Immunological Roles of TmToll-2 in Response to Escherichia coli Systemic Infection in Tenebrio molitor

The antimicrobial roles of Toll-like receptors have been mainly identified in mammalian models and Drosophila. However, its immunological function in other insects has yet to be fully clarified. Here, we determined the innate immune response involvement of TmToll-2 encountering Gram-negative, Gram-positive, and fungal infection. Our data revealed that TmToll-2 expression could be induced by Escherichia coli, Staphylococcus aureus, and Candida albicans infections in the fat bodies, gut, Malpighian tubules, and hemolymph of Tenebrio molitor young larvae. However, TmToll-2 silencing via RNAi technology revealed that sole E. coli systemic infection caused mortality in the double-strand RNA TmToll-2-injected group compared with that in the control group. Further investigation indicated that in the absence of TmToll-2, the final effector of Toll signaling pathway, antimicrobial peptide (AMP) genes and relevant transcription factors were significantly downregulated, mainly E. coli post-insult. We showed that the expression of all AMP genes was suppressed in the main immune organ of insects, namely, fat bodies, in silenced individuals, while the relevant expressions were not affected after fungal infection. Thus, our research revealed the immunological roles of TmToll-2 in different organs of T. molitor in response to pathogenic insults.


Introduction
Toll and Toll-like receptors (TLRs) are a conserved family of pattern recognition receptors (PRRs) initially identified for their role in dorso-ventral axis formation in Drosophila embryos [1]. The TLR/nuclear factor-kappaB (NF-κB) pathway activates immune responses in Drosophila [2,3]. Since the first Toll protein (Toll1) and eight other Toll receptors  were characterized in Drosophila, 10 TLRs in human (TLR1-10) and 13 TLRs in mouse (TLR1-13) have been identified [4]. TLR activation and function can vary between vertebrates and invertebrates. In vertebrates, the Toll pathway is activated by direct interaction with pathogen-associated molecular patterns (PAMPs) [5]. Whereas, in invertebrates, Toll signaling is initiated by binding to the cytokine-like molecule Spätzle (Spz) [4]. Thus, it is proposed as evolutionary conversion in Toll/TLR immune functions [6]. Although the Toll pathway in invertebrates has immunity and developmental activities, mammalian TLRs are simply involved in immunity and have no developmental roles [7].
The structural properties of insect and mammalian TLRs share analogies and differences [8]. Insect Toll and vertebrate TLRs are characterized by leucine-rich repeat (LRR) domains containing cysteine-rich motifs in the extracellular domain, a single-pass transmembrane domain, and a cytoplasmic Toll/Interleukin-1 receptor (TIR) domain [9]. Although the TIR domain has similarities to that of the IL-1 receptor family, unlike the intracellular domain of Tolls, IL-1 receptors consist of an immunoglobulin-like domain [10]. Conversely, mammalian TLRs contain a single cysteine-rich cluster at the C-terminal end of LRRs, except Toll9; other Drosophila Tolls hold cysteine-rich clusters at either, or both, the N-and C-terminal ends of LRRs [11]. Likewise, Drosophila Toll9 was shown to be the closest TLR to vertebrate counterparts, suggesting functional similarities among these receptors in different clades [9].
Similar to its receptor structure, the Toll signaling pathway in insects is divided into three steps: (1) extracellular recognition of pathogenic antigens, including but not limited to Lys-type peptidoglycan (Lys-PGN) of Gram-positive bacterial cell walls, β-glucans of fungi, and virulence factors, by PGN recognition proteins (PGRPs) and glucan-binding proteins (GNBPs); (2) activation of a distinct proteolytic cascade from the developmental activation of Toll signaling, leading to the cleavage and activation of the zymogen Spätzle and binding of a cleaved ligand to the extracellular domain of TLRs; and (3) dimerization of the receptor cytosolic domain, TIR, activation of a phosphorylation cascade of an intracellular signaling cascade, and expression of effector molecules [12,13]. After the activation of PGRP-SA/GNBP1 and GNBP3 by Lys-PGN of Gram-positive bacteria cell walls and β-1,3-glucans of yeast and some fungi, respectively, a cascade of CLIP-domain zymogens mediate signal amplification [12]. In addition to the serine protease cascade, modular serine protease (ModSP) and Grass in turn activate the Spätzle-processing enzyme (SPE) [14,15]. The protease Persephone is likely involved in this proteolytic cascade activation [13]. However, the precise mechanism by which protease directly cleaves and activates SPE is yet to be clarified [16]. The interaction of the mature active C-terminal C-106 domain of Spz leads to the dimerization of intracytoplasmic TIR domains, which in turn, interact with distinct death-domain-containing proteins [2]. Upon the binding of processed TIR to the death-domain-containing adaptor molecules, Myeloid differentiation primary response 88 (MyD88), it is associated with the death domain of Tube [17]. Through a bifunctional death domain, Tube recruits the Ser/Thr kinase Pelle [18]. Subsequently, NF-κB transcription factors are released from their inhibitors through the K48 ubiquitination and degradation of Cactus. Nevertheless, the Cactus kinase is not identified [12]. Freed Rel transcriptional factors, Dorsal-related immunity factor (Dif), and Dorsal are then translocated into the nucleus, bind to κB-response elements, and induce the transactivation of antimicrobial peptides (AMPs), which are the hallmark of insect innate immune responses [12,15,19].
Studies on immune responses of arthropods have focused on dipterans [16]. Moreover, a relatively large insect model should be established for comprehensive and accurate biochemical studies (sufficient hemolymph sampling) by using species belonging to different taxonomic groups [6]. To this extent, Toll signaling pathway and their components in T. molitor were investigated in response to various pathogenic sources, including Grampositive and Gram-negative bacteria and fungi. Seven Toll genes (TmToll-2, -3, -6, -7, -8, -9, and -10) were identified in T. molitor. However, the functional importance of these isoforms, except TmToll-2, is poorly understood [20]. In the present study, we explored the immunological importance of TmToll-2 against microbial infection. Our study revealed that TmTLR2 is important for immune-mediating responses to Gram-negative E. coli ( Figure S1). Our findings might provide a basis for developing possible therapeutic strategies against human pathogens.

