Third we also found negative correlations between TREM mRNA
Third, we also found negative correlations between TREM2 mRNA expression and the methylation rate of 4 CpG sites in the intron 1 of TREM2. Gene methylation rates are generally associated with gene expression. Thus, heavily methylated areas of genes are usually less active at the transcriptional level (gene expression turned off), whereas areas with less methylation are more active (gene expression turned on) (Labbé et al., 2016). Although very recent studies showed hypermethylation at the Transcript Start Site (TSS)-associated region of the TREM2 gene despite higher TREM2 mRNA expression in AD brains (Smith et al., 2016, Celarain et al., 2016), there have been some reports that intron 1 methylation is related to gene expression (Xue et al., 2014, Godler et al., 2011, Balmer and LaSalle, 2001). Epigenetic changes can be impacted by age (Fraga et al., 2005), sex (El-Maarri et al., 2007), and factors such as obesity (Dick et al., 2014) and smoking (Lee and Pausova, 2013, Wan et al., 2015). On the other hand, none of the TREM2 DNA methylations were associated with sex, age, age of onset, disease duration, any of the neuropsychological tests, smoking history, APOE –ε4 allele carrier, medication state and psychiatric symptoms.
Recent genome wide methylation studies also revealed that DNA methylation in peripheral leukocytes related to both aging and AD and could be exploited for identification of AD biomarkers (Li et al., 2016). Circulating leukocytes play an important role in the innate immune response against pathogens and numerous studies have shown that peripheral myeloid cells can infiltrate Silmitasertib tissue and reduce the deposition of Aβ plaques (review in Zenaro et al., 2016). Furthermore, recent data indicate that infiltrating monocytes rather than resident microglia express TREM2, a receptor involved in myeloid cell phagocytosis, supporting the role of peripheral myeloid cells in AD pathogenesis (Jay et al., 2015). Thus, hypomethylation at CpG sites in intron 1 of TREM2 in leukocytes can be a biomarker of AD pathogenesis. Considering subtle changes (∼20%) in TREM2 expression or methylation with many control subjects overlapping with the cases, it is difficult to claim that TREM2 expression/methylation change by itself may be a “biomarker for AD”. However, it could be one of the elements of a panel of biomarkers for AD (Watanabe et al., 2015).
There were several limitations of this study. Sample size was relatively small. We did not examine whether methylation rates of TREM2 in leukocytes are equivalent to those in the brain. Although several studies have shown that DNA methylation in peripheral blood often correlates with that in brain (Wockner et al., 2014, Davies et al., 2012), it should be verified whether the same results can be obtained in brain samples. Although there were no significant correlations between the TREM2 mRNA expression level or methylation rates and clinical characteristics of AD, multiple linear regression revealed that age and sex might influence both TREM2 mRNA expression level and methylation rates. Further research on the relationship among TREM2 and age and sex is needed.
In conclusion, we reported for the first time that TREM2 mRNA expression was elevated and that TREM2 DNA methylation rates were decreased in peripheral leukocytes in AD subjects. These results suggest that hypomethylation at CpG sites in intron 1 of TREM2 in leukocytes can be a biomarker of AD pathogenesis. Our study may be an important step in understanding epigenetic mechanisms underlying AD pathogenesis.
Acknowledgments We wish to thank Ms. Chiemi Onishi for technical assistance. This work was partially supported by a Health and Labor Science Research Grant from the Japanese Ministry of Health, Labour and Welfare and a Grant-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science and Technology, JSPS KAKENHI Grant Numbers 15K09808 and 16K21207.