LAUSR

Circulation. 2019;139:722–723. DOI: 10.1161/CIRCULATIONAHA.118.035782 722 Lutz Schomburg, Prof, Dr rer nat Olle Melander, Prof, MD, PhD To the Editor: We read with interest the article by Kikuchi et al,1 which presents experimental evidence for an involvement of selenoprotein P (SELENOP) in controlling pulmonary artery smooth muscle cells. The researchers studied SELENOP expression in hypertensive rats, transgenic mice, pulmonary artery smooth muscle cells, and clinical samples. Hypertensive patients and control subjects were compared and used to relate survival odds to serum SELENOP concentrations exceeding or not exceeding 26.9 mg/L (Figure 1K). This fascinating report is characterized by a selection of most appropriate in vitro and in vivo models combined with high-quality analytical techniques and a scholarly performed discussion. However, in our opinion, the SELENOP concentrations reported appear unphysiologically high. Biosynthesis of SELENOP ceases once a replete selenium status is achieved,2 and supplementation studies indicate maximal concentrations of 6 to 7 mg/L.2,3 With 51 kDa and 7.5 selenium atoms per SELENOP, a concentration of 7 mg/L corresponds to 137 nmol/L SELENOP, equivalent to 1.0 μmol/L SELENOP-associated selenium. In view that SELENOP contributes ≈50% to circulating Se levels in humans, this concentration is equivalent to 2.0 μmol/L total serum selenium and accords with the majority of clinical studies. Kikuchi et al1 report SELENOP concentrations as high as 100 mg/L, equivalent to 2 μmol/L SELENOP and translating into SELENOP-associated serum selenium exceeding 14 μmol/L and total serum selenium concentrations consequently exceeding 25 μmol/L. This value is far above the highest concentrations ever reported, even compared with a recent supplementation trial for assessing the maximum tolerated selenium dose in human patients.4 We assume that the commercial assay used is not suitable for quantifying human SELENOP from clinical samples. Unfortunately, suppliers of antibodies or immune assays rarely characterize their products in sufficient detail and leave the researchers in the uncomfortable situation of purchasing the expensive material and validating the analytical performance. Unreliable assays have already led to reports of SELENOP concentrations in healthy subjects differing by 10 000-fold, as highlighted recently.5 Notably, these studies usually fail to present a positive correlation of total selenium and SELENOP in the study cohort, which would constitute a suitable test for data quality. Besides this problematic issue of the fine study presented by Kikuchi et al,1 we fully appreciate the comprehensive experimental approach of the colleagues, the wealth of valuable results presented, and the timely quest for an accessible and eagerly awaited biomarker for this life-threatening condition.