Further evidence for plasma progranulin as a biomarker in bipolar disorder

Background: A recent study suggested that progranulin (encoded by the fronto-temporal dementia risk gene GRN) plasma levels are decreased in bipolar disorder (BD). Replication of this finding is however lacking. Methods: Progranulin plasma levels of bipolar patients (n¼104) and healthy controls (n¼80) were measured by enzyme-linked immunosorbent assay (ELISA). Participants were also genotyped for three single nucleotide polymorphisms (SNPs) in the GRN gene (rs2879096, rs4792938 and rs5848), and the effect of genetic variation on progranulin levels was examined. Results: Plasma progranulin levels were decreased in BD (ANCOVA, p¼0.001). Furthermore, age was significantly and positively correlated with plasma progranulin (Pearson's correlation, r¼0.269, po0.001). Also, lithium treatment but no other medication had a significant effect on progranulin plasma levels (ANCOVA, p¼0.007). Specifically in BD, the GRN SNP rs5848 was associated with progranulin plasma levels (Kruskal–Wallis test, po0.005). Limitations: Subgroup analysis regarding bipolar I vs. bipolar II subtype and polarity of the episode at sampling (manic vs. depressed vs. mixed vs. rapid cycling vs. euthymic) could only be performed with limited validity due to the relatively small sample size. The suitability of peripheral progranulin as a biomarker for BD is limited due to the overlap between patients and controls. Conclusion: The findings strengthen the evidence for progranulin being involved in pathomechanisms of bipolar disorder, and suggest a genetic determinant of progranulin concentrations that is relevant specifically in bipolar patients.