Preferential Induction of Anonical IMD and Toll Innate Immune Receptors by Bacterial Challenges in Triatoma Pallidipennis Primed with Gram-Negative and Gram-Positive Bacteria, Respectively

I. INTRODUCTION

Insects lack an adaptive immune defense system against invading microorganisms but respond with humoral and cellular effector mechanisms whose components have counterparts in vertebrates' innate immune responses [1-4]. These were mostly unraveled from studies in Drosophila melanogaster and Aedes aegypti [5-7]. The principal effectors comprise lectins, melanin, and antimicrobial peptides (AMPs) [8-13] produced mainly in the fat body, hemocytes and the digestive tract [14].

The Immune Deficiency (IMD) [15] and Toll [16] signaling pathways, whose counterparts in vertebrates are the tumor necrosis factor receptor (TNFR) and the interleukin-1 receptor (IL-1R), and the Toll-like receptors (TLRs) [17], respectively, are the major signaling pathways to produce AMPs [18]. Pathogen-associated molecular patterns (PAMPs) of invading microorganisms, that are recognized by pattern-recognition receptors (PRRs) [12, 13, 19, 20], activate these immune pathways. The canonical pathways were described mainly in the Dipterans Drosophila flies and Aedes mosquitoes.

Accordingly, Gram-negative bacteria are recognized by a transmembrane receptor that recruits and activates the IMD cascade leading to the expression of the AMP genes cecropin, attacin, diptericin, drosomycin .

Gram-positive bacteria and fungi bound to proSpatzle induce its proteolytic cleavage, Spatzle binds to the Toll receptor on the cell membrane [23], which in turn triggers the Toll cascade leading to the expression of the AMP genes drosomycin, defensin B, and metchnikowin [19, 24, 25). Although Gram-negative and Gram-positive bacteria preferentially activate IMD and Toll signaling pathways, respectively, there are cross interactions among these and other immune pathways such as JAK/STAT (26-28], and they could be synergetic [29].

The Toll and IMD pathways are fully present in most insects , but this is not true for several Hemipterans, including reduvids. In this family, ortholog molecules, including the PRRs, pgrp-lc and pgrp-la, and AMPs have been identified . In addition, there is evidence for the presence of key members of the IMD pathway (IMD and Relish) in R. prolixus, . But genes coding this pathway appear incomplete or absent in Triatoma dimidiata, Triatoma infestans and Triatoma pallidipennis . Nevertheless, it was documented in T. pallidipennis the preferential participation of the IMD pgrp-lc and toll receptor genes and the Relish transcription factor in the regulation of responses to Gram-positive and Gram-negative infections, respectively, but as in other insects, cross induction was observed .

In contrast to vertebrates, the mechanisms for adaptation and selection of immune effector molecules are lacking in insects. However, enhanced protection after a previous exposure to a pathogen, termed immune priming , has been documented in Diptera, Coleoptera, Lepidoptera, Homoptera, Hymenoptera, and Orthoptera (revised by Contreras et al. ). This immune priming is functionally homologous to the adaptive immune memory of vertebrates .

As no adaptive modifications of the protective molecules, such as AMPs occur, the molecular mechanisms participating in the induction of immune priming are mainly unknown. The increased response to a second encounter is associated with an increased production of immune effector molecules . Still, the participation of the components of the immune signaling cascades remains poorly explored .

We present herein evidence for immune priming responses to sublethal doses of Micrococcus luteus (Gram-positive) and Escherichia coli (Gram-negative) bacteria in T. pallidipennis, a primary vector of Chagas disease in Mexico [43].

In addition, we documented the preferential participation of the IMD and Toll immune pathways receptors in the responses to each bacterial challenge, respectively.

II. METHODS

2.1 Insects Rearing

Triatoma pallidipennis was obtained from a colony initiated with specimens collected in different locations in the Morelos State, Mexico. The colony was maintained in the insectary of the National Institute of Public Health of Mexico.

