Several studies indicate that there is a potential role for circulating miRNA levels as valuable biomarkers for certain physiological events or disease. It has been shown in different studies that a large portion of the human population has been exposed to JCPyV as they carry antibodies against the JCPyV VP1 capsid protein [30, 31]. This might raise the question why certain individuals are seropositive while others are not and whether there is an intrinsic driving factor that determines susceptibility for the virus. Within the subpopulation of exposed subjects a smaller subset excretes large amounts of JCPyV in the urine [31, 32]. It is unclear why some subjects excrete the virus while others don’t. We therefore wanted to investigate whether miRNA expression levels in plasma are correlated to JCPyV serology or JCPyV viruria. In a first experiment 755 human miRNAs were analyzed in a small set of samples from seronegative subjects, shedders and non-shedder. Based on these results, 10 miRNAs were selected for further analysis and in a subsequent experiment another 100 samples were investigated for these 10 miRNAs. None of them were found to have differential expression between seropositive and seronegative subjects or between shedders and non-shedders. The fact that in the initial profiling experiment, none of the p-values remained significant (p < 0.05) after correction for multiple testing already indicated that the statistical significance of the differences observed was possibly caused by chance .
The lack of different miRNA patterns between seropositive and seronegative subjects is at first sight not that surprising as the presence or lack of antibodies against JCPyV might simply be explained by the fact that some subjects have and some other subjects have not been exposed to the virus. The lack of differential miRNA patterns between both groups might indeed be indicating that a subject’s JCPyV serostatus is driven by viral exposure and is only limitedly influenced by the background of the subject itself. One should however be mindful that JCPyV VP1 serology might actually underestimate real infection rates . As a consequence this study would then effectively be comparing two groups of infected individuals, which may account for the lack of a difference. It should also be mentioned that to date 12 human polyomaviruses have been identified, all with high seroprevalence, making it likely that an individual is infected with several human polyomaviruses [34–37]. The differential plasma miRNA profile that may exist between JCPyV seropositive and seronegative individuals may therefore be masked by infection with other human polyomaviruses. Further studies might therefore require full human polyomavirus serotyping of the subjects examined.
The lack of different miRNA patterns between shedders and non-shedders on the other hand was more surprising as both groups have been exposed to the virus (both are seropositive) but only shedders appear to have active replication of the virus. One might have expected that some difference in biological or genetic background is needed to explain this observation. However, although limited to the analysis of circulating miRNAs only, our data presented in this study show this is not the case.
There is one observation that might be of interest for further study, even though it is not JCPyV related. Based on the analysis of hsa-miR-642, the population of 100 healthy subjects clearly can be separated in 2 groups, one with relatively high levels of hsa-miR-642 and one with low or undetectable levels of hsa-miR-642. Also, more female subjects appear to fall in the low expression group compared to male subjects. As it has been shown that hsa-miR-642 is an adipocyte-specific miRNA, it will be of interest to see whether this hsa-miR-642 based clustering is linked to specific metabolic parameters, such as Body Mass Index or blood cholesterol and fatty acids levels, or insulin resistance .
It should also be noted that only 755 miRNAs were analyzed in this study and it cannot be excluded that differential expression might be present for other miRNAs that were not analyzed here. Also, differences in circulating miRNAs that are expressed at low levels may have been missed. Other techniques that are not limited in the number of miRNAs, such as small RNA sequencing using deep sequencing platforms, have been shown to be very useful to investigate this [39, 40]. Besides the use of a different analytical platform, one might also consider using PAXgene blood instead of plasma. This might show a totally different picture as it does not only include circulating miRNAs but also cellular miRNAs present in blood cells [41, 42].
Although this study did not result in the identification of circulating miRNAs related to JCPyV infection, circulating miRNAs might still hold potential as biomarkers for PML progression in immunocompromised patients. It will therefore be of interest to investigate the miRNA expression profile of samples obtained from PML patients.