Sequence Analysis of TmToll-2
In this study, a Toll-2 homolog from T. molitor (TmToll-2, Accession number: OP566501) was identified by an EST and RNA-seq search by using the T. castaneum protein sequence as a query. A 2510 bp open reading frame (ORF) encoded a protein of 834 amino acids. Phylogenetic analysis based on the full-length amino acid sequences of TmToll-2 and other insect Toll receptors indicated that it clustered with Toll-2 proteins from T. castaneum, Sitophilus oryzae, Manduca sexta, Vanessa cardui, Bicyclus anynana, Galleria mellonella, Neodiprion fabricii, Nilaparvata lugens, Homalodisca vitripennis, TLR4 from Bombyx mori, Toll-9, isoform C from Drosophila melanogaster, and TLR Tollo from Anopheles arabiensis; and Mus musculus TLR2, the isoform X1, was used as the outgroup (Figure 1). Phylogenetic analysis showed that TmToll-2 in the order Coleoptera formed a group with other isoforms of Toll-2 from T. castaneum and S. oryzae.

Developmental and Tissue Expression of TmToll-2
The mRNA expression of TmToll-2 was evaluated through qRT-PCR at different developmental stages ( Figure 2A) and in different tissues of late-instar larvae ( Figure 2B) and adults ( Figure 2C). Our data illustrated the developmental and innate immunity accepts of TLR receptors and revealed that the highest expression pattern of TmToll-2 occurred in 1-day-old adults and the embryonic stage in a descending order. In young larvae and adults of T. molitor, the gene expression of TmToll-2 was the highest in gut tissues but negligible in other tissues (less than 0.1-fold change).

Figure 2.
Developmental stage-and tissue-specific expression patterns of TmToll-2 measured by qRT-PCR. (A) Relative mRNA expression levels of TmToll-2 in eggs, young larvae, late-instar larvae (LL), pre-pupae, 1-to-7-day-old pupae (P1-P7), and 1-to-5-day-old adults (A1-A5) were presented. Expression levels were the highest in the eggs and 1-day-old adults. The mRNA expression decreased at the larval stage and were the lowest at the young larval stage. TmToll-2 tissue expression patterns in late instar larvae (B) and adults (C) were also examined. Total RNA was extracted from different tissues, including the integument (IT), Malpighian tubule (MT), gut (GT), hemocytes, and fat bodies (FB) of late instar larvae and the IT, MT, GT, hemocytes, FB, ovary, and testis (TE) of 5day-old adults. Total RNA was isolated from 20 mealworms, and T. molitor 60S ribosomal protein 27a (TmL27a) primers were used as internal control (N = 3). One-way ANOVA and Tukey's multiple-range test were used for comparisons. Bars with the same letter are not significantly different by Tukey's multiple-range test (p < 0.05).

