experimental conditions utilizing the UGent bioinformatic webtool for Venn diagrams (http://bioinformatics.psb.ugent.be/webtools/Venn/). Subsequent enrichment of distinctive gene terms was conducted in the MSigDB v7.2 database (https://www.gsea-msigdb.org/gsea/index.jsp) for molecular signature pathway outputs.