https://orcid.org/0000-0002-9641-363X
Abstract
Chagas disease, caused by the protozoan Trypanosoma cruzi, is a zoonosis primarily found in rural areas of Latin America. It is considered a neglected tropical disease, and Triatoma dimidiata is the main vector of the parasite in Central America. Despite efforts, Chagas disease continues to be a public health concern, and vector control remains a primary tool to reduce transmission. In this study, we tested the hypothesis that highly abundant bacteria in the gut of T. dimidiata inhibit the growth of T. cruzi. To achieve this, bacterial diversity in the gut of T. dimidiata specimens from Costa Rica was characterized by metabarcoding of the 16S rRNA, microbial isolation was performed, and the effect of freeze-dried supernatants of the isolates on T. cruzi was investigated. Metabarcoding showed that the most abundant genera in the gut were Corynebacterium, Tsukamurella, Brevibacterium, and Staphylococcus. Barcoding and sequences comparison confirmed that 8 of the 30 most abundant amplicon sequence variants (ASVs) were isolated, and 2 of them showed an inhibitory effect on the growth of T. cruzi epimastigotes. These bacteria correspond to isolates of Tsukamurella and Brevibacterium, which were respectively the second and sixth most abundant ASVs in the gut of T. dimidiata. Notably, only the isolate of Brevibacterium showed a significant difference in growth inhibition against epimastigotes of both T. cruzi strains tested. These findings suggest that the gut microbiota of T. dimidiata may play an active role in modulating parasite development.
Introduction
Chagas disease, or American trypanosomiasis, is a parasitic zoonosis with high endemism in rural and low-income localities in many countries of Latin America (Prata 2001, Rassi and de Rezende 2012). This disease is caused by the protozoan parasite, Trypanosoma cruzi, whose transmission cycle includes different mammal reservoirs and is vectored by hematophagous insects in the subfamily Triatominae (Hemiptera: Reduviidae), mainly in the genera Triatoma, Panstrongylus, and Rhodnius (Coura and Viñas 2010). In vector transmission, the parasite infects humans through contact of lacerated skin or mucous membranes with the insect’s fecal droplets carrying metacyclic trypomastigotes, an event that often occurs during or immediately after blood feeding (Rassi and de Rezende 2012, Bern 2015). It is estimated that approximately 7 million people worldwide are infected with T. cruzi, although most acute infections go unnoticed and less than 30% of cases progress to chronic disease (Prata 2001, Bern 2015, World Health Organization 2022). Moreover, about 12,500 deaths are attributable to cardiac and gastrointestinal sequels of chronic Chagas disease every year (Strickland 2000, Prata 2001, Moncayo and Silveira 2017). Furthermore, American trypanosomiasis is considered a “neglected tropical disease”, given that the economic investments for developing a cure, a vaccine to prevent disease, or novel control methods are limited (World Health Organization 1992, Lee et al. 2013, Tarleton et al. 2014). Currently, there are 2 effective drugs to treat acute infection (benznidazole and nifurtimox), but they show lower effectiveness in curing chronic disease, as the parasites may persist in various organs and cause irreversible chronic tissue damage (Cançado 2002, Villar et al. 2019).
The regional Initiatives for Chagas Disease Control propose several alternatives to reduce the spread of this parasitosis. They include screening blood products to eliminate transfusion transmitted T. cruzi infections and reducing vector populations by indoor residual spraying of pyrethroid insecticides, improvement of human dwellings, and habitat removal/modification (Schofield 1992, Schofield and Dias 1999, Peterson et al. 2019). However, some vector species that are common in rural or urban areas also maintain sylvatic populations, which represent an important obstacle for insect elimination programs (Schofield 1994, Beard et al. 2001). In the last 20 years, several strategies have emerged as possible alternatives to block parasite transmission. For example, a para-transgenesis approach has been assayed in Rhodnius prolixus using Rhodococcus rhodnii, a bacterial symbiont in the insect gut, which has been genetically modified to express an antiprotozoal compound and make the insect resistant to T. cruzi infection (Durvasula et al. 1997, Beard et al. 2001, Ratcliffe et al. 2022). However, such chemical effects may also occur naturally in the intestine of the triatomine vectors. For instance, the recent idea that microbial symbionts may be naturally selected as mutualists due to their capacity to produce antibiotic-like compounds that protect their host against antagonists has prompted the discovery of novel compounds with pharmaceutical potential (Kaltenpoth et al. 2005, Seipke et al. 2012, Cambronero-Heinrichs et al. 2019, Chevrette and Currie 2019, Chevrette et al. 2019). Therefore, studying the microbial community in the gut of triatomine bugs could lead to the discovery of defensive mutualisms, where beneficial microbes inhibit parasitic protozoans in the insect host.
