*2.7. Statistical Analyses*

Concentration data were analyzed with the nonparametric Mann–Whitney test. In order to assess viability and radial growth speed, time was used as a regression variable. Survival data were compared with the Kaplan Meier test. For gene expression analysis, statistical significance was assessed using a one-way analysis of variance followed by Tukey's multiple comparison test or an unpaired Student's *t*-test when it corresponded, depending on the number of experimental groups under analysis. Graph Pad Prism 8.0 software (GraphPad Software, San Diego, CA, USA) was used for statistical analyses.

#### **3. Results and Discussion**

#### *3.1. Molecular and Phenotypic Characterization*

We reported the isolation of a native fungal strain from a cadaver of an adult specimen of the Chagas disease vector, *T. infestans*. After a first morphological characterization suggesting that the fungus belongs to *Beauveria* sp., we characterized the isolate by molecular tools in order to confirm the identity at species level. The alignment of both ITS and EF1-α fragment sequences showed high similarity (99.5 and 99%, respectively) with *B. bassiana sensu lato* genes (Figure 1A,B) and was named and deposited in local mycological repository (see Material and Methods Section) as Bb-C001.

**Figure 1.** Phylogenetic tree generated by the maximum likelihood analysis of the ribosomal internal transcribed spacer (ITS) (**A**), and elongation factor 1-α (EF1-α) (**B**) sequences of *Beauveria bassiana* Bb-C001 (boxed) and related isolates. The fungal isolates, origin and access numbers to the GenBank are detailed. Numbers on branches are bootstrap support from 1000 replicates.

Then, as *B. bassiana* strain GHA was previously used in field trials in 2008 and 2009 [5,6] in Salvador Mazza municipality, distant ~130 km from the site of Bb-C001 collection, we assayed a molecular identification using a SCAR marker in order to discard a potential reisolation of naturally dispersed GHA strain from Salvador Mazza to Santa Victoria Este. Sequence-characterized amplified region (SCAR) is a very useful PCR-based technique to differentiate between fungal isolates since it can be developed to be strain-specific. In this sense, Castrillo and coworkers have set a SCAR marker specific for GHA, sensitive enough to differentiate this strain from others *B. bassiana* isolates [26]. Thus, the absence of the SCAR amplicon corresponding to 455 bp in Bb-C001 (Figure 2) confirms that this isolate is different from Bb GHA.

**Figure 2.** PCR amplification of the sequence-characterized amplified region (SCAR) marker (455 bp) specific for *B. bassiana* strain GHA (lane 1) which is absent in the Bb-C001 isolate (lane 3). Lane 2, molecular weight marker.

Conidia production did not show any statistically significant differences between both strains according to the Mann–Whitney test (*U* = 40.00; *p* = 0.999), with mean values of 4.5 ± 1.6 × 10<sup>12</sup> conidia mL−<sup>1</sup> for GHA and 4.0 ± 2.4 × 10<sup>12</sup> conidia mL−<sup>1</sup> for Bb-C001 (Figure 3A). Germination percentage did not exhibit statistically significant differences between strains (*p* = 0.59) during the time period studied (0, 10, 20, and 30 days). Regression analysis showed that there is significant loss of viability with time for both stains, GHA (b = −0.4% viability day-1; *p* = 0.02) and Bb-C001 (b = −0.4% viability day-1; *p* = 0.01) (Figure 3B). Radial growth did not exhibit statistically significant differences between the assessed strains (*p* = 0.48), showing an increment in diameter with time. Values obtained were 0.074 ± 0.003 mm h−<sup>1</sup> (GHA) and 0.071 ± 0.002 mm h−<sup>1</sup> (Bb-C001) (Figure 3C).

**Figure 3.** Comparative physiological characteristics of the Bb-C001 and GHA isolates of *Beauveria bassiana*. (**A**) Production of conidia. (**B**) Germination of conidia collected from PDA cultures at different time periods. (**C**) Growth diameter of fungi grown in PDA at different time periods.