Temporal Expression of TmToll-2 Post-Systemic Infection
The mRNA expression level in T. molitor young larvae was measured and examined post-systemic infection with E. coli, S. aureus, and C. albicans in the MTs ( Figure 3A), gut ( Figure 3B), hemolymph ( Figure 3C), and fat bodies ( Figure 3D) 3, 6, 9, 12, and 24 h after injection to determine the immunological roles of TmToll-2 in deflecting pathogenic attacks. PBS was used as the control. The results showed that TmToll-2 mRNA expression was upregulated in response to all infectious sources and varied in tissue-and time-dependent manner. The upregulation of the TmToll-2 gene expression was the highest in the hemolymph and MTs in a descending order against C. albicans 9 h post infection (up to 350-and 90-fold, respectively). Moreover, the expression of TmToll-2 was induced by 150and 100-fold against S. aureus and E. coli, respectively, because of the involvement of the Toll signaling pathway in the recognition of Gram-positive and Gram-negative invasion. However, the minor gut expression of TmToll-2 was induced mainly after E. coli infection by approximately 3-fold and S. aureus by approximately 2-fold but not after C. albicans infection. Likewise, TmToll-2 expression was relatively low in fat bodies in response to all microorganisms. Relative mRNA expression levels of TmToll-2 in eggs, young larvae, late-instar larvae (LL), pre-pupae, 1-to-7-day-old pupae (P1-P7), and 1-to-5-day-old adults (A1-A5) were presented. Expression levels were the highest in the eggs and 1-day-old adults. The mRNA expression decreased at the larval stage and were the lowest at the young larval stage. TmToll-2 tissue expression patterns in late instar larvae (B) and adults (C) were also examined. Total RNA was extracted from different tissues, including the integument (IT), Malpighian tubule (MT), gut (GT), hemocytes, and fat bodies (FB) of late instar larvae and the IT, MT, GT, hemocytes, FB, ovary, and testis (TE) of 5-day-old adults. Total RNA was isolated from 20 mealworms, and T. molitor 60S ribosomal protein 27a (TmL27a) primers were used as internal control (N = 3). One-way ANOVA and Tukey's multiple-range test were used for comparisons. Bars with the same letter are not significantly different by Tukey's multiple-range test (p < 0.05).

Temporal Expression of TmToll-2 Post-Systemic Infection
The mRNA expression level in T. molitor young larvae was measured and examined post-systemic infection with E. coli, S. aureus, and C. albicans in the MTs ( Figure 3A), gut ( Figure 3B), hemolymph ( Figure 3C), and fat bodies ( Figure 3D) 3, 6, 9, 12, and 24 h after injection to determine the immunological roles of TmToll-2 in deflecting pathogenic attacks. PBS was used as the control. The results showed that TmToll-2 mRNA expression was upregulated in response to all infectious sources and varied in tissue-and timedependent manner. The upregulation of the TmToll-2 gene expression was the highest in the hemolymph and MTs in a descending order against C. albicans 9 h post infection (up to 350-and 90-fold, respectively). Moreover, the expression of TmToll-2 was induced by 150and 100-fold against S. aureus and E. coli, respectively, because of the involvement of the Toll signaling pathway in the recognition of Gram-positive and Gram-negative invasion. However, the minor gut expression of TmToll-2 was induced mainly after E. coli infection by approximately 3-fold and S. aureus by approximately 2-fold but not after C. albicans infection. Likewise, TmToll-2 expression was relatively low in fat bodies in response to all microorganisms.

Effect of TmToll-2 RNAi on T. molitor Larval Survival
After we observed the TmToll-2 mRNA induction post systemic infection, we further investigated the innate immune responses in T. molitor larvae treated with dsTmToll-2 RNAi. Initially, we checked the survival rates of the infected TmToll-2-silenced larvae compared with those of dsTmVer-treated larvae as the control group. Furthermore, 4 days after dsTmToll-2 RNAi injection, TmToll-2 mRNA levels were decreased by 77% ( Figure  4A), confirming the efficiency of RNAi. Subsequently, T. molitor larvae were monitored during 10 days after E. coli ( Figure 4B), S. aureus ( Figure 4C), and C. albicans ( Figure 4D) injection. We did not report either significant or meaningful mortality rate between S. aureus-and C. albicans-infected dsTmToll-2-and dsTmVer RNAi-treated groups, except E coli-infected larvae. Therefore, a high survivability rate was reported after non-entomopathogen infections.