Previous studies have analyzed the diversity of microorganisms that inhabit the intestine of several vector species of T. cruzi, including variations in factors like geographical location, sex, developmental stage, T. cruzi infection status, and blood meal source (Handler and James, 2000, Matthews et al. 2011, Gumiel et al. 2015, Díaz et al. 2016, Dumonteil et al. 2018, Lopez-Ordonez et al. 2018, Mann et al. 2020, Jiménez-Cortés et al. 2021, Polonio et al. 2021). However, the differential microbiome composition of gut compartmentalization (foregut, midgut, and hindgut) has not been completely explored, since studies usually analyze the entire gut, focus on a single portion, or leave out one or more sections (Oliveira et al. 2018, Waltmann et al. 2019, Hu et al. 2020, Tobias et al. 2020, Murillo-Solano et al. 2021, Eberhard et al. 2022). In addition, not many studies describe the ecological and functional role of the possible mutualists in the gut of triatomine bugs. Traditionally, gut mutualists in hematophagous animals have been described in nutritional associations, where microbes supplement the host’s diet with B-complex vitamins, which the blood lacks (Duron and Gottlieb 2020). This is also the case for some bacterial isolates obtained from the gut of triatomines (Brecher and Wigglesworth 1944, Harington 1960, Hill et al. 1976, Beard et al. 2001). Conversely, different from a nutritional approach, a report about an isolate of Serratia marcescens obtained from R. prolixus showed an anti-trypanosome effect in an in vitro coculture assay (Azambuja et al. 2004). It was later shown that the strain of this bacterium can attach to the surface of T. cruzi epimastigotes during coculture (Castro et al. 2007, Garcia et al. 2010).
The present study tested the hypothesis that highly abundant bacteria in the gut of T. dimidiata can inhibit the growth of T. cruzi. With this purpose, the microbial diversity in the gut of T. dimidiata specimens from Costa Rica was characterized, and the inhibitory activity of bacteria isolated from the gut was tested against T. cruzi. The approach included a combination of 16S rRNA metabarcoding and culture-dependent methods for the isolation of representative bacteria, which were later screened for their inhibitory activity against epimastigotes of 2 T. cruzi strains. The results contribute to understanding the interactions between the vector microbiota and the parasite and provide evidence that microorganisms inhabiting the gut of triatomines can inhibit the growth of T. cruzi.
Materials and Methods
Insect Collection and Screening for Trypanosoma cruzi
Two collection efforts were performed to obtain wild Triatoma dimidiata. Collections were performed in a rural dwelling located in Chilamate de Poás, Alajuela, Costa Rica (10°03ʹ04″ N, 84°16ʹ08″ W) (Fig. 1A). The house’s peridomicile was characterized by poultry enclosures adjacent to firewood storage areas and an extra-domiciliary surrounding forest (Fig. 1B–D). The farm owners reported T. dimidiata and opossum sightings on the property. Two collection campaigns were performed respectively in April and May of 2020. Insects were collected by removing the habitat and capturing as many adults as possible by hand. Insects were kept in plastic containers and transported alive to the laboratory, where they were screened for the presence of T. cruzi within 24–48 h from capture.
Triatoma dimidiata sampling details. A) The sampling site was located at Chilamate de Poás, Alajuela, Costa Rica (10°03ʹ04″N 84°16ʹ08″W). B–C) The rural dwelling had a henhouse and stacked wood in the peridomicile. D) Triatomines were found around the henhouse. E) Three different sections of the insect gut were analyzed: anterior midgut (AM), posterior midgut (PM), and rectum (R).
To detect T. cruzi, a sample of every adult insect’s intestinal content was observed under the microscope. For this, the abdomen of each insect was gently pressed, and the intestinal content obtained was diluted in a drop of sterile saline solution (NaCl 0.9%), which was observed under a light microscope (400× magnification) to look for moving forms of T. cruzi (Stevens et al. 2021). The observation of any number of parasites preliminarily classified the insect as positive to T. cruzi, and at least 40 microscopic fields were observed before classifying any adult insect as negative. Positive and negative insects were placed in different containers and maintained in the laboratory before dissection. Considering that differences in blood meal sources can affect the microbiota of Reduviidae bugs (Murillo-Solano et al. 2021), all insects were fed once with mouse blood, using different mice for positive and negative insects. Only female insects were selected for subsequent assays in order to reduce variability related to sex.