#### *3.2. Fungal Virulence Assessment*

Nine days after infection with fungal isolates, no survivors were observed in all *T. infestans* development stages studied (Figure 4). Mean lethal time (MLT) observed for Bb-C001 were 4.6 ± 0.5 days (first instar nymph), 7.2 ± 0.2 days (third instar nymph), and 6.7 ± 0.5 days (adults), whereas MLT obtained for GHA was 5.5 ± 0.4 days (first instar nymph), 7.4 ± 0.4 days (third instar nymph), and 7.9 ± 0.3 (adults). In relation to the survival rate, no statistically significant differences were observed according to the Kaplan and Meier tests between both strains. The isolation and use of native strains of entomopathogenic fungi generate a greater specificity in the infection process, since they are better adapted to the local natural conditions and their host [22,33–35].

**Figure 4.** Survival curves of *Triatoma infestans* exposed to both Bb-C001 and GHA isolates of *Beauveria bassiana*, and controls (healthy insects). (**A**) First instar nymphs. (**B**) Third instar nymphs. (**C**) Adults.

#### *3.3. Fungal Toxin Expression*

The expression pattern of Bb-C001 genes encoding for nonribosomal peptides was dependent on the type of nymphs inoculated (Py-S or Py-R), the conidial concentration used (1 × 10<sup>2</sup> or 1 × 10<sup>5</sup> conidia mL−1), and the time period assayed (three, six, or nine days after fungal inoculation). Even at low values, all three genes (*BbbslS*, *BbbeaS*, and *BbtenS*) were detected inside both Py-S and Py-R nymphs three days after inoculation. In Py-S insects treated with 1 × 10<sup>2</sup> conidia mL−1, the *BbbslS* gene peaked at day six, showing significant higher values than three and nine days post inoculation. *BbbeaS* also peaked at day six, but with lower values than *BbbslS*. *BbtenS* showed the lowest values at the three times assayed (Figure 4A). In Py-S treated with 1 × 10<sup>5</sup> conidia mL−1, *BbbslS* and *BbbeaS* also exhibited the highest values at day six post inoculation, with not significantly different values as on day nine after inoculation (Figure 5A). With respect to Py-R, also higher values were detected at day six post inoculation, *BbtenS* peaked with 1 × 10<sup>2</sup> conidia mL−1, and *BbbslS* with 1 × 10<sup>5</sup> conidia mL−<sup>1</sup> (Figure 5B). Comparing both Py-S and Py-R, *BbbslS* and *BbbeaS* genes were higher expressed inside Py-S than in Py-R nymphs at day six post immersion in 1 × 10<sup>2</sup> conidia mL−<sup>1</sup> (*BbbslS*, *p* < 0.001; *BbbeaS*,

*p* < 0.01) and in 1 × 10<sup>5</sup> conidia mL−<sup>1</sup> (*BbbeaS*, *p* < 0.001), and also at day nine post inoculation with 1 × 10<sup>5</sup> conidia mL−<sup>1</sup> (*BbbslS*, *p* < 0.05; *BbbeaS*, *p* < 0.01).

**Figure 5.** Normalized relative quantities (NRQ) of *Beauveria bassiana* transcripts encoding tenellin (*BbtenS*), beauvericin (*BbbeaS*) and bassianolide (*BbbslS*) synthetases into fourth instar nymphs of pyrethroid-susceptible (**A**) and pyrethroid-resistant (**B**) *Triatoma infestans* at different time periods after insect immersion in conidial suspensions. Values are means of three replicates ± SEM. Different letters indicate significant differences for a single gene through time. Asterisks indicate significant differences in gene expression at each time point. \* *p* < 0.05; \*\* *p* < 0.005; \*\*\*\* *p* < 0.00005.

