Synthetic peptides in the diagnosis of human echinococcosis

Echinococcus granulosus and Echinococcus multilocularis are the two tapeworm species responsible for the majority of human echinococcosis cases. Infection of this zoonosis can be acquired all over the world by accidental ingestion of eggs. The prevalence of echinococcosis is highest in areas with extensive animal husbandry.
Most diagnostic tests for the detection of antibodies against echinococcosis in humans are based on crude or partly purified native antigen extracts. The production of standardized diagnostic-grade native antigen is not possible. Native antigen suffers from batch-to-batch variation depending on purity, origin and developmental stage of the parasite material. In addition, native antigen is limited in availability.
Substantial progress in the development of standardized reagents has been achieved by the production of recombinant antigens. Consequent further development of antigens might lead to the production of synthetic antigen.
This PhD thesis investigated the applicability of synthetic peptides ranging from 24 to 47 amino acids in length for the use as synthetic antigens in serological diagnosis of cystic and alveolar echinococcosis. A bioinformatic selection procedure was established for identification of potentially antigenic protein sequences. From each of these parent proteins one or several peptides were selected for chemical synthesis. Peptides were designed from predicted domains of alpha-helical coiled-coils and intrinsically unstructured regions. These two motifs are capable of adopting their natural conformation even if synthesized as isolated fragments. The diagnostic performance of a synthetic antigen depends on its ability to fold into its native structure and thus be recognized by antibodies that had been naturally induced in the course of an infection. Peptide candidates were designed from different sets of parent proteins which had been selected by three different approaches: i) a genomics approach, ii) a proteomics approach, and iii) a transcriptomics approach.
These three approaches identified 6 promising peptide candidates. The synthetic peptide performing best obtained a sensitivity of 74% for the detection of CE infection and 43% for the detection of AE infection in adult patients. Specificity was 94%. All identified peptides, apart from one candidate, were not able to discriminate between CE and AE infection.
The major drawback of using peptides for serodiagnosis was their reduced sensitivity compared to EgHF, the native antigen used in routine diagnostics. Our results indicated that test sensitivity could be increased by combination of several peptides into a pool of synthetic antigen. This strategy can solve the problem of decreased sensitivity in future assay development.
The most promising candidate, peptide longD8-9, was investigated as potential marker for serological follow-up of treatment success in young patients. This represents a novelty, as synthetic peptides have not yet been tested as marker for serological follow-up of CE patients.
The performance of longD8-9-ELISA compared well to that of EgHF-ELISA, but it was superior in that within three years after treatment, more cured CE patients reached seronegativity in peptide ELISA compared to EgHF ELISA. Non-cured CE patients produced highly positive test results until the end of the follow-up period.
This thesis provides proof-of-principle for the discovery of diagnostically relevant peptides by bioinformatic selection complemented with screening on a high-throughput microarray platform. It also showed the value of synthetic peptides as potential markers not only in primary diagnosis, but also in monitoring of treatment outcome in follow-up patients.