CUN. CHEM. 39/5, 865-869 (1993) CLINICAL CHEMISTRY, Vol. 39, No. 5, 1993 865 Interference of Lysophosphatidyicholine in Hormone Radioimmunoassays Nathalie Lepage,”2 Kenneth D. Roberts,14 and Jean Langlais23 The interference of synthetic and naturally occurring detergents in immunoassays is well documented. In the present study, we evaluated the effect of lysophosphati- dyicholine (LPC) and found that the lysophospholipid interfered with formation of the antigen-antibody complex in hormone immunoassays. In the presence of LPC (100 moI/L), progesterone was overestimated by 29%. Fur- thermore, physiological concentrations of LPC (140 moIfL) interfered with the assays of cortisol, progester- one, and aldosterone, resulting in overestimations of 35%, 30%, and 27%, respectively. The addition of albu- min decreased the interference by LPC to 7% in the assay of cortisol and progesterone when the LPC:albumin ratio was unity. Adding cholesterol (100 mol/L) also reduced by 50% the interference induced by LPC. Finally, treating plasma to increase the endogenous LPC concentration also resulted in interference in the cortisol assay. Thus, interpretation of the results of these assays should take into consideration the endogenous serum albumin:LPC ratio. IndexIng Terms: variation, source of phospholipids . albu- min #{149} cholesterol . aidosterone progesterone . androstenedione Clinical immunoassays are oriented toward simplic- ity and speed of execution. Current steroid assays use direct testing of the plasma, without consideration of the blood constituents. it is well known, however, that phospholipids and fatty acids interfere with steroid immunoassays (1, 2). Bile acids and detergents such as sodium dodecyl sulfate, Triton X-100, and anilinonaph- thalenesulfonic acid also interfere with RIAs of cAMP and gastrin (3) as well as assays of vitamin B12 (4) and digoxin (5). Here we report and characterize the inter- ference of lysophosphatidyicholine (LPC) in hormone RIAs.4 The LPC molecule, which has both detergent and lipid properties (6), is found in plasma either bound to albumin or in the free form (7). The mean physiological concentration of LPC in human plasma is 140 molfL (8). The three enzymes involved in the synthesis of LPC from phosphatidyicholine are phospholipase A2 (PLA2; EC 3.1.1.4), lecithin:cholesterol acyltransferase (LCAT; ‘Department of Biochemistry, University of Montreal, and 2Maisve4mont Hospital Research Center, 5415 l’As- somption Blvd., Montreal, Quebec, Canada, HiT 2M4. 5Mdress correspondence to this author, at Maisonneuve-Rose- mont Hospital Research Center. 4Nonstandard abbreviations: LPC, lysophosphatidylcholine; PLA2, phospholipase A2; LCAT, lecithin:cholesterol acyltransfer- ase; PAIl, platelet-activating factor acetylhydrolase; PBS, phos- phate-buffered saline; and DPC, Diagnostic Products Corp. Received June 9, 1992; accepted November 30, 1992. EC 2.3.1.135), and platelet-activating factor acetylhy- drolase (PAH; EC 3.1.1.48). PLA2 converts phosphati- dyicholine to LPC and free fatty acid (9); LCAT is involved in the same reaction plus the esterification of cholesterol (10). PAIL hydrolyzes phosphatidylcholine to yield LPC when this enzyme is in contact with oxidized low-density lipoprotein (11). In the present study, we investigated the effect of LPC on the formation of the antigen-antibody complex, us- ing immunoassays of the hormones cortisol, aldoste- rone, progesterone, and androstenedione and the pro- tein calmodulin. The four steroids were selected in view of the frequent demand for their analysis as well as for their wide range of polarities. The interference proper- ties of LPC were observed by adding exogenous LPC to the assay systems and also by treating plasma samples with PLA2 or with LCAT to increase the endogenous LPC concentration. Finally, we investigated the effects of albumin and cholesterol, natural LPC-binding mole- cules, on the interference mediated by LPC. Materials and Methods Assays Calibration curves were prepared as described by the suppliers or were diluted in phosphate-buffered saline (PBS; Sigma Chemical Co., St. Louis, MO) for studies made in the absence of serum. LPC (purity 99%, from egg yolk; Sigma) was added at various concentrations to standards to assess interference. Albumin (essentially free of fatty acids; Sigma) and cholesterol (Sigma) were added to test their effect on countering the effects of LPC. The first three kits below involve the use of polyclonal antibodies covalently linked to polypropylene tubes. For all five assays, radioactivity was measured with a LKB- Wallac (Turku, Finland) Model 1272 Climgamma counter. Cortisol. Experimental procedures were followed as described by Kallestad (Austin, TX). A first set of experiments tested the effect of LPC (0-1000 imol/L, final concentration), alone or with plasma constituents, on the cortisol calibration curve. A second set of exper- iments was performed with untreated plasma or with plasma treated with PLA2 or LCAT. Progesterone. Procedures were followed as described by Diagnostic Products Corp. (DPC; Los Angeles, CA). The interference of LPC (140 tmolIL, final concentra- tion) on the calibration curve was studied. In a second set of experiments, albumin at concentrations from 0 to 150 molJL was included to test its ability to overcome the interference from LPC, 100 JLmol/L. Aldosterone. Procedures were followed as described by DPC. LPC interference (0 to 1858 pmol/L) was tested on this calibration curve.