IKKγ/NEMO Is Required to Confer Antimicrobial Innate Immune Responses in the Yellow Mealworm, Tenebrio Molitor

IKKγ/NEMO is the regulatory subunit of the IκB kinase (IKK) complex, which regulates the NF-κB signaling pathway. Within the IKK complex, IKKγ/NEMO is the non-catalytic subunit, whereas IKKα and IKKβ are the structurally related catalytic subunits. In this study, TmIKKγ was screened from the Tenebrio molitor RNA-Seq database and functionally characterized using RNAi screening for its role in regulating T. molitor antimicrobial peptide (AMP) genes after microbial challenges