Insect Dissection
After the first collection effort, 8 wild females (4 T. cruzi-negative and 4 T. cruzi-positive) were used for construction of 16S rRNA gene libraries. Insects from the second collection effort, 9 wild females (3 T. cruzi-positive and 6 T. cruzi-negative), were used for bacterial isolation. Dissections were conducted ten days after the laboratory blood meal. The females were placed in Petri dishes and then, for 20 min, in an ice-filled box to cold-inactivate them. To avoid external contamination during bacterial isolation, dissection of inactive specimens was performed in aseptic conditions, using sterilized instruments and in a laminar flow cabinet. First, insects were surface disinfected using ethanol 70%. Then, specimens were dissected, head, legs, and hemelytra were removed, and the connexivum was excised longitudinally, which facilitated the removal of the abdominal tergites. Finally, the exposed gut was carefully removed from the abdominal cavity and placed in a sterile Petri dish. The first section of the intestine (foregut) was not recovered, but the midgut and the rectum were obtained. The recovered section of the intestine was divided for subsequent methods into anterior midgut (AM), posterior midgut (PM), and rectum (R) (Fig. 1E).
For 16S rRNA gene libraries, sections from 8 wild female insects were placed individually in sterile plastic vials (3 vials per insect, a total of 24 samples) and stored at −80 °C until DNA extraction. For microbial isolation, gut sections from 9 wild (second sampling effort) females were pooled according to infection status (T. cruzi-positive or T. cruzi-negative) giving a total of 6 composite samples or pools (AM, PM, and R pools from T. cruzi-positive insects; and AM, PM, and R pools from T. cruzi-negative insects;). Pool preparation and microbial isolation was conducted immediately after the dissection process.
Total DNA Isolation, Construction of 16S rRNA Gene Libraries, and Illumina Sequencing
DNA from the individual gut sections (24 sections in total) was obtained using the PrepFiler BTA Forensic extraction kit (Thermo Fischer Scientific, USA), according to the manufacturer´s instructions. Extracts were used in a one-step PCR protocol, using the primers S35/S36 to confirm or not the microscope detection of T. cruzi (Mubiru et al. 2014). The amplicon library was constructed based on the amplification of the V4 hypervariable region of the 16S rRNA gene using universal primers 515F and 806R (Caporaso et al. 2011). Libraries were sequenced on a paired-end Illumina platform to generate 250bp paired-end raw reads (Illumina Novaseq, Novogene Bioinformatics Technology Co., Ltd., CA, USA).
Bioinformatic Analysis
The DADA2 version 1.18 was used to process the fastq files and generate a table of amplicon sequence variants (ASVs), higher-resolution analogs of the traditional OTUs (Callahan et al. 2016). Briefly, primers and adapters were removed, and the quality profiles of the reads was inspected. Sequences were filtered and trimmed with a quality score <30, estimating error rates, modeling, and correcting amplicon errors, and then the sequence variants were inferred. The forward and reverse reads were merged to obtain the full denoised sequences, chimeras were removed, and the ASV table was constructed. Taxonomy was assigned to the ASVs with the function assign taxonomy of DADA2, which uses as input the set of sequences to be classified and the reference sequences of SILVA database version 138 (Quast et al. 2013). Next, a second taxonomic assignment of the ASVs was carried out using the tool IDTAXA of DECIPHER (Murali et al. 2018) with the same version of the SILVA database and using the RDP database version 18 (https://rnacentral.org/expert-database/rdp). The consistency between the taxonomic assignments of the different programs was verified, followed by performing a manual curation. In cases of discrepancies, a comparison with the BLAST tool of NCBI Genbank was applied. This process generated 879.527 sequences from the 10 samples (3 positive and 7 negative for T. cruzi). The average number of sequences per sample was 109.940 ranging from 66.465 to 176.445. The raw sequencing data were deposited in the sequence-read archive of GenBank under the BioProject ID: PRJNA799595.
Microbial Isolation
For the isolation protocol, the 6 pools of the gut sections (3 from T. cruzi-positive females, 3 from T. cruzi- negative females) were macerated and vortexed in sterile plastic vials, using 1 ml of autoclaved NaCl 0.9% solution per sample. We prepared decimal dilutions of the samples (1 × 10−1 and 1 × 10−3), and they were plated on brain-heart infusion (BHI) agar (DIFCO Laboratories, Detroit, MI). A rich medium and aerobic conditions were used in order to isolate as many microbes as possible. Petri dishes were incubated at 28 °C and checked every 2 days for 4 wk, looking for colonies. All colony morphotypes per sample were isolated and transferred to new BHI Petri dishes. Then, each isolate was morphologically described, characterized by Gram staining, and stored at −80 °C in 20% glycerol (Sigma-Aldrich, St. Louis, MO, USA).