Fifth-stage nymphs were maintained at C and 70–80% relative humidity under a photoperiod of 12 h light and 12 h dark. They were fed rabbit blood after molting, using artificial feeders. All experiments were conducted using 10-12 days-post-feeding-fifth-instar nymphs. The protocols were approved by the Biosafety, Ethics and Research Committees of the National Institute of Public Health (file number, CB17-229, CB:1491, CI:1500).

2.2 Cultures of Escherichia Coli and Micrococcus luteus

Gram-negative Escherichia coli (DH5α8739 strain atcc.org) was grown overnight at C in Luria-Bertani Broth (Dibico, Mexico) with agitation. 100μL was added to 2mL of fresh medium in a test tube and incubated until reaching an optical density of 0.7 (O.D. 600 nm).

The culture was centrifuged at 8000 rpm at 4°C, and the pellet recovered in PBS-pH 7.4 (137mM NaCl, 2.7mM KCl, 10mM sodium phosphate). Bacteria concentrations were quantified by colony forming units (CFU) in Luria Bertani-Broth medium with 1.5% bacteriological agar (BD Bioxon, Becton Dickinson, Mexico). Lyophilized Gram-positive Micrococcus luteus (Sigma-Aldrich, M-0508) was diluted in 5mL of PBS-pH 7.4. The viability of the sample was determined in an aliquot of 10μL using Trypan Blue at 0.4% in

PBS. (Corning USA) and counted in a Neubauer chamber. Translucent white colonies were counted individually in an optical microscope (Olympus Optical Co, LTD, Japan) at 100X. Bacteria concentrations were determined by counting CFU of cultures in Luria Bertani-Broth medium with 1.5% bacteriological agar (BD Bioxon, Becton Dickinson, Mexico).

2.3 Infection of T. Pallidipennis with Gram-Negative and Gram-Positive Bacteria

Groups of ten fifth-stage nymphs, 10 to 12 days post-feeding, cold-anesthetized (4 °C), were injected through the cuticular inter-tegument between the head and the thorax with live CFU of M. luteus, CFU of E. coli in 20μL of PBS, control groups were inoculated with 20 μL of sterile PBS using a syringe (31GX8mm needle).

On 1-, 15-, and 21 days post-inoculation (priming), five specimens of each group were dissected, and their fat body tissues were recuperated. Tissue samples were stored in 200 LTRIzol (Thermo Fisher Scientific, Waltham, MA, USA) at until processing for quantitative real-time polymerase chain reaction (qPCR) of transcripts.

After 21 days post-inoculation, five to seven nymphs per group were challenged with the same quantities of M. luteus, E. coli bacteria or PBS as previously described. At one- and three-day post-challenge, the fat bodies of five cold-anesthetized specimens of each group were individually obtained in PBS. Tissue samples were stored in Eppendorf tubes in 200 L TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) at until processed for total RNA extraction. Each treatment had three replicates per group.

2.4 RNA Extraction and cDNA Synthesis.

Total RNA from fat body tissues was extracted using TRIzol (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer's recommendations. Briefly, about 50 mg of each fat body sample in 200 L TRIzol in Eppendorf tubes (Thermo Fisher Scientific) were macerated using a biovortex. After adding 40 L of chloroform (Sigma-Aldrich, St. Louis, MO, USA), the preparations were mixed and centrifuged for 15 min at 10,000g at 4 °C. The aqueous phase was recovered, and 200 L of cold isopropanol (Sigma-Aldrich) was added, mixed, and incubated at -20 °C for one h. The samples were centrifuged at 10,000g for 10 min, and the pellets were washed with 500 L 75% ethanol and centrifuged at 10,000g for 15 min (Scilogex, D3024R centrifuge). The supernatants were removed, and the pellets were suspended in 40 L diethyl pyrocarbonate (DEPC, Sigma-Aldrich)-treated water. RNA was quantified with a NanoDrop 1000 spectrophotometer v. 3.7 (Thermo Fisher Scientific) and visualized using electrophoresis in agarose gels stained with EpiQuik DNA stain (EpiGentek, Farmingdale, NY, USA). Ten micrograms of total RNA were treated with 1 unit of DNase I (Thermo Fisher Scientific, Waltham, MA, USA) for 30 min at 37 °C, and subsequently inactivated with 1uL 25mM EDTA (ethylenediamenetetraacetic acid) at 75°C for 15 min.