Effect of TmToll-2 RNAi on T. molitor Larval Survival
After we observed the TmToll-2 mRNA induction post systemic infection, we further investigated the innate immune responses in T. molitor larvae treated with dsTmToll-2 RNAi. Initially, we checked the survival rates of the infected TmToll-2-silenced larvae compared with those of dsTmVer-treated larvae as the control group. Furthermore, 4 days after dsTmToll-2 RNAi injection, TmToll-2 mRNA levels were decreased by 77% ( Figure 4A), confirming the efficiency of RNAi. Subsequently, T. molitor larvae were monitored during 10 days after E. coli ( Figure 4B), S. aureus ( Figure 4C), and C. albicans ( Figure 4D) injection. We did not report either significant or meaningful mortality rate between S. aureusand C. albicans-infected dsTmToll-2-and dsTmVer RNAi-treated groups, except E. coli-infected larvae. Therefore, a high survivability rate was reported after non-entomopathogen infections.
group. Interestingly, dsTmToll-2 RNAi increased the mRNA levels of some AMPs in response to pathogens in the whole body samples ( Figure   , and survival rates were studied for 10 days post-pathogen injection (n = 10 per group). Larval survival rates at 10 days post-microbial injection were 80% after E. coli injection and 100% after S. aureus and C. albicans injection compared with the levels in the dsTmVer-injected control group. Data were reported as averages of three biologically independent replicates. Asterisks indicate significant differences between dsTmToll-2-and dsTmVer RNAi-injected groups. Survival analysis was performed using Kaplan-Meier plots (logrank chi-squared test; * p < 0.05). . TmToll-2-silenced larvae were injected with E. coli (B), S. aureus (C), and C. albicans (D), and survival rates were studied for 10 days post-pathogen injection (n = 10 per group). Larval survival rates at 10 days post-microbial injection were 80% after E. coli injection and 100% after S. aureus and C. albicans injection compared with the levels in the dsTmVer-injected control group. Data were reported as averages of three biologically independent replicates. Asterisks indicate significant differences between dsTmToll-2and dsTmVer RNAi-injected groups. Survival analysis was performed using Kaplan-Meier plots (log-rank chi-squared test; * p < 0.05).