DNA Extraction and Identification of the Isolates
Genomic DNA of bacterial isolates was extracted using the NucleoSpin Tissue Kit (Macherey-Nagel, Germany), according to the manufacturer’s instructions, using colonies from solid media cultures (24 h to 48 h). For molecular identification, one-step PCR protocols were applied to amplify fragments of the 16S rRNA gene (16S), using primers 27/1492R, according to the protocol by Lane (1991). PCR products were purified using ExoSAP-IT (Applied Biosystems, CA, USA) and sent to Macrogen Inc. (Seoul, South Korea) for sequencing. DNA sequences were edited and assembled with DNA Baser software (Heracle BioSoft). To select the bacteria that show an identity of 100% with the ASVs in our analysis, the sequences of the 30 most abundant ASVs from the metabarcode and all the isolates were aligned using the MAFFT 7 CRBC Server (Katoh et al. 2019) and trimmed using MEGA11 (Tamura et al. 2021). To cluster the isolates and ASVs, a phylogenetic analysis was run in MEGA11 using the Maximum Likelihood method and Tamura-Nei model (Tamura and Nei 1993, 2021). The bootstrap consensus tree was inferred from 1,000 replicates, and the analysis involved 256 positions. Sequences from isolates and ASVs that clustered together were then aligned using BLAST to determine their identity (Altschul et al. 1990). Those isolates with 100.00% identity with a specific ASV were identified as the same bacteria and used in the following experiments.
Supernatant Preparation for Antagonism Assays
For bioactivity experiments, a freeze-dried preparation of the supernatant from a liquid culture of each isolate was used. To avoid any effect on epimastigotes due to residual growth medium in the supernatants, bacteria were incubated in the same medium that was later used in the experiment with T. cruzi, Liver Infusion and Tryptone (LIT) broth (Fernandes and Castellani 1966), starting at 0.05 OD and using overnight cultures as inoculum. Cultures were orbitally shaken (120 rpm) for 24 h at 30 °C. After this, supernatants were obtained by centrifugation (8,000 × g for 15 min) followed by filtration (0.22 μm, Millipore, USA). Triplicates of every isolate were freeze-dried at −80 °C and 0.2 mbar for 3 days. Each dry supernatant was weighted, diluted in water to 122 mg/ml, and stored at −80 °C for later use.
Evaluation of Supernatant Activity on T. cruzi Growth
To measure the effect of excreted secondary metabolites from the individual bacterial isolates on the protozoan, the effect of freeze-dried supernatants on the growth of T. cruzi epimastigotes was evaluated. Such effects were measured as changes on a growth curve of epimastigotes, using a colorimetric protocol based on the metabolic reduction of the dye MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium] to formazan, using PMS (1-methoxy phenazine methosulfate) as intermediate electron acceptor. The protocol was adapted from previous studies that test for purified compounds (Henriques et al. 2011, Chowdhury et al. 2018).
Two T. cruzi isolates were used as challenge strains for the assay: (i) ALF (NCBI: MH020170, DTU-I, TcI), isolated from T. dimidiata in Costa Rica (Bonilla et al. 2019); and (ii) Dm28c (DTU-I, TcI), originally obtained from an opossum (Didelphis marsupialis). The latter has a sequenced genome and is widely used as an experimental model (Contreras et al. 1988, Grisard et al. 2014).
The freeze-dried supernatants were diluted to 20 mg/ml in LIT broth (1% Streptomycin-Penicillin), and a suspension of epimastigotes was inoculated at an initial density of 1 × 106 cells/ml. An exponential culture of the protozoan that was previously incubated and quantified was used as the inoculum. Triplicates were conducted for every microbial freeze-dried supernatant, and the tubes were incubated for 96 h at 28 °C. Then, 100 μl of each sample was dispensed in a 96-well microplate, and 20 μl of a MTS-PMS (2 mg/ml—1 mg/ml) solution was added to every well, followed by an incubation of 120 min at 28 °C, after which the sample absorbance was measured at 490 nm. Of note, the time required for the isolates to reach the stationary phase was determined beforehand. Results were compared to the positive control containing benzimidazole (10 μM; Sigma-Aldrich, St. Louis, MO, USA) and no freeze-dried supernatants. Results were normalized to the negative control using the freeze-dried LIT medium. Finally, Kruskal–Wallis and Dunn’s tests were applied to determine significant differences (P-value < 0.05) between the sample supernatants and the negative control.
To evaluate the possibility of generalized toxic or other nonspecific inhibitory effects of the freeze-dried supernatants, the ability of isolates to inhibit the growth of bacteria and yeast was assessed using an agar disc diffusion method (Cambronero-Heinrichs et al. 2019). Test bacteria (Escherichia coli ATCC 25992, Staphylococcus aureus ATCC25923, Bacillus subtilis CGMCC 1.2428, Pseudomonas aeruginosa PAO1) and yeast (Candida albicans ATCC 10231) were challenged against a 10 μl aliquot of each freeze-dried supernatant (20 mg/ml), following the protocol described previously. Either Kanamycin (6 mg/mL, Sigma-Aldrich, St. Louis, MO, USA) or Clotrimazole (6 mg/ml, Sigma-Aldrich, St. Louis, MO, USA) was used as the positive control, while water was the negative control. Petri dishes were incubated at 37 °C, except for the experiments using B. subtillis (30 °C), and the presence of inhibition was determined after 24 h.
