Blood samples were centrifuged at 1000× g for ten minutes and the supernatant was stored at −80 °C for subsequent analysis. Magnetic protein G beads (Dynal, Oslo, Norway) were incubated with the patient’s sera. After several washings, the patient’s antibodies were covalently bound to the beads using ethanolamine. The bead-antibody complexes were then incubated with homogenized retinal antigens. The antigens bound to the patient’s autoantibodies were be eluted, concentrated and analysed by SELDI time-of-flight (TOF) MS ProteinChips with two different chromatographic surfaces (CM10 cation exchange and H50 reversed phase). The samples were measured with a SELDI-TOF MS ProteinChip system (Bio-Rad, Hercules, CA, USA) on a PBS-IIc ProteinChip Reader. Raw data was transferred to CiphergenExpress 2.1 database software (Bio-Rad, Hercules, CA, USA) for workup and analysis. An in-house developed Proteomics Software Project (PSP) statistically evaluated the spectra using different statistical approaches to guarantee a high specificity and sensitivity of antibody patterns for the observed study groups. The PSP additionally searched for highly significant biomarkers directing a Statistical based analysis using above mentioned algorithms. The identification of biomarkers was performed by MALDI-TOF/TOF MS analysis (Bruker, Billerica, MA, USA). We aimed to generate at least eight highly specific biomarkers (significance level α = 0.05 and power (1 − ß) = 90%) for “wet” AMD.
Statistical calculations of sample sizes were based on experiences from previous studies (e.g., [28 (link)]): the calculated number of cases is sufficient to detect an effect on the serum antibody profiles, given a significance level α = 0.05 and power (1 − ß) = 90%. The statistical analysis demonstrated that the antibody composition against retinal antigens within sera change over time. A comparison to the control group showed if the modifications are beneficial, i.e., the serum compositions become more similar to the serum of healthy subjects or not. A subsequent biomarker identification using MALDI-TOF/TOF MS (Bruker, Billerica, MA, USA). revealed valuable hints on the systemic effects. After electrophoretic separation, proteins were be typically digested, crystallized on matrix and analysed on a MALDI target. The obtained peptide mass fingerprint data were exported into BioTools (Version 3.1, Bruker, MA, USA) and used for an internal Mascot database search (Matrix Science, London, England; Uniprot release 07, 2012), leading to protein identifications.
Statistical calculations of sample sizes were based on experiences from previous studies (e.g., [28 (link)]): the calculated number of cases is sufficient to detect an effect on the serum antibody profiles, given a significance level α = 0.05 and power (1 − ß) = 90%. The statistical analysis demonstrated that the antibody composition against retinal antigens within sera change over time. A comparison to the control group showed if the modifications are beneficial, i.e., the serum compositions become more similar to the serum of healthy subjects or not. A subsequent biomarker identification using MALDI-TOF/TOF MS (Bruker, Billerica, MA, USA). revealed valuable hints on the systemic effects. After electrophoretic separation, proteins were be typically digested, crystallized on matrix and analysed on a MALDI target. The obtained peptide mass fingerprint data were exported into BioTools (Version 3.1, Bruker, MA, USA) and used for an internal Mascot database search (Matrix Science, London, England; Uniprot release 07, 2012), leading to protein identifications.
Free full text: Click here
