Background/Purpose

Saliva is a biofluid secreted by the salivary glands (SGs) that is critical for the health of the oral cavity. In Sjögren’s Syndrome (SS), changes in salivary biomarkers are not only useful for diagnosis, but may also elucidate the mechanisms underlying SG dysfunction. The RNA content of saliva has been shown to be useful for monitoring the health of oral tissue and the oral microbiome. Next-Generation Sequencing (NGS) offers a high throughput method for comparing the salivary transcriptomes of SS patients and healthy volunteers (HV). We have previously shown that RNA is protected within exosomes and we focused this study in using salivary exosomes isolated from the parotid. Parotid saliva has the advantage of being pure without the contamination generated from all type of cells found in the whole saliva.

Methods

Total RNA was extracted from exosomes isolated from parotid saliva from 4 healthy volunteers and 4 primary SS patients. The amount and the quality of the RNA were assessed using Nanodrop, Qubit and Bioanalyzer. The Ion Torrent Proton sequencer  was used sequencing according to manufacturer protocols. Reconstructed reads were aligned using the TMAP (Torrent Mapper) algorithm sequentially to miRBase 19, hg19, viral, and bacterial genomes, i.e. reads left unmapped were used as input for each subsequent step.  The bacterial reference included 14,549 contigs representing 1,132 genomes retrieved from the Human Oral Microbiome Database, and the viral reference included 1,741 viruses retrieved using the NCBI assembled genomes FTP website.  Read counts were generated using the HTSeq python module, and differential expression analysis was done in R using the DESeq2 package. Ingenuity Pathway Analysis (IPA) was used to analyze pathway enrichment and visualization. Microbial expression and taxonomy were analyzed used Pathosystems Resource Integration Center (PATRIC).

Results

Overall the percentage of all reads mapping to each reference were similar between SS and HV: miRBase (5.96% in HV, 5.427% in pSS), hg19 (81.82% HV, 81.96% pSS), viral (6.94% HV, 3.26% pSS), and bacterial (1.121% HV, 1.737% pSS). On average, 4.14% of the input was left unmapped in HVs and 7.62% in pSS patients after all mapping. The differences between percent input reads in patients and HVs was only significant for the bacterial reference, and for unmapped reads. Expression pairing between miRNAs and their target genes using IPA showed significant enrichment for canonical pathways for pancreatic adenocarcinoma signaling, and estrogen-mediated S-phase Entry.  Top differentially regulated bacterial genera (fold-change greater than or less two and FDR-corrected p-value less than 0.01) include Streptococcus, Selenomonas, and Actinomyces.  Notable upregulated viruses include Tomato yellow leaf curl virus and streptococcus phage Cp-1.  Human papillovirus, type 41 was found to be significantly downregulated in pSS patient saliva.

Conclusion

This is the first known effort to compare the transcriptome of SS patients’ salivary exosomes versus healthy controls. Using RNA-Seq, we were able to identify important human genes as well as alterations in the salivary microbiome that could shed light on the mechanisms of salivary dysfunction in pSS.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-descriptive-and-comparative-study-of-the-transcriptome-from-salivary-exosomes-of-sjogrens-syndrome-patients-using-rna-seq/