Results
Analysis of Microbial Communities
DNA sequences of acceptable quality were obtained from only 10 samples: 2 T. cruzi-positive samples from the PM, 1 T. cruzi-positive sample from the rectum, 3 T. cruzi-negative samples from the PM, 2 T. cruzi-negative samples from the AM, and 2 T. cruzi-negative sample from the rectum. Therefore, there were not enough samples to evaluate significant differences in the bacterial community composition related to gut section and T. cruzi infection status. However, descriptive data is reported and was linked to the identities of bacterial isolates.
The bacterial community in the gut of the T. dimidiata samples processed showed a total of 3,189 ASVs, according to the analysis of sequences of the V4 region of the 16S rRNA gene. All the bacterial sequences were assigned to 50 phyla and 113 classes. An average number of sequences per sample of 105,438 was obtained with a minimum of 66,465 and maximum of 176,445. As shown in Fig. 2A, Actinobacteriota was the most abundant phylum, representing 670,621 of all sequences, followed by Proteobacteria with 164,497, and Firmicutes with 149,696.
Bacterial community composition in the guts of the T. dimidiata analyzed in the study. Taxonomic composition of the 10 samples analyzed at the phylum A) and order B) level. Sample codes start with an insect’s individual number (samples with the same number come from the same insect). Capital letters in the code correspond to the gut section: anterior midgut (AM), posterior midgut (PM), and rectum (R); the lowercase letter indicates Trypanosoma cruzi infection status: negative (n) or positive (p).
Archaea Sequences Represented Less Than 0.02% of the Sequences
The bacterial order Corynebacteriales was highly abundant in almost all samples (Fig. 2B) (12.5–90.4% of all sequences). Interestingly, 2 of the 3 samples from individuals infected with T. cruzi showed an exceptionally low abundance of this order compared to all the other samples. Among this order, Corynebacterium (2.4–88.1%) and Tsukamurella (1.6–80.3%) were the most abundant genera (Fig. 3). Other abundant genera identified in the samples were Brevibacterium (9.2–27.9%), Cutibacterium (1.3–5.6%), Staphylococcus (1.8–39.5%), Lactobacillus (0–4.5%), Escherichia/Shigella (1.0–24.9%), Acinetobacter (2.5–4.8%), and Pseudomonas (1.5–1.6%) (Fig. 3). Genera such as Serratia, Rickettsiales, Chromobacterium, Bacillus, Brevibacullus, and Enterococcus composed less than 0.2% of the sequences found in the insects analyzed (Fig. 3).
Prokaryotic composition of the microbiota of the different T. dimidiata gut sections. Taxonomic composition of the 30 most abundant ASVs in all samples. Heatmap intensity represents the logarithmic transformed percentage abundance. Samples that begin with the same number come from the same insect.
Isolation of Microbes
In total, 19 bacterial isolates were obtained from the insect guts (Table S1 of Supplementary Material). All bacterial isolates were Gram-positive and almost all of them showed an identity of 100% with 8 of the 30 most abundant ASVs (Table 1). These ASVs include 5 Actinomycetota in the genera Corynebacterium (ASV1), Tsukamurella (ASV2), Brevibacterium (ASV6), Brachybacterium (ASV13), and Dietzia (ASV14); and 3 Bacillota, all in the genus Staphylococcus (ASV3, ASV17, ASV22). These bacteria include the 3 most abundant ASVs (ASV1, ASV2, and ASV3), which together represent more than 50% of the sequences from the metabarcoding analysis. Although different gut sections were used for isolation, metabarcoding data showed that the different isolates obtained could be found in any of the gut sections processed, regardless of T. cruzi infection status. Information about the source of all the isolates can be found in Supplementary Table S1.
Amplicon sequence variants (ASVs) from the gut of wild Triatoma dimidiata specimens that were obtained as isolated bacteria
| ASVs | Phylum (previous classification) | Designated genus |
|---|---|---|
| ASV1 | Actinomycetota (Actinobacteria) | Corynebacterium |
| ASV2a | Actinomycetota (Actinobacteria) | Tsukamurella |
| ASV3 | Bacillota (Firmicutes) | Staphylococcus |
| ASV6a | Actinomycetota (Actinobacteria) | Brevibacterium |
| ASV13 | Actinomycetota (Actinobacteria) | Brachybacterium |
| ASV14 | Actinomycetota (Actinobacteria) | Dietzia |
| ASV17 | Bacillota (Firmicutes) | Staphylococcus |
| ASV22 | Bacillota (Firmicutes) | Staphylococcus |
| ASVs | Phylum (previous classification) | Designated genus |
|---|---|---|
| ASV1 | Actinomycetota (Actinobacteria) | Corynebacterium |
| ASV2a | Actinomycetota (Actinobacteria) | Tsukamurella |
| ASV3 | Bacillota (Firmicutes) | Staphylococcus |
| ASV6a | Actinomycetota (Actinobacteria) | Brevibacterium |
| ASV13 | Actinomycetota (Actinobacteria) | Brachybacterium |
| ASV14 | Actinomycetota (Actinobacteria) | Dietzia |
| ASV17 | Bacillota (Firmicutes) | Staphylococcus |
| ASV22 | Bacillota (Firmicutes) | Staphylococcus |
aMicrobe isolates that showed a significant reduction on the growth of T. cruzi epimastigotes.