Effect of TmToll-2 Gene Silencing on the Expression of Antimicrobial Peptide and NF-κB Genes
An interesting result of TmToll-2 gene silencing indicated that the absence of this gene did not affect the survivability of T. molitor larvae following Gram-positive and fungal infections. However, the TmToll-2 mRNA expression significantly upregulated in hemolymph and Malpighian tubules in response to bacterial and fungal infections. Therefore, to verify these observations, we checked the hallmark of T. molitor innate immunity and the regulation of 15 AMP genes: TmTenecin-1, -2, -3, and -4 (TmTene1, 2, 3, and 4); TmAttacin-1a, -1b, and -2 (TmAtt1a, 1b, and 2); TmDefensin (TmDef ); TmDefensin-like (TmDef-like); TmColeoptericin-A, -B, and -C (TmColeA, B, and C); TmCecropin-2 (TmCec-2); and TmThaumatin like protein-1 and -2 (TmTLP1 and 2) in TmToll-2-silenced larvae challenged with E. coli, S. aureus, and C. albicans. We also confirmed the TmToll-2 knockdown efficiency. In fat bodies, which are the main immune organ of insects, 13 out of 15 AMP genes were downregulated in response to E. coli, and 14 out of 15 AMP genes were downregulated in response to S. aureus ( Figure 5 A-O). In accordance with previous reports of AMP production in T. molitor [20][21][22], we showed that TmDef, TmDef-like, and TmCec2 are involved in immune responses against Gram-positive, Gram-negative, and fungal infections ( Figure 5E,F). Likewise, the downregulation of TmTLP2 following C. albicans infection in TmToll-2-silenced larvae presented the analogy of previous findings ( Figure 5O). In the gut ( Figure 6A-O), eight AMP genes were downregulated in TmToll-2-silenced larvae after E. coli, S. aureus, and C. albicans infections. However, the main effect of TmToll-2 gene silencing in the gut could be observed after E. coli infection, particularly TmTene2 (5000-fold), TmTene4 (2000-fold), TmColeA (2000-fold), and TmColeC (3500-fold), compared with that in dsTmVer-treated larvae ( Figure 6B Figure 7L). Additionally, the AMP genes were marginally negatively regulated after C. albicans in the TmToll-2-silenced group. Interestingly, dsTmToll-2 RNAi increased the mRNA levels of some AMPs in response to pathogens in the whole body samples ( Figure 8A-O), gut, and Malpighian tubules, precisely the levels of Cecropin, Coleoptericin, Attacin, Thaumatin-like protein, and Tencin families (Figure 6A     In accordance with the same protocol used to evaluate the expression of AMP gene after TmToll-2 knockdown, we further examined the NF-kB pathway genes TmDorX1 TmDorX2, and TmRelish in the whole body, fat bodies, gut, and Malpighian tubules (Fig  ure 9A-D) to determine the exact transcription factors downstream of the main signaling pathways responsible for AMP production. Consistent with previous findings and AMP expression levels, the results showed that the expression level of TmDorx2, downstream of the Toll signaling pathway, decreased following E. coli and S. aureus infection in all th examined tissues, whereas the level of TmRelish, downstream of Imd signaling, was no strongly affected [23][24][25][26]. Additionally, the comparison of the expression pattern o TmDorX1 in different tissues explained the main negative and positive regulation of AMP gene expression; in the whole body, TmDorX1 was positively regulated ( Figure 9A). Con In accordance with the same protocol used to evaluate the expression of AMP genes after TmToll-2 knockdown, we further examined the NF-kB pathway genes TmDorX1, TmDorX2, and TmRelish in the whole body, fat bodies, gut, and Malpighian tubules ( Figure 9A-D) to determine the exact transcription factors downstream of the main signaling pathways responsible for AMP production. Consistent with previous findings and AMP expression levels, the results showed that the expression level of TmDorx2, downstream of the Toll signaling pathway, decreased following E. coli and S. aureus infection in all the examined tissues, whereas the level of TmRelish, downstream of Imd signaling, was not strongly affected [23][24][25][26]. Additionally, the comparison of the expression pattern of TmDorX1 in different tissues explained the main negative and positive regulation of AMP gene expression; in the whole body, TmDorX1 was positively regulated ( Figure 9A). Conversely, in the fat bodies and gut, the NF-kB gene expression was negatively regulated ( Figure 9B,C). In the fat bodies and gut, the NF-κB genes expression was negatively regulated. TmVer dsRNA was assessed as a negative control, and T. molitor ribosomal protein (TmL27a) was used as an internal control. All experiments were performed in triplicate. Asterisks indicate significant differences between dsTmToll-2-and dsTmVer RNAi-treated groups determined using Student's t-test (p < 0.05).

Discussion
The Toll signaling pathway is important for the development and immunity of insects in their life. However, intensive studies have described the roles of TLRs in infectious diseases and innate immunity; furthermore, the current understanding of these receptors and relevant signaling is chiefly limited to mammalian and some invertebrate models such as Drosophila, mosquitoes, moths, beetles, and shrimp. These receptors are evolutionarily conserved among all living organisms from mammals to plants [6]. Homologies between toll receptors in vertebrates and invertebrates can be characterized by TIR and LRR domains [27]. Except for Toll-1 in D. melanogaster [28], the specific functions, including their immunological roles, of other proteins within this family remain elusive.
With economic importance, T. molitor was developed into a distinctive immune study model. We identified seven Toll genes in T. molitor. TmToll-7 roles in the innate immune response following Gram-negative E. coli infection through AMP production was reported [20]. Here, we attempted to reveal further roles in the TLR family by examining TmToll-2 activity in warding off pathogenic invasions via RNAi technology.
Initially, our phylogenic analysis indicated that the Toll-2 cluster in insects is distinguished from mammalian TLRs, suggesting distinct evolutionary events and consequently different co-players and functions [29]. In addition to their roles in dorso-ventral axis formation, TLRs have a dynamic expression pattern throughout the development of Drosophila [30]. Moreover, Toll-2, -6, -7, and -8 in T. castaneum, which forms a group with In the fat bodies and gut, the NF-κB genes expression was negatively regulated. TmVer dsRNA was assessed as a negative control, and T. molitor ribosomal protein (TmL27a) was used as an internal control. All experiments were performed in triplicate. Asterisks indicate significant differences between dsTmToll-2-and dsTmVer RNAi-treated groups determined using Student's t-test (p < 0.05).

