Deep serological profiling of the Trypanosoma cruzi TSSA antigen reveals different epitopes and modes of recognition by Chagas disease patients

Abstract

Background: Trypanosoma cruzi, the agent of Chagas disease, displays a highly structured population, with multiple strains that can be grouped into 6-7 evolutionary lineages showing variable eco-epidemiological traits and likely also distinct disease-associated features. Previous works have shown that antibody responses to 'isoforms' of the polymorphic parasite antigen TSSA enable robust and sensitive identification of the infecting strain with near lineage-level resolution. To optimize the serotyping performance of this molecule, we herein used a combination of immunosignaturing approaches based on peptide microarrays and serum samples from Chagas disease patients to establish a deep linear B-cell epitope profiling of TSSA.

Methods/principle findings: Our assays revealed variations in the seroprevalence of TSSA isoforms among Chagas disease populations from different settings, hence strongly supporting the differential distribution of parasite lineages in domestic cycles across the Americas. Alanine scanning mutagenesis and the use of peptides of different lengths allowed us to identify key residues involved in antibody pairing and the presence of three discrete B-cell linear epitopes in TSSAII, the isoform with highest seroprevalence in human infections. Comprehensive screening of parasite genomic repositories led to the discovery of 9 novel T. cruzi TSSA variants and one TSSA sequence from the phylogenetically related bat parasite T. cruzi marinkellei. Further residue permutation analyses enabled the identification of diagnostically relevant or non-relevant substitutions among TSSA natural polymorphisms. Interestingly, T. cruzi marinkellei TSSA displayed specific serorecognition by one chronic Chagas disease patient from Colombia, which warrant further investigations on the diagnostic impact of such atypical TSSA.

Conclusions/significance: Overall, our findings shed new light into TSSA evolution, epitope landscape and modes of recognition by Chagas disease patients; and have practical implications for the design and/or evaluation of T. cruzi serotyping strategies.

Fig 1. Microarrays design and reactivity evaluation.

A) Complete sequence of T. cruzi TSSA reference variants (TSSAI [AFF60282.1], TSSAII [XP_808931.1], TSSAIII [TXP_805410.1] and TSSAIV [AAQ73324.1]). Polymorphic positions are indicated with pink dots. B) Diagram showing peptide nomenclature, and the calculations of the overall reactivity of a specific sequence and residue. Peptides were named by the identity and relative number of the residue occupying its middle position (taking as 1 the Methionine proposed as site of TSSA translation initiation), using standard single-letter abbreviations. C) Diagram showing chimeras generated from variable sequences.

Fig 2. Differential seroprevalence of TSSA variants across the Americas.

A) Arrays made up of consecutive 16mer peptides and spanning the complete sequences of TSSA reference variants were probed with serum samples from chronic Chagas disease patients of different geographic origin, and the reactivity of each TSSA (indicated with a color scale) was calculated as the sum of the reactivities of its constituent peptides. B) TSSA reactivity features of the population analyzed as a whole (top), or upon stratification in Southern cone countries’ samples (middle) and Northern South America and North American samples (bottom). In every case, the overall TSSA seroprevalence is indicated by pie charts to the left. The bar charts show the primary reactivity (the most reactive TSSA isoform, left chart), and the number of isoforms recognized by TSSA-reactive sera (right chart). The y-axis refers to the number of samples.

Fig 3. B-cell epitope fingerprinting in TSSAII.

Peptide arrays comprising distinct sets of consecutive kmers, overlapped by k-1 residues and spanning the sequence of TSSAII^24-62 (or the complete TSSAII sequence for 15mers) were probed with the indicated pools of T. cruzi-infected sera. The mean reactivity of each peptide is shown. Residues included in epitopes A and B, as defined by 8mers arrays, are indicated in purple and green, respectively. Epitope C (in pink) was solely highlighted by the Brazilian pool of sera. Reactivities for peptides pE59 to pS62 were calculated solely with 15mer data.

Fig 4. Phylogenetic relationships and features of TSSA proteins.

A) Unrooted phylogenetic tree constructed from deduced TSSA amino acid sequences using the Maximum Likelihood estimation based on p-distance (TcMARK TSSA distance = 0.434). Bootstrap values are represented by a light blue dot on each branch. B) Correlation plot showing, for each unique TSSA sequence, the isoform assigned and the informed genomic classification of the strain in which it was found. The number of sequences per isoform/DTU combination is indicated with size-scaled dots. C) Comparison of the deduced amino acid sequences from TSSA variants identified in this work and TSSA reference sequences. Amino acid identity with the TSSAII reference sequence is indicated by dots. The predicted signal peptide and GPI anchor are indicated.

Fig 5. TSSA polymorphisms and antibody recognition.

Peptide arrays comprising sets of fully overlapped 15mers, spanning residues 30 to 55 of TSSAII reference variant (TSSAII^30-55) and bearing substitution(s) at polymorphic positions were probed with pools of chronic Chagas disease sera of different geographic origin. A) Reactivity of wild-type (wt) and single-substituted sequences for each pool. B) Mean reactivity of each residue in the context of wt (top row) or single-substituted sequences is indicated with a color scale. For each sequence, the replaced residue is marked in bold. C) Reactivity of sequences bearing combinations of either non-relevant (NR, green dots) or relevant (R, red dots) substitutions. Amino acid identity at the indicated positions in sequences bearing 4 relevant substitutions (for the Argentina pool sample) is shown in the boxes, where original and replaced residues are depicted in black and red, respectively. D) Reactivity of sequences bearing every possible combination of R and NR substitutions. For the Argentina pool sample, the identity of residues at positions 38, 39 and 40 from molecules displaying high (upper box), intermediate (middle box) or low reactivity (lower box) is indicated as in (C).

Fig 6. TSSAII mutational scanning.

Peptide arrays comprising sets of fully overlapped 16mers, spanning residues 31 to 50 of TSSAII reference variant (TSSAII^31-50) and bearing an Alanine substitution (or Glycine, in case of A43 and A47) at the indicated position were probed with 32 sera from individuals showing primary reactivity to TSSAII. A) Heatmap showing the overall impact of a specific replacement for each sample. Signal difference is represented by a color range for reactivity loss (orange) or gain (blue). Samples were divided into three groups based on their TSSAII reactivity profile (see S1 Fig). B and C) TSSAII reactivity profile and sequence logos representing the impact of each replacement for all analyzed sera (B) or for individual samples: AR_P6 (epitope A dominance), BO_P3 and BR_P4 (undefined dominance) and BR_P3 (epitope B dominance) (C). Peptide reactivity (left Y axis) is indicated by the mean ± SD values of biological (B) or technical replicates (C). The mean reactivity change caused by residue mutation (right Y axis) is indicated in orange (negative) or blue (positive). The proposed limits for epitopes A and B are indicated above the logos.