Amplicon sequence variants (ASVs) from the gut of wild Triatoma dimidiata specimens that were obtained as isolated bacteria
| ASVs | Phylum (previous classification) | Designated genus |
|---|---|---|
| ASV1 | Actinomycetota (Actinobacteria) | Corynebacterium |
| ASV2a | Actinomycetota (Actinobacteria) | Tsukamurella |
| ASV3 | Bacillota (Firmicutes) | Staphylococcus |
| ASV6a | Actinomycetota (Actinobacteria) | Brevibacterium |
| ASV13 | Actinomycetota (Actinobacteria) | Brachybacterium |
| ASV14 | Actinomycetota (Actinobacteria) | Dietzia |
| ASV17 | Bacillota (Firmicutes) | Staphylococcus |
| ASV22 | Bacillota (Firmicutes) | Staphylococcus |
| ASVs | Phylum (previous classification) | Designated genus |
|---|---|---|
| ASV1 | Actinomycetota (Actinobacteria) | Corynebacterium |
| ASV2a | Actinomycetota (Actinobacteria) | Tsukamurella |
| ASV3 | Bacillota (Firmicutes) | Staphylococcus |
| ASV6a | Actinomycetota (Actinobacteria) | Brevibacterium |
| ASV13 | Actinomycetota (Actinobacteria) | Brachybacterium |
| ASV14 | Actinomycetota (Actinobacteria) | Dietzia |
| ASV17 | Bacillota (Firmicutes) | Staphylococcus |
| ASV22 | Bacillota (Firmicutes) | Staphylococcus |
aMicrobe isolates that showed a significant reduction on the growth of T. cruzi epimastigotes.
Bioactivity of Isolates on T. cruzi
After testing the activity of the supernatants from bacterial isolates against 2 T. cruzi strains, the supernatants of 2 of the ASVs within the 30 most abundant microbes caused a significant decrease in epimastigote growth when compared to the negative control for the T. cruzi ALF strain (ASV2 and ASV6), while only the supernatant of ASV6 showed this effect for the Dm28c (Fig. 4). These isolates were classified in the genera Tsukamurella (ASV2) and Brevibacterium (ASV6).
Inhibitory activity of isolate supernatants against T. cruzi epimastigotes. Bacteria isolate supernatants that showed a significant epimastigote growth inhibition when compared to the negative control are marked with an asterisk. Results are shown only for the 8 isolates that belonged to the 30 most abundant ASVs.
All isolate supernatants were considered negative to nonspecific antibiosis against bacteria and yeast.
Discussion
This study shows that abundant Actinobacteriota in the gut of T. dimidiata can inhibit the growth of T. cruzi epimastigotes, an effect caused at least in part by excreted components of some of the bacteria isolated. The gut microbiota of triatomines have coevolved with parasitic protozoa, and this close interaction may have selected anti-T. cruzi activity in the insect host. In addition, different strains of T. cruzi were differently affected by the supernatants of the isolated bacteria in this study. However, it is important to note that this study was limited to the effect of the supernatants on epimastigotes, but the complex multidirectional chemical ecology in the gut of T. cruzi-infected triatomines was not tested.
Although the small number of samples sequenced in the present study prevents further statistical analyses of the metabarcoding data, results show that Actinobacteriota was the most abundant taxa in the gut of female T. dimidiata, followed by members of Firmicutes and Proteobacteria, which agrees with previous reports for triatomines (Salcedo-Porras et al. 2020, Eberhard et al. 2022). These 3 phyla have also been identified as the most frequent bacteria in the gut of adult T. dimidiata from Belize (Polonio et al. 2021). However, some phylotypes of highly relevant bacteria that have been documented in other species of triatomines were not found in the samples analyzed in this study, for example, the genus Wolbachia, which has been only observed in Rhodnius species (Eberhard et al. 2022).