First-strand cDNA synthesis was performed in 25 L reactions containing 10 g total RNA using an oligo dT 18 mero primer (Thermo Fisher Scientific) with 200 units of Reverse Transcriptase enzyme IV (SuperScript ®, Thermo Fisher Scientific). The synthesis reactions were incubated for one h at 42 °C, and then inactivated at 75°C for 15 min. The synthesized cDNA was diluted 1:10 with DEPC water and stored at -70 °C until use.

2.5 PCR of Tppgrp-lc, Tptoll, and AMP Transcripts

The transcription of Tppgrp-lc, Tptoll, Tpdefensin B, and Tpprolixicin was confirmed in cDNA templates by RT-PCR, using the T. pallidipennis -actin gene as control. Oligonucleotides sequences and PCR reaction conditions are reported in [35].

2.6 Quantitative PCR

We used qPCR to analyze the expression of Tppgrp-lc and Tptoll receptors and prolixicin and defensin B transcripts in individual cDNA samples of fat body tissue after the priming and challenge with M. luteus and E. coli as previously reported . Briefly, each reaction was performed in a final volume of 10 L, containing 2 L of cDNA (1:10), 1.5 pmol of each oligonucleotide, and 5 L of SYBR Green 2X Mix (NZY qPCR Green Master

Mix, nzytech, Lisbon, Portugal). qPCR was performed in a Rotor-Gene Q 5plex (Qiagen, Hilden, Germany). The qPCR conditions used were as follows: for 2min, for 5 min, 35 cycles of for 20s and for 60s. A melt curve analysis was conducted to confirm the specificity of the reaction. Controls without templates were included with each primer set to verify the absence of exogenous DNA and oligonucleotide dimers.

2.7 Statistical Analysis

The relative differences in the expression of transcripts were calculated using the method [44]. As endogenous control, we used the -actin gene. The values obtained from the analysis (Ctvalue of problem transcript – Ctvalue of -actin) were used to compare each transcript between groups ( ) in all experiments (e.g., priming group – PBS group). ANOVA not RM tests were performed to determine differences in gene expression between each treatment and their controls adjusted by {"image_source":{"path":"images/7539c9099ebb29d40edbc805e90b856b80b27ff66c1facf68b0f5bc7e16fbf5d.jpg"},"content":"","chart_caption":[],"chart_footnote":[]}

Tukeys analysis. Graphs were made using GraphPad Prism 6. P-values of P < 0.05 were considered significant.

C)

III. RESULTS

The expression of Tppgrp-lc, Tptoll, Tpdefensin B, and Tpprolixicin was estimated by PCR in the fat body of specimens of all experimental groups. Inoculation with E. coli and M. luteus increased the transcription of immune response genes, and significant differences were observed in some transcripts after the first bacterial infections and after challenges.

Tprolixicin increased significantly in insects infected (priming) with E. coli (4.08-fold, SE 2.00–6.15) on day one after priming, and its expression was reduced on days 15 to 21 after priming, but it increased to levels higher than those of the priming response (13.68-fold, SE 7.67-20.82–P = 0.0146 and 51.19-fold, SE 19.18-149.08 P = 0.0064) on days 1 and 3 respectively after challenge with the same bacteria. (Fig. 1A, Table1).