Discussion
The Toll signaling pathway is important for the development and immunity of insects in their life. However, intensive studies have described the roles of TLRs in infectious diseases and innate immunity; furthermore, the current understanding of these receptors and relevant signaling is chiefly limited to mammalian and some invertebrate models such as Drosophila, mosquitoes, moths, beetles, and shrimp. These receptors are evolutionarily conserved among all living organisms from mammals to plants [6]. Homologies between toll receptors in vertebrates and invertebrates can be characterized by TIR and LRR domains [27]. Except for Toll-1 in D. melanogaster [28], the specific functions, including their immunological roles, of other proteins within this family remain elusive.
With economic importance, T. molitor was developed into a distinctive immune study model. We identified seven Toll genes in T. molitor. TmToll-7 roles in the innate immune response following Gram-negative E. coli infection through AMP production was reported [20]. Here, we attempted to reveal further roles in the TLR family by examining TmToll-2 activity in warding off pathogenic invasions via RNAi technology.
Initially, our phylogenic analysis indicated that the Toll-2 cluster in insects is distinguished from mammalian TLRs, suggesting distinct evolutionary events and consequently different co-players and functions [29]. In addition to their roles in dorso-ventral axis formation, TLRs have a dynamic expression pattern throughout the development of Drosophila [30]. Moreover, Toll-2, -6, -7, and -8 in T. castaneum, which forms a group with T. molitor, with 62% amino acid identities in our phylogenic analysis, share a common characteristic-encoding more LRRs, and Drosophila long Tolls function in embryonic development [9]. Accordingly, TmToll-2 expression patterns vary in an insect's life; in addition to the embryonic stage, the highest expression in 1-day-old adults highlights the TLR significance in the entire development of insects. The early adult stage expression of TmToll-2 can be attributed to compartment boundary restrictions, which are mediated by Toll-1 and -2 as adhesion molecules in Drosophila [31]. Moreover, the highest TmToll-2 expression in the gut of T. molitor larvae and adults in tissue-specific gene expression experiments is supported by former reports, which show that Imd signaling within an insect's gut is not the solo player in an innate immunity match; instead, a possible crosstalk exists between Imd and Toll signaling pathways [32,33].
Toll signaling in T. molitor can be activated after various microbial challenges, including but not limited to Gram-positive, Gram-negative, and fungal infections [20,25,26]. The DAP-type peptidoglycan (DAP-PGN) of Gram-negative bacteria can be sensed by Tenebrio recognition protein PGRP-SA and induces activation [34]. Here, we reported that S. aureus, C. albicans, and E. coli induce TmToll-2 mRNA expression, and the relevant expression was distinguished in the hemolymph of T. molitor larvae. Previous studies showed that the Toll-dependent wound healing activity is initiated by the influx of extracellular calcium ions in epidermal cells in the wound site of Drosophila, which eventually triggers Spätzle cleavage and activation [35]. Additionally, transglutaminase activity and calcium ions from injury in insects mediates the prophenoloxidase-activating (proPO) system [8]. The same protease cascade leading to the PO activity can mediate Spätzle cleavage and Toll pathway-dependent AMP production in M. sexta [13,36]. Consequently, the extensive induction of TmToll-2 in the hemolymph samples of the pathogen-injected groups could be triggered by PO activity, the injection site of epidermal cells with oxidative stress, and proteolytic cascade activation following microbial sensation by Toll-related PRRs. Because of the constant hemolymph flow into Malpighian tubules [37], the expression pattern of TmToll-2 in this tissue was also relatively high.
TmToll-2-silenced larvae were more susceptible to Gram-negative infection than to the controls. In accordance with the survivability result, AMP expressions showed to be in favor of T. molitor larvae after E. coli in the T. molitor vermilion (TmVer) double-strand RNA (dsTmVer) injected group. The negative regulation of all AMPs in the fat bodies of insects following Gram-negative bacterial infection was consistent with our previous report on TmToll-7, in which AMPs downregulated in the silenced larvae following E. coli infections [20]. Moreover, we observed that the mRNA expression levels of TmTen2, TmTen4, TmColeA, TmColeC, TmCec2, and TmTLP1 were negatively regulated in TmToll-2-silenced individuals. The secretion of cecropin and coleoptericins is dependent on Imd signaling in Drosophila and weevil Sitophilus, respectively [38,39]. Additionally, the gut multifunction in insect metabolism and immunity is extensively described [40]; conversely, Imd and Toll signaling pathways play major roles in insects' innate immunity, and the Toll pathway in Drosophila does not have a role in gut immunity [41]. Hence, reactive oxygen species (ROS) production and Imd signaling function as a homeostasis regulator of the gut microbiota in Drosophila and red palm weevil (RPW) [42,43]. Similar to our former studies in T. molitor and RPW [14], the present study exhibited that TmToll-2 could mediate immunity within the gut by regulating the AMP production, confirming that (i) the Imd signaling pathway mainly functions in immunity within the gut [44][45][46]; (ii) Gram-negative bacteria are recognized by PRRs of Toll signaling [47,48]; (iii) and possible cross-talks existed between Toll and Imd signaling pathways [33,49]. Interestingly, in Drosophila, AMP expression in fat bodies can be regulated through ROS signaling in the gut, and hemocytes serve as a signal-relaying organ between the gut and fat bodies after oral infection [50]. This finding can explain why T. molitor larval fat bodies invest the least energy on the expression of the TmToll-2, because the signaling pathway, its relevant transcription factor, and AMPs as final effectors can be also activated by other organs after the infections. Surprisingly, the positive regulation of the mRNA expression of APMs in the whole body samples suggested that major immune organs, such as fat bodies, gut, and Malpighian tubules, affect the total AMP production in T. molitor larva; other tissues and relative cells, such as the integument and hemocytes, are vital players in the overall immunity of insects [51]. Furthermore, the mRNA production of AMPs in the whole body can explain a high survival ratio post systemic infection by different pathogens. Studies on NF-κB function under pathological conditions revealed that the mRNA levels of several AMPs are regulated by TmRelish (downstream of Imd signaling) and TmDorX2 (downstream of Toll signaling) pathways in the fat body [45,52]. Accordingly, our NF-κB results were consistent with the mRNA expression of AMPs. This result suggested that the negative relationship between TmDorX2 and TmRelish is responsible for the downregulation of AMP genes in dissected tissues. Moreover, in whole body samples, TmDorX1 is positively regulated, indicating the upregulation of AMP genes in the relevant samples.
Overall, our molecular analysis provided insights into the innate immunity of T. molitor. We illustrated that TmToll-2 regulates antimicrobial activities in epithelial tissues such as Malpighian tubules and the gut, in addition to fat bodies. TmToll-2 is activated by Grampositive and Gram-negative bacteria in T. molitor, similar to B. mori but in contrast to some insects such as M. sexta and D. melanogaster [2]. In addition, the activation of AMP genes was nonspecific after E. coli, S. aureus, and C. albicans challenges. Further comprehensive studies of possible ligands and simulators of TLRs and possible cross-talks of this signaling with other pathways should provide a clear perspective of the underlying mechanisms involved in innate immunity.