Among the Actinobacteriota detected, most sequences belonged to the family Corynebacteriaceae (order Mycobacteriales). This family and Mycobacteriales in general, have been reported to be highly abundant in multiple species of blood-feeding triatomines (Díaz et al. 2016,Brown et al. 2020, Murillo-Solano et al. 2021). In particular, the symbiotic Rhodococcus rhodnii, first found in Rhodnius prolixus, is a model microorganism for describing the symbiosis between members of Mycobacteriales and triatomine bugs (Dotson et al. 2003). It is known that R. rhodnii produces vitamins of the B complex that supplement the nutrition of its host (Pachebat et al. 2013), suggesting that bacteria among Corynebacteriaceae may also play a nutritional role in other triatomines. Furthermore, bacteria in the genus Corynebacterium are known for synthesizing vitamins from the B complex (Lee and Hwang 2003, Al-Dilaimi et al. 2014). Corynebacterium was the most abundant genus in the present metabarcoding analysis, and it may be possible that organisms in this genus accomplish a nutritional role in T. dimidiata, analogous to R. rhodnii (Pachebat et al. 2013).
Other studies have established that microbial colonization in different gut sections can be determined by bloodmeal source, hormone-regulated perimicrovillar membrane development, and the presence of insect effectors from the immune response or salivary and digestive compounds (Garcia et al. 2010, de Fuentes-Vicente et al. 2018, Dumonteil et al. 2018). Therefore, further analyses of gut sections from a larger number of T. dimidiata are warranted to elucidate possible differences in their microbial communities. In addition, differences in the gut microbiota according to the presence or absence of T. cruzi have been previously observed in T. dimidiata, as well as other species of Triatoma and Rhodnius, suggesting that the presence of the parasites affects the structure of the microbial community of the triatomine’s gut (Vieira et al. 2015, Díaz et al. 2016, Waltmann et al. 2019, Polonio et al. 2021). Nonetheless, the type of samples analyzed in previous studies varies greatly between studies (feces, intestine tissues, whole bug, etc.).
In general, it has been accepted that T. cruzi infections produce a dysbiosis effect (Vieira et al. 2015, Díaz et al. 2016, Waltmann et al. 2019). This dysbiosis may be caused by an antimicrobial effect of T. cruzi, the resulting immune response in the insect host, and opportunistic species taking advantage of the colonization by the protozoan (Gresse et al. 2017, Kriss et al. 2018). Moreover, bacteria may cause glucose depletion and inhibit T. cruzi colonization (Orantes et al. 2018). In general, T. cruzi is expected to compete with the gut microbiota for resources and may modify its immediate bacterial community by reduced nitrite and nitrate production and increased phenoloxidase and antibacterial activities (Castro et al. 2012). Additionally, it has been previously demonstrated that T. cruzi alters the production of antimicrobial peptides by its host, which may enhance dysbiosis (Boulanger et al. 2006). An alternative hypothesis is that the augmented populations of some bacterial taxa may also respond to opportunistic species taking advantage of the protozoan colonization, as occurs in other animals dealing with certain microbes or pathogens (Sepkowitz 2002). For example, certain Serratia marcescens have been shown to promote or enhance the susceptibility to arboviral infection in mosquitoes (Apte-Deshpande et al. 2014, Wu et al. 2019), contrary to the effect observed with Wolbachia symbionts (Gao et al. 2020). The hypothesis of favored opportunistic populations is also compatible with the higher richness of microbes usually observed in triatomines infected with T. cruzi (Salcedo-Porras et al. 2020).
On the other hand, several studies suggest that the microbiota may play a protective role against T. cruzi infection in the insect vector. For instance, Arsenophonous appears to be a dominant genus in different species of reared Triatoma, and has been described as an intracellular symbiont in multiple insects (Nováková et al. 2009, Díaz et al. 2016). Some scientists claim that this genus may negatively interact with trypanosomes, but the bacterial genus is neither prevalent in samples from the present study nor in other studies (Murillo-Solano et al. 2021).
The antibiosis assays, performed to analyze the hypothesis that bacteria present in the intestine of T. dimidiata may affect T. cruzi growth, revealed that the isolates of 2 ASVs were able to inhibit the growth of at least one strain of T. cruzi. Moreover, these isolates did not influence the growth of several other microbes tested, suggesting specificity. The bacterial isolates that showed activity in these assays were identified as 2 highly abundant ASVs in the gut of T. dimidiata through the DNA metabarcoding analyses. It is important to note that the rich medium and aerobic conditions used for isolation of bacteria limit the results to those that grow in these specific conditions. In addition, the inhibition assay used may capture the effect of excreted secondary metabolites or other trypanosome-inhibiting compounds but will miss compounds that are delivered directly from a bacterium to the parasite via bacterial secretion systems. Moreover, it is possible that certain secondary metabolites may only be produced in vivo when bacteria are exposed to the complex web of microbial, parasite, and host factors present in gut of the insect host. Therefore, these additional possibilities should be investigated further to fully understand the role of triatomine gut microbiota in T. cruzi infection and transmission.
