B-cell peptide epitopes as diagnostic targets for Q fever in sheep

Tamara Kozytska^1,2Claudia Gerlach¹Maksym Danchenko³Gabriela Flores-Ramirez³Hanka Brangsch¹Heinrich Neubauer¹Martin Pfeffer²Mathias Wilhelm Pletz⁴Ludovit Skultety³Katja Mertens-Scholz^1,4*

• ¹Friedrich-Loeffler-Institut Bundesforschungsinstitut fur Tiergesundheit Institut fur bakterielle Infektionen und Zoonosen, Jena, Germany
• ²Universitat Leipzig Veterinarmedizinische Fakultat, Leipzig, Germany
• ³Slovenska akademia vied, Bratislava, Slovakia
• ⁴Universitatsklinikum Jena, Jena, Germany

Q fever, caused by Coxiella burnetii, is a zoonotic disease of global relevance with domestic ruminants as main reservoirs. Serological diagnosis, especially ELISA, often suffers from limited sensitivity and specificity due to antigenic variability and cross-reactivity. In this study, a combined proteomic and literature research approach was used to identify immunoreactive proteins and predict linear B-cell epitopes as alternative diagnostic targets. Total protein extracts of a C. burnetii field isolate from sheep were separated by 2-dimensional gel electrophoresis and immunoreactive proteins were detected by Western blotting using pooled sheep sera obtained from various flocks with known Q fever status. Immunoreactive proteins were identified by LC-MS/MS and used for linear B-cell epitopes prediction for peptide synthesis. Peptides (n=30), were initially screened by fluorescent ELISA against nine field serum pools (90 individual sera) and most promising peptides (n=15) were individually tested with 79 single sera. Diagnostic performance was assessed by receiver operating characteristic (ROC) analysis and by a multi-peptide rule ("positive if ≥1 peptide reactive"). A total of 156 seroreactive proteins, including 51 previously reported antigens were detected, among others Com1, CBU_0482 and Mip. Although the selected, 15 peptides showed a specific reaction with pooled sera they showed limited diagnostic performance with an area under the curve (AUC) of 0.5 – 0.7 when using single serum samples (n=79). Multi-peptide combinations (6–8 peptides) increased sensitivity (Se) to 80% and specificity (Sp) to 75%. While single peptides lacked discriminatory power, multi-epitope combinations reached acceptable accuracy, and may be used as complementary tool for commercial ELISAs. However, larger bioinformatic approaches and validation studies are required to identify specific peptides of high diagnostic accuracy.