Effect of Sample Collection Tubes on Cerebrospinal Fluid Concentrations of Tau Proteins and Amyloid β Peptides

Tau protein and its phosphorylated forms, and amyloid β peptides ending at amino acid 42 (Aβ42) are used as cerebrospinal fluid (CSF) biomarkers of Alzheimer disease (AD) (1)(2)(3)(4)(5)(6)(7)(8). Because preanalytical factors may affect results (4)(9)(10), we measured these biomarkers in CSF samples in collection tubes made of different materials. After approval by the Ethics Committee of the University of Erlangen-Nuremberg, patients and/or their closest relatives gave written informed consent. CSF was obtained from patients (n = 20; mean age, 66 years; 9 women and 11 men) with various neuropsychiatric diseases, including dementias (n = 9), mild cognitive impairment (n = 8), recurrent depression (n = 1), bipolar disorder (n = 1), and emotionally unstable personality (n = 1). We included only CSF samples showing no or only minor contamination with erythrocytes (<150/μL), no intrathecal humoral or cellular immune response, and normal or only slightly disturbed blood–CSF barrier as measured by albumin quotient. After collecting 4–5 mL of CSF for routine analyses, we transferred four 0.5-mL fractions directly into 4 test tubes in this order: (a), polycarbonate (PC; Sarstedt; cat. no. 60.9922.936); (b), a copolymer of polystyrene and acrylonitrile [modified polystyrene (PX); Sarstedt; cat. no. 60.9924.952]; (c), polystyrene (PS; Sarstedt; cat. no. 62.553.542PS); and (d), polypropylene (PP; Sarstedt; cat. no. 62.554.502PP). Fractions were centrifuged simultaneously (1600g for 15 min at room temperature) immediately after the spinal puncture and were frozen at −80 °C within 30–40 min. The material was never thawed and refrozen. We used AΒx–40 and AΒx–42 methods from The Genetics Co. and Aβ1–42, total Tau, and P-tau181P from Innogenetics. Because different antibodies used in the respective assays show different epitope specificities, they enable detection of either specifically one species, namely Aβ1–42 [the assay of Innogenetics (11)], or the “family” of Aβ peptides ending with a COOH terminus at residue 42 and beginning at different NH2 termini (AΒx–42; the assays of The Genetics Co.). All 4 corresponding aliquots from a given patient were applied simultaneously to 1 ELISA plate. All measurements were performed in duplicate. Results are presented in reference to the concentration obtained in PP (set as 100%) and analyzed by ANOVA for repeated measurements, followed by the Scheffé post hoc test. AΒx–40 concentrations were lower in the PS tube than in all other material types (Table 11 ; n = 13; P <0.001), whereas concentrations in PC, PX, and PP were similar. CSF AΒx–42 concentrations were lower in the PS tube than in all other material tubes (Table 11 ; n = 19; P <0.001), whereas there were no significant differences among PC, PX, and PP. The AΒx–42/AΒx–40 concentration ratio was only slightly higher in PC than in PX (n = 13; P = 0.04), with no differences among other tubes. Aβ1–42 was lower on PS (Table 11 ) than in PC (P <0.01), PX, and PP (P <0.001). Aβ1–42/AΒx–42 was higher in PS than in PP and PC (n = 19; P <0.05). Tau was significantly lower (P <0.001) in the PS tube compared with all other tubes. The results also differed between PC and PX (P <0.05; Table 11 ). Neither Aβ1–42/AΒx–40 ratio (n = 13) nor P-tau181P (n = 16) differed significantly among test tubes (Table 11 ). Intraassay variability of the biomarkers was similar among the 4 material types with CVs of 1.8%–5.4%). Concentrations of none of the biomarkers differed significantly between the first and the last fraction of the CSF. Our study suggests that amyloid β peptide concentration ratios normalized to the concentration of AΒx–40 (i.e., AΒx–42/AΒx–40 and Aβ1–42/AΒx–40) and P-tau181P are much less prone to methodologic error introduced by interactions of the biomarkers’ molecules with the test tube surfaces compared with pure Aβ peptides concentrations or total tau. The decreased concentrations of Aβ1–42 and total tau in the PS tube agree with the results of Andreasen et al. (4). CSF Aβ peptides and tau may adsorb to PS. Because the AΒx–42/AΒx–40 and Aβ1–42/AΒx–40 concentration ratios seem not to be affected by the test tube material, we assume that Aβ peptides with COOH termini ending at residues 42 and 40 adhere to PS to a similar extent, which might be surprising, as Aβ peptides ending at amino acid position 42 (Aβ1–42 and AΒx–42) are more hydrophobic than is AΒx–40 (2)(3). The increase in Aβ1–42/AΒx–42 in PS compared with PC and PP suggests increased adherence of AΒx–42 molecules to PS, which might be explained by the fact that AΒx–42 is actually a set of different peptides, beginning at different N-terminal amino acid positions, most probably with different hydrophobic properties. Phosphorylation of the tau molecule increases its hydrophobicity and, presumably, decreases its adsorption to nonpolar plastic. We suggest that amyloid β peptide concentration ratios are more reliable biomarkers than are the pure Aβ peptides concentrations, as Aβ peptides ratios are not altered by interaction with the surface of the collection tubes. Relative concentrations and concentration ratios of biomarkers in reference to the results obtained in PP set at 100% P <0.01 compared with all other materials. P <0.05 compared with PC. P <0.05 compared with PP. Relative concentrations and concentration ratios of biomarkers in reference to the results obtained in PP set at 100% P <0.01 compared with all other materials. P <0.05 compared with PC. P <0.05 compared with PP. This study was supported by the following grants from the German Federal Ministry of Education and Research (BMBF): Kompetenznetz Demenzen (01 GI 0420) and HBPP-NGFN2 (01 GR 0447). We gratefully appreciate the technical help of Christine Schödel.