{"image_source":{"path":"images/8d6ded9777e43302b00f1fe0ffd6888f24fd0965af91df89e1074b782179ea54.jpg"},"content":"","chart_caption":[],"chart_footnote":[]} {"image_source":{"path":"images/a236034ccd8297544cc6462f1411267a22c064eb583fe811ea2be51a1d9ceb2a.jpg"},"content":"","chart_caption":[],"chart_footnote":[]}

{"image_source":{"path":"images/670a85d82b6dea332a2b85e03ee408b0413fb5d701b3d741da67d7fa662135e8.jpg"},"content":"","chart_caption":[{"type":"text","content":"Figure 1: Relative expression of Tpprolixicin (A), Tpdefensin B (B), Tppgrp-lc (C), and Tptoll (D) transcripts in fat body of priming and challenge T. pallidipennis with E. coli or M. luteus. In priming bugs, Tpprolixicin and Tptoll transcripts increased against both bacteria after a day post-priming and these increase was greater against M."}],"chart_footnote":[]} luteus after one and three-days post-challenge (A and D). Similar results were observed in Tpdefensin B and Tppgrp-lc, but the transcripts expression was greater against E. coli (B and C). Relative expression (2- CT) is the quantified change of expression between transcripts, asterisks indicate significant p value ( , , and ), bars represent the mean transcript levels CI. Groups were normalized with unchallenged and pbs group adjusted for -actin.

Table 1: Comparasion of the relative expression of Tpprolixicin, Tpdefensin B, Tpprgp-lc and Tptoll transcripts during the first bacterial infection (1-day post priming) and the second bacterial infection (1 and 3 days post challenge). Priming values are shown in gray, challenge values with increased expression are shown in green and those without significant increase are shown in red. All values are shown in fold change. Standard errors and statistical significance are shown in parentheses.

Like the infection with E. coli, the infection (priming) with M. luteus resulted in significant increases in the expression of Tprolixin (5.27-fold, 2.46-9.45), but its expression was more significant after 1 and 3 days of the challenge (51.40-fold, SE 27.85-90.50 P < 0.0001 and 112.81-fold, SE 29.44-222.86 P < 0.0001), respectively (Fig.1A, Table1). The increase in the levels of Tpprolixin transcripts induced by the challenge with M. luteus was higher than those by the challenge with E. coli (Fig. 2, Table 2).

Priming with E. coli produced an increase of Tpdefensin B transcripts on day one post-priming (13.36-fold, SE 7.03-25), and these decreased at days 15 to 21 post-priming, but the challenge with the same bacterium produced increased levels of the transcript (34-fold, SE 8.69–82.70 P > 0.05 and 90.29-fold, SE 43.10–170 P < 0.0001) on days 1 and 3 post-challenge with the same bacterium, respectively (Fig. 1B, Table 1).

Priming with M. luteus resulted in a moderate increase of Tpdefensin (5.48-fold, SE 2.07–12.70) on day one post-priming, and its level decreased on day 15 and remained low until day 21 post. priming. Challenge with the same bacterium did not induce statistically significant increases on days 1 and 3 post-challenge (9.21-fold, SE 4.25–28.24 P > 0.05 and 26.73-fold, SE 5.33–70.52 P > 0.05). However, the increases in Tpdefensin B transcripts induced by the challenge with E. coli were higher than those by the challenge with M. luteus (Fig. 2, table 2).

{"image_source":{"path":"images/ce8d98ccd2fd9e561461a68fc9ca9f3df117a5fbfc42fcc84b69a03a2324dabf.jpg"},"content":"","chart_caption":[{"type":"text","content":"Figure 2: Differential relative expression of AMPs and receptors of T. pallidipennis after three days post-challenge with E. coli or M. luteus. Tpdefensin B and Tppgrp-lc transcripts were more intensely expressed when inoculated with E. coli, while Tpprolixicin Tptoll were more expressed in response to M. luteus."}],"chart_footnote":[]}

Table 2: Differential expression of AMPs, and Tppgrp-lc and Tptoll receptors during the 1-and 3-days post challenge with E. coli or M. luteus. Tpdefensin B and Tppgrp-lc were more expressed against E. coli, while Tpprolixicin and Tptoll against M. luteus. The differential values with increased expression are shown with an asterisk. Differential expression was obtained by comparing the values of a specific transcript in response to infection of each bacterial species. The values with increased expression are shown in green and those with no significant increase are shown in red. All values are shown in fold change.

