Chagas Disease Diagnosis with Trypanosoma cruzi-Exclusive Epitopes in GFP

Abstract

Serological tests are critical tools in the fight against infectious disease. They detect antibodies produced during an adaptive immune response against a pathogen with an immunological reagent, whose antibody binding characteristics define the specificity and sensitivity of the assay. While pathogen proteins have conveniently served as reagents, their performance is limited by the natural grouping of specific and non-specific antibody binding sites, epitopes. An attractive solution is to build synthetic proteins that only contains pathogen-specific epitopes, which could theoretically reach 100% specificity. However, the genesis of de novo proteins remains a challenge. To address the uncertainty of producing a synthetic protein, we have repurposed the beta barrel of fluorescent proteins into a receptacle that can receive several epitope sequences without compromising its ability to be expressed. Here, two versions of a multiepitope protein were built using the receptacle that differ by their grouping of epitopes specific to the parasite Trypanosoma cruzi, the causative agent for Chagas disease. An evaluation of their performance as the capture reagent in ELISAs showed near-complete agreement with recommended diagnostic protocols. The results suggest that a single assay could be developed for the diagnosis of Chagas disease and that this approach could be applied to other diseases.

Keywords: Chagas disease; ELISA; Green Fluorescent Protein; Serodiagnosis; Trypanosoma cruzi.

Figure 1

Sequences of the ß-receptacle, DxCruziV1, and DxCruziV2. Panel (A) shows the amino acid sequence selected as the ß-receptacle along with a back translation to its DNA sequence. For insertion into pET28(a), a 5′ NdeI site and a 3′ XhoI site were used to remove the multiple cloning sites. Based on the sites previously used to insert extraneous sequences (Table S2), codons were chosen to generate unique restriction sites for HindIII, SacII, BamHI, and RsrII. The addition of a single amino acid (bold beneath the underlined codon) could form the unique sites AatII, KpnI, AflII, and SacI, which did not eliminate fluorescence. Colored amino acids depict amino acids altered via the directed evolution of eGFP (Table S3). Panel (B) shows the protein sequences of the ß-receptacle, DxCruziV1, and DxCruziV2 aligned using Muscle in SnapGene (V7.0.2, GSL Biotech, San Diego, CA, USA). Conserved amino acids are shown in red.

Figure 2

Projected structure of DxCruziV1 and its performance as an antibody capture molecule for in-house ELISAs. Panel (A) depicts three views of the tertiary structure predicted on the I-Tasser server with the inserted epitopes in red and the core structure in green. Panel (B) shows the optical density measured at 405 nm for patient samples previously diagnosed by the LACENs in the states of Maranhão, Sergipe, Ceará, and Paraíba IN Brazil to be positive (green diamonds; n = 60) and negative (red dots; n = 75) for Chagas disease. Panel (C) shows the reactivity indexes of healthy individuals (n = 24) and patients with inactive cutaneous Leishmaniasis (Leish-C; n = 1), visceral Leishmaniasis (Leish-V; n = 32), malaria (n = 12), or dengue (n = 20) calculated with the cutoff value obtained from an ROC analysis of the data in Panel (B).

Figure 3

Spot synthesis analysis of DxCruziV1. The coding sequence of DxCruziV1 was converted into a library of 64 consecutive peptides of 14 residues with a 6-amino-acid overlap that was synthesized in duplicate directly onto a cellulose membrane as a grid of 3 rows (A–C or D–F) and 24 columns. Panel (A) depicts the chemiluminescent image of the membrane after incubation with a pool of sera (n = 10) from Chagasic patients (Rows A–C) or non-Chagasic individuals (Rows D–F) for the detection of human IgG antibodies. Panel (B) displays the respective quantification of the signals measured from the membrane in Panel (A). The blue dotted lines indicate the cutoff point used to select positives from negatives. Below the line indicates background. The list of peptides and their positions is presented in Table S4.

Figure 4

Projected structure of DxCruziV2 and its performance as the antibody capture molecule for in-house ELISAs. Panel (A) depicts three views of the projected tertiary structure from the I-Tasser server, with the inserted epitopes in red and the core structure in green. Panel (B) shows the absorbance values for patient samples previously diagnosed by LACENs in the states of Maranhão, Sergipe, Ceará, and Paraíba in Brazil as positive (green diamonds; n = 60) and negative (red dots; n = 75) for Chagas disease. Panel (C) shows the reactivity indexes of healthy individuals (n = 24) and patients with inactive cutaneous Leishmaniasis (Leish-C; n = 1), visceral Leishmaniasis (Leish-V; n = 32), malaria (n = 12), or dengue (n = 20) calculated with the cutoff value obtained from an ROC analysis of the data in Panel (B).

Figure 5

Analytical and geographical sensitivity of DxCruziV1 and DxCruziV2. In-house ELISAs were prepared with 500 ng/well of DxCruziV1 or DxCruziV2. Reactive indices for a 1:2 serial dilution up to 1:256 of the WHO International Biological Standard for Chagas disease 09/188 (Panel (A)) and 09/186 (Panel (B)) detected using anti-human IgG secondaries antibodies conjugated with horseradish peroxidase (HRP; orange symbols) or alkaline phosphatase (AP; blue symbols). Each data point represents the median of three independent measurements, except for DxCruziV2 HRP at 1:8 dilution of 09/186, which showed zero reactivity in one measurement and was excluded. The influence of the geographical origins of patient samples on performance was evaluated via Brown–Forsythe and Welch ANOVA tests between DxCruziV1 and DxCruziV2 (Panel (C)) or DxCruziV2 against different locations (Panel (D)) with the p-value displayed. Comparisons displayed were insignificant. * The target dilution factor suggested by the WHO [32]. ^† Maximum dilution factor with a reactive index > 1 determined via commercial tests in Brazil from a previous study [49]. AM—Amazonas; MA—Maranhão; CE—Ceará; PB—Paraíba; SE—Sergipe.