Insect Rearing and Preparation of Microorganisms
T. molitor larvae were reared under dark conditions at 26 ± 1 • C and 60% ± 5% relative humidity in an environmental chamber established in our laboratory. Larvae were fed with an artificial diet consisting of 1.1 g sorbic acid, 1.1 mL propionic acid, 20 g bean powder, 10 g brewer's yeast powder, and 200 g wheat bran in 4400 mL distilled water. The feed was autoclaved at 121 • C for 15 min. All experiments were conducted with healthy 10th-12th instar larvae.
Gram-negative E. coli (strain K12), Gram-positive S. aureus (strain RN4220), and fungus C. albicans (strain AUMC 13529) were used as representatives of different pathogenic sources to investigate the roles of TmTLR2 in immunological challenges. E. coli and S. aureus were cultured in Luria-Bertani [13] broth, and C. albicans was cultured in Sabouraud's dextrose broth at 37 • C overnight. The microorganisms were harvested, washed twice in 1× phosphate-buffered saline (PBS; 8.0 g NaCl, 0.2 g KCl, 1.42 g Na 2 HPO 4 , 0.24 g KH 2 PO 4 in 1 L distilled water, pH 7.0), and centrifuged at 3500 rpm for 15 min. Subsequently, the samples were suspended in PBS, and concentrations were measured at 600 nm (OD600) through spectrophotometry (Eppendorf, Hamburg, Germany). E. coli and S. aureus were diluted to 1 × 10 6 cells/µL, and C. albicans was diluted to 5 × 10 4 cells/µL for immune challenge studies. The relevant optimization of microbial concentration was adjusted in accordance with our previous studies [20,53,54].
Total RNA was extracted by using a Clear-S total RNA extraction kit (Invirustech Co., Gwangju, Republic of Korea) in accordance with the manufacturer's instructions. Then, 2 mg of total RNA was used as the template to synthesize cDNA by using the Oligo (dT)12-18 primers under the following reaction conditions: 72 • C for 5 min, 42 • C for 1 h, and 94 • C for 5 min. MyGenie96 Thermal Block (Bioneer, Daejeon, Republic of Korea) and AccuPower ® RT PreMix (Bioneer) were used in accordance with the manufacturer's instructions. cDNA was stored at −20 • C until further use.
The relative mRNA expression level of TmToll-2 was investigated via quantitative real-time polymerase chain reaction (qRT-PCR) by using an AccuPower ® 2X Greenstar TM qPCR Master Mix (Bioneer, Daejeon, Republic of Korea) with synthesized cDNAs and specific primers (Table 1) at an initial denaturation of 95 • C for 5 min, followed by 45 cycles at 95 • C for 15 s, and 60 • C for 30 s. T. molitor ribosomal protein (TmL27a) was used as an internal control, and the results were analyzed using the 2 −∆∆Ct method [58]. The results were presented as means ± standard error (SE) of three biological replicates.