Preferential Induction of Canonical IMD and Toll Innate Immune Receptors by Bacterial Challenges in Triatoma Pallidipennis Primed with Gram-Negative and Gram-Positive Bacteria, Respectively

Tppgrp-lc increased after infection with E. coli (6.31-fold, SE 3.60–11.10) but transcript levels decreased on days 15 and 21. After challenge with the same bacterium, Tppgrp-lc increased significantly on three days post inoculation, (33.78-fold, SE 14.92-73 P < 0.0001) (Fig. 1C, Table1).

After priming with M. luteus, increases of Tppgrp-lc levels were observed on day one post-inoculation (4.47-fold, SE 2.14-6.03), and the transcript levels remained low up to 21 days. After a challenge with the same bacterium, transcription increased at levels higher than those observed during the first infection. Increased levels were higher after three days post-challenge (25.97-fold, SE 12.29–39.65 P < 0.0064). On day one, no statistically significant increases were observed (9.61-fold, SE 6.40–12.50 P > 0.05). The increase in Tppgrp-lc transcripts induced by the challenge with E. coli was higher than those by the challenge with M. luteus.

Tptoll transcripts increased after priming with E. coli (2.07-fold, SE 1.40-2.84) levels below those of pre-priming were observed on days 15 and 21. After challenge, Tptoll transcripts increased (9.42-fold, SE 6.02–12.01 P = 0.0003 and 12.98-fold, SE 10.28-16.28 P < 0.0001) on the first- and third-day post-injection, respectively. Using M. luteus, Tptoll transcripts increased during the priming (12.52-fold, SE 4.92-15.60), and the challenge with the same bacterium resulted in a new rise in TpToll, reaching higher levels that those observed after the first infection (24.02-fold, SE 17.80–29.12 P < 0.0001) on day three post-challenge. The increase in the levels of Tptoll transcripts induced by the challenge with M. luteus was higher than those by the challenge with E. coli (Fig. 2).

IV. DISCUSSION

We documented in T. pallidipennis the participation of Tppgrp-lc and Tptoll receptors in the induction of immune priming, and in the enhanced immune response after a second challenge with the same bacteria. In addition, we confirmed our previous observations on the preferential participation of these receptors in the recognition and induction of the primary immune innate response to bacterial infections.

As we employed sub-lethal doses of E. coli and M. luteus for the challenge, we could not assess any possible protection effect of the priming induced by first bacterial infections. Nevertheless, the defensin B and prolixin transcription responses after a second bacterial challenge fulfilled the essential parameters of priming: specificity and long-lasting biphasic response . However, in analyzing the molecular specificity of the reactions, some precisions are required. Although the infection with Gram-positive and Gram-negative bacteria induced the transcription of both AMPs, and transcription levels decreased with time in both infections. The Tpdefensin B initial response was higher after infection with E. coli than with M. luteus, confirming the preferential induction of the IMD pathway by the Gram-negative bacterium. The response to M. luteus could indicate possible Tppgrp-lc isoforms responding to Gram-positive bacteria, or the direct induction of components of the IMD pathway. Similar primary responses were observed in D. melanogaster

A second challenge with E. coli resulted in higher levels of Tpdefensin B transcript, while a second challenge with M. luteus did not induce significant changes in its transcription. After priming, the differences between the responses support a preferential response to the Gram-negative bacterium, which is in accord with a preferential IMD pathway induction. Although we did not explore possible changes in the transcription of other components of the IMD pathway, changes in Tppgrp-lc transcription after challenge with E. coli followed a similar pattern to that of Tpdefensin B. In contrast, after challenge with M. luteus, this was not significant. We speculate that the increased Tppgrp-lc receptor, augmented the recognition of the E. coli used for challenge, thus the enhanced Tpdefensin B response. This confirms a relative degree of the specificity of the priming response within the cross interacting IMD and Toll system, a condition favouring a broad-spectrum innate Immune response .