The bacteria isolates that inhibited T. cruzi growth matched ASV2 and ASV6, respectively identified as Tsukamurella sp. and Brevibacterium sp. The ASV2 was able to inhibit only one of the T. cruzi strains. Bacteria of the genus Tsukamurella are usually employed as a source of mycolic acid in coculture studies to induce the production of active metabolites by other Actinomycetota bacteria (Igarashi et al. 2010, Hoshino, Okada, et al. 2015, Hoshino, Wakimoto, et al. 2015, Sugiyama et al. 2015, 2016, Hoshino et al. 2017). However, antimicrobial compounds produced entirely by Tsukamerella spp. have not been reported. Only the isolate identified as Brevibacterium sp. (ASV6) produced a significant reduction against both strains of T. cruzi. Brevibacterium spp. have been found in different human samples, poultry, mammal fur, and soils (Rojas-Gätjens et al. 2022). Some species are known for producing volatile and sulfur-containing compounds, such as methanethiol, which is toxic to eukaryotes and relevant to food safety (Forquin et al. 2009, Forquin-Gomez et al. 2014). Members of this genera have also been found in association with other organisms, such as marine sponges and sloths (Indraningrat et al. 2016, Rojas-Gätjens et al. 2022), and they can produce multiple bacteriocins with specific activities (Valdes-Stauber and Scherer 1994, Motta and Brandelli 2002, 2003).
The present study showed differences in the inhibition of microbial isolates against the T. cruzi strain of the same genotype (TcI), something to be expected as the discrete typing unit, including all Tc1 isolates, is a diverse and heterogeneous group (Tibayrenc 2003). Such diversity can result in differences in susceptibility to drugs such as benznidazole, which is widely used to treat Chagas disease (Vela et al. 2021). Reports using Serratia marcescens also showed that this bacterium could inhibit the growth of certain isolates of T. cruzi but not others; for example, inhibition of the Y strain (TcII) but not the strain DM28c (TcI) (Azambuja et al. 2004). Therefore, some of the microbial inhabitants in the gut of triatomines that can inhibit T. cruzi growth may be protecting the insects against infection with specific genotypes or strains of the protozoan, depending on its regular exposure to those strains.
Conclusion
This study describes one of the functional roles that highly abundant bacteria may have in the gut of T. dimidiata. We propose that bacteria in the insect’s gut can negatively affect the growth of T. cruzi epimastigotes by secreted and inhibitory compounds, although this effect can be dependent on the strain of the parasite. Overall, this new functional information is highly relevant for understanding the multikingdom interaction occurring in the gut of triatomines. The gut microbiota of T. dimidiata may be affecting the development of T. cruzi. Moreover, results suggest that studying the chemical ecology and microbes in the gut of triatomines could lead to the discovery of molecules that could be used against T. cruzi.
Supplementary Material
Supplementary material is available at Journal of Medical Entomology online.
Acknowledgments
The authors thank Iván Coronado, Geison Rivera-Bemúdez, Guillermo Barquero, Jovita Chaves, and Victor Castro-Gutierrez for technical support during sampling and laboratory analysis. We are especially grateful for the artistic figures designed by Grettel Andrade.
Funding
This work was supported by the Vicerectory of Research of Universidad de Costa Rica (project numbers C0186 and B5735) and the National Center of Biotechnological Innovations (CENIBiot) of Costa Rica. Additionally, this study received the FEMS Congress Attendance Grant to participate in FEMS2022, Belgrade.
Ethical statement
This study was approved by the Biodiversity Commission of the University of Costa Rica (Resolution No. 235; project C0186), in agreement with the laws and regulations concerning access to biodiversity in Costa Rica.
Conflict of interest statement
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Author Contributions
Juan Carlos Cambronero-Heinrichs (Conceptualization [Lead], Data curation [Equal], Formal analysis [Equal], Investigation [Lead], Methodology [Equal], Writing—original draft [Lead], Writing—review & editing [Equal]), Diego Rojas-Gätjens (Data curation [Equal], Formal analysis [Equal], Investigation [Equal], Methodology [Equal], Writing—original draft [Equal]), Mónica Baizán (Investigation [Equal], Visualization [Equal]), Johan Alvarado-Ocampo (Investigation [Equal], Writing—original draft [Equal]), Keilor Rojas-Jimenez (Data curation [Equal], Formal analysis [Equal], Methodology [Equal], Writing—review & editing [Equal]), Randall Loaiza (Methodology [Equal], Resources [Equal], Supervision [Equal], Writing—review & editing [Equal]), Max Chavarría (Methodology [Equal], Supervision [Equal], Writing—review & editing [Equal]), Olger Calderon-Arguedas (Methodology [Equal], Writing—review & editing [Equal]), and Adriana Troyo (Conceptualization [Equal], Funding acquisition [Lead], Project administration [Lead], Resources [Equal], Supervision [Lead], Writing—review & editing [Lead])