RNA Interference Analysis
TmToll-2 gene silencing was performed as reported previously [20,37,59]. Briefly, the primers containing the T7 promoter sequence at their 5 end were designed using SnapDragon-Long dsRNA Design (Table 1) to synthesize the dsRNA of the TmToll-2 gene. The primary PCR for the TmToll-2 gene was performed using an AccuPower PfuPCR PreMix (Bioneer) with cDNA and specific primers of the TmToll-2 gene ( Table 1). The second PCR was conducted with primers tailed with T7 promoter sequences and 100× dilution of the second PCR products. According to the developmental expression pattern of TmToll-2, cDNA was synthesized from the whole bodies of the pupae on day 5 and used as a template under the following cycling conditions: an initial denaturation step at 94 • C for 2 min followed by 35 cycles of denaturation at 94 • C for 30 s, annealing at 53 • C for 30 s, and extension at 72 • C for 30 s, with a final extension step at 72 • C for 5 min. The PCR products were purified using a Clear-STM PCR/Gel DNA fragment purification kit (Invirustech Co., Gwangju, Republic of Korea). The dsRNA was synthesized using an AmpliScribe T7-Flash transcription kit (Epicentre Biotechnologies, Madison, WI, USA) in accordance with the manufacturer's instructions. After synthesis, the dsRNA was purified by precipitation with 5 M ammonium acetate and 80% ethanol and quantified using an Epoch spectrophotometer (BioTek Instruments, Inc., Winooski, VT, USA). Then, 2 µg of synthesized dsTmToll-2 RNAi was injected into 10th-11th instar larvae for gene silencing, and dsTmVer was used as a control.
4.5. Effect of TmToll-2 Gene Silencing on Larval Mortality against Microbial Challenge dsTmToll-2 RNAi (2 µg/µL) was first injected into early-instar larvae (instars 10-12; n = 30) by using disposable needles mounted onto a micro-applicator (Picospritzer III Micro Dispense System; Parker Hannifin, Hollis, NH, USA) to measure the effect of TmToll-2 in the T. molitor mortality following a pathogenic invasion. An equal amount of dsTmVer was injected into the larvae at the same stage as the negative control. The efficiency of TmToll-2 knockdown was evaluated through qRT-PCR, and over 95% knockdown was achieved 4 days after injection. TmToll-2-silenced and dsTmVer-injected larval groups were challenged with E. coli (10 6 cells/µL), S. aureus (10 6 cells/µL), or C. albicans (5 × 10 4 cells/µL) in triplicate experiments. The number of the surviving challenged larvae was monitored for 10 days, and the survival rates of TmToll-2-silenced larvae were compared with those of the control larvae. The relevant analysis was performed using Kaplan-Meier plots [60].

Statistical Analysis
All experiments were carried out in triplicate, and data were subjected to one-way ANOVA. Tukey's multiple range tests were used to evaluate the difference between groups (p < 0.05). The fold change in gene expression levels compared with the internal control (TmL27a) and external control (PBS) levels was calculated using the 2 −∆∆Ct method.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.