On the other hand, Tprolixin and Tptoll transcription levels increased after the first infection with both bacteria, but they were higher after infection with M. luteus. The participation of Tptoll and other Toll receptors in responses to Gram-positive, indicates higher receptor affinity to the Gram-positive bacterium. Toll pathway induction by Gram-positive and Gram-negative bacteria was observed in other insects [45, 46] and could explain our observations. In the same way, the transcription of Tprolixin and Tptoll was higher after a second challenge with both bacteria, but the challenge with M. luteus resulted in higher transcription levels than with the challenge with E. coli. The increase of the Toll receptor after priming could explain the enhanced response to M. luteus during the challenge.

Our observations reflect the diversity in the induction of the innate immune pathways among insects. It could be explained by cross-interactions between the members of two central innate immune cascades. The capacity of the receptors to interact with diverse molecules in Gram-positive and Gram-negative bacteria is shown by diverse toll receptors that could mediate the activation of immune responses against Gram-negative bacteria . On the other hand, as bacteria contain a mixture of antigens, it is possible that although Gram-negative bacteria preferentially induce the IMD immune system and Gram-positive the Toll system, some bacterial antigens may induce the other immune cascade than the canonically, up to now accepted. Nevertheless, the higher responses after the challenge are in order with the specificity of the priming inductor.

The mechanisms underlying immune priming remain poorly understood; however, specific receptor induction is expected for specific responses. Specific, or in the case of T. pallidipennis preferential, induction began with activating a particular receptor. Our results indicate that transcription of Tppgrp-lc and Tptoll receptors increased after the first exposition to their respective bacterium, indicating that a mechanism activates the production of more receptors than those already present before exposition. Similar results were observed in Anopheles gambiae exposed to bacteria .

Increased numbers of receptors could be responsible of the primary response, and having already more receptors could explain an increased response to the second exposure. As stablished in canonical priming (biphasic response), the transcription effectors Tpprolixin and Tpdefesin B subdued after three weeks, so did their receptors, probably because exhaustion of the respective antigens. Interestingly, an even higher receptors' transcription occurred after the challenge. This indicates possible not yet defined mechanisms responsible for induction of the transcription during the first pathogen encounter, specifically (or preferentially) triggered again with the challenge.

We did not explore the participation of other members of the IMD and Toll immune cascades that could play a role in directly modulating the immune response at the transcription level. There is evidence in Drosophila that Toll, but not IMD, is necessary, although insufficient for priming induction (48), but its participation still requires revealing. Epigenetic DNA/RNA dynamic modifications by methylation could regulate gene expression [49], possibly explaining "memory", but the underlying mechanism for enhancement is not responded. Endoreplication (DNA synthesis without cell division) has been documented in An. Albimanus infected with Sacharomyces cerevisiae [50] and Plasmodium berghei, probably increasing the number of immune genes copies during the first encounter with the pathogens. This, along with the transcription of regulatory elements of the cell cycle [51], could indicate the involvement of enhanced transcription in the immune response but does explain further enhancement after challenge.

In conclusion, we have documented in the Hemipteran T. pallidipennis the response to Gram-negative and Gram-positive bacteria with the main features of priming: biphasic and enhanced AMPs transcription response after a second challenge. Both phases were associated to the preferential induction of IMD and Toll receptors, respectively. The mechanisms underlying the induction of their transcription warrant further research.

ACKNOWLEDGEMENTS

This study was supported by the Consejo Nacional de Ciencia y Tecnología de México (CONACyT) (Grant 1500). The authors would like to thank Carlos Isaac Roldan, for helping with maintaining the triatomine colony.

The authors declare that they have no conflict of interest related to the findings or techniques in the present study.

List of abbreviations:

qPCR (quantitative polymerase chain reaction) pgrp-lc (Peptidoglycan recognition protein-long chain)

AMPs (Antimicrobial peptides)

IMD (Immune deficiency signalling pathway)