Abstract
Exosomes are membrane-bound nanovesicles released by cells into their extracellular environment via fusion of the multivesicular endosome with the plasma membrane. These cell-derived vesicles are present in numerous body fluids, including blood, saliva, urine and lymph, under healthy and pathological conditions. They play a crucial role in cell-to-cell communication by virtue of the exchange of their contents which include proteins, lipids and nucleic acids. The contents can be transferred from one cell to another, including transfer from normal to cancer cells and between cancer cells that can lead to transcriptional and translational changes in the recipient cell. The contents of exosomes are influenced by various physiological and pathological stimuli received by the parent cell. Exosomal contents may thus provide a valuable source for screening and early detection of cancer, as well as monitoring cancer progression and interventions.
The management of patients with oral cancer is a tremendous worldwide liability. Despite significant efforts, the worldwide five-year relative survival rate from oral cancer is generally less than 50%. This low survival rate has remained unchanged for more than three decades. Early diagnosis of oral cancer is associated with improved survival; however, the invasive nature, technical requirements and skill required to perform intra-oral biopsies limits their usefulness as a part of community oral cancer screening. The use of saliva-based diagnostic testing to assess the presence of established salivary biomarkers would be advantageous if it could be shown to be a reliable method to detect early-stage tumours and identify high-risk patients. Saliva is a thick fluid that is 98% water, and the remaining 2% contains proteins, potentially including salivary biomarkers that have diagnostic potential. There have been various advances in saliva analysis, but to date, no saliva-based oral cancer diagnostic tests have made it to the market. The discovery of exosomes in saliva has provided a pathway to detect oncogenic biomarkers protected within salivary exosomes. If we can show that salivary exosomes are a reliable and effective method for detecting oral cancer biomarkers, potentially saliva can be used to develop a non-invasive oral cancer diagnostic test to run community screening programs.
The aim of this study was to isolate exosomes and assess the expression of genes of interest (GOI) FOXM1, HOXA7, CCNB1, DNMT1 and HSPA1A genes and FOXM1, DNMT1 and HSP70 proteins of interest (POI) in exosomes, both in vitro, using oral squamous cell carcinoma (OSCC) and normal primary gingival keratinocyte (PKG) cell lines and ex vivo, in exosomes derived from plasma and saliva samples from OSCC patients and healthy controls. The oncogenes and oncoproteins selected for this study had been documented previously to be overexpressed in oral cancers. The hypothesis tested was that FOXM1, HOXA7, CCNB1, DNMT1 and HSPA1 genes and FOXM1, DNMT1 and HSP70 proteins would be highly expressed in exosomes derived from OSCC patients in comparison with healthy controls.
A major hurdle in studying exosomes is their isolation. Their small size and the fact that they do not have any specific surface protein markers make them difficult to isolate. Thus, this research initially looked into determining and optimising a method to isolate exosomes, which was both cost-effective and yielded enough exosomes for downstream analysis. An Exoquick-TC plus exosome isolation kit (kit) was compared with the classic ultracentrifugation (UC) method for exosome isolation from the concentrated cell-conditioned medium of a representative OSCC cell line (SCC-4). Cryo-Electron Microscopy (Cryo-EM) showed vesicles were delimited by a bilayered membrane and had either a round or slightly elongated shape with diameters ranging between 50 and 150nm in samples obtained using both methods. Dynamic light scattering resulted in multiple peaks for kit exosomes, whereas a single peak was observed for UC exosomes. The multiple peaks most likely came from proteins co-precipitating with exosomes. A significantly higher total RNA was observed in UC exosomes in contrast to kit exosomes (P < 0.0001). This was reflected in subsequent mRNA analysis using quantitative polymerase chain reaction (qPCR), where UC exosomes had lower cycle threshold (Ct) values compared to kit exosomes. UC proved a better option and was used for exosome isolation for all subsequent experiments.
The presence of the GOI and POI were evaluated using the concentrated cell-conditioned medium of normal oral keratinocyte (PGK) and SCC-4 and SCC-25 cell lines. The results showed a significantly higher expression level of all the GOI in exosomes derived from SCC-4 and SCC-25 exosomes than PGK exosomes. The POI results suggested a higher expression of all POI in SCC-4, and DNMT1 and HSP70 in SCC-25 exosomes when compared with cells. Even though their expression was lower in SCC-4 and SCC-25 cells compared to PGK cells, DNMT1 and HSP70 were highly expressed in SCC-4 and SCC-25 exosomes compared to PGK exosomes implying that the cells preferentially package certain proteins for transport via exosomes to recipient cells.
The selected GOI and POI were then further evaluated in exosomes derived from blood plasma and saliva of OSCC patients compared with healthy controls. Due to the low quantity of total RNA, only HSPA1A and housekeeping genes were detected in plasma and saliva exosome samples despite adding a preamplification step. The mean expression level of HSPA1A was higher in both OSCC plasma and saliva exosomes compared to controls. Enzyme linked immunosorbent assay (ELISA) results showed a higher expression of FOXM1, DNMT1 and CCNB1 in OSCC plasma exosomes than controls. There was a marginally lower expression of HSP70 in OSCC plasma exosomes. A similar trend was observed in OSCC saliva exosomes for FOXM1, DNMT1 and HSP70, whereas a lower expression level of CCNB1 was observed in OSCC saliva exosomes compared to controls. Even though a majority of the GOI and POI were overexpressed in OSCC samples, due to the small sample size and a high variation in expression levels between samples, the differences of the means were not statistically significant except for FOXM1 protein, which was significantly higher in OSCC plasma.
This is the first study, to the best of my knowledge, which has shown a similar trend of expression of biomarkers in exosomes in both saliva and plasma. It was also observed that despite having less total RNA and total proteins, a higher expression of the GOI and POI was observed in salivary exosomes compared to plasma exosomes. It is thus proposed that saliva may be a suitable alternative to blood in the future for the detection of disease biomarkers. The results also provide the first evidence that the selected proteins were detected and quantified in exosomes using ELISA. This suggests a change of focus from mRNAs to proteins since they are more stable and easier to detect and quantify. While more precise exosome isolation methods are still underway, it was shown that saliva exosomes are very important for future research and potential clinical application as they carry genes and proteins in greater quantities when compared with plasma exosomes.