The same three genes were measured in GHA strain during infection of the same host [13]. Although both strains exhibited a peak of expression at day six post inoculation, only *BbbeaS* was highly expressed in by both *B. bassiana* strains into Py-S insects. The accompanying gene with high expression was different, namely *BbtenS* in GHA [13], and *BbbslS* in Bb-C001 (this study). Although this result is bounded to a small piece of research, it might be possible that the expression of the plethora of secondary metabolites described in entomopathogenic fungi is being isolate-dependent but also is likely to depend upon the exposure level (i.e., the inoculum) to the pathogen and the physiological state or host condition (e.g., the pyrethroid-related behavior). In this regard, Pedrini et al. [5] have previously studied the susceptibility of both Py-S and Py-R insects to *B. bassiana*. They found no differences in fungal virulence towards different nymphal stages from both insect populations [5]. The same study demonstrates that although a cuticle thickening and higher surface hydrocarbon content of Py-R bugs (compared with Py-S insects) might be related to a reduced penetration of the pyrethroid, and thus contribute to decrease the effective dose of insecticide, these differences do not seem to affect the fungal contact and penetration through the cuticle [5]. However, the current study found some differences in fungal toxin production either inside Py-S or Py-R insects. We can speculate that these different barriers might be responsible for differences of the fungal inoculum starting the infection, and thus, ultimately, provoke differential expression of secondary metabolites within the insect body invasion process.

#### *3.4. Insect Immune Response*

The immune-related gene expression was also different between Py-S and Py-R insects, and the inductions were observed mostly at day nine post inoculation. Comparing with controls, Py-S nymphs treated with 1 × 10<sup>2</sup> conidia mL−<sup>1</sup> showed induction in the all three defensin genes at day nine post inoculation, and *Tidef-6* was induced also at day six post inoculation. At the higher dose assayed (1 × 10<sup>5</sup> conidia mL−1), the genes induced were *Tidef-1* at day nine and *Tidef-6* at day six post inoculation (Figure 6). *Tilimpet-2* was not induced either with time period or dose used.

**Figure 6.** Expression pattern of limpet (**A**) and defensin (**B**–**D**) genes in *Beauveria bassiana*-infected pyrethroid-susceptible nymphs of *Triatoma infestans*. Normalized relative quantities (NRQ) are shown at different time periods after insect immersion in conidial suspensions. Values are means of three replicates ± SEM. For each gene, different letters indicate significant differences for each treatment through time. Asterisks indicate significant differences in gene expression at each time point. \* *p* < 0.05; \*\* *p* < 0.005; \*\*\* *p* < 0.0005.

In Py-R insects, *Tilimpet-2* was induced at day three after treatment with 1 × 10<sup>2</sup> conidia mL−<sup>1</sup> and at day nine with 1 × 10<sup>5</sup> conidia mL−1, as same as the three defensin genes. *Tidef-2* was also induced at day six after treatment with 1 × 10<sup>2</sup> conidia mL−<sup>1</sup> (Figure 7). Comparing the immune response in both Py-S and Py-R at day nine post inoculation with 1 × 10<sup>2</sup> conidia mL−1, *Tidef-1*, *-2* and *-6* were more expressed in Py-S than in Py-R nymphs (*p* < 0.001, *p* < 0.001, *p* < 0.05, respectively). On the contrary, in insects inoculated with 1 × 10<sup>5</sup> conidia mL−<sup>1</sup> at same time period, Py-R showed higher expression in *Tidef-2* (*p* < 0.001) and *Tidef-6* (*p* < 0.001).

Two variants of the limpet transcription factor were previously characterized in *T. infestans*, and their function have been linked with the humoral innate immune response [14]. As *Tilimpet-2* showed to be more involved than *Tilimpet-1* in defensin regulation, and the most affected defensin genes after limpet silencing were *Tidef-1*, *Tidef-2* and *Tidef-6* [14], we selected these genes to follow the host response in the current study. Although we found similar results as those in the previous work, the role of *Tilimpet-2* in this process (i.e., peaking before defensing genes) was observed only in Py-R insects (day three and day six for limpet and defensins, respectively), but no differences in its expression were detected in Py-S insects. As the defensin genes were more expressed in Py-S nymphs, it might be possible that the induction of *Tilimpet-2* has taken place before day three and thus it has not been detected by this study design.

**Figure 7.** Expression pattern of limpet (**A**) and defensin (**B**–**D**) genes in *Beauveria bassiana*-infected pyrethroid-resistant nymphs of *Triatoma infestans*. Normalized relative quantities (NRQ) are shown at different time periods after insect immersion in conidial suspensions. Values are means of three replicates ± SEM. For each gene, different letters indicate significant differences for each treatment through time. Asterisks indicate significant differences in gene expression at each time point. \* *p* < 0.05; \*\* *p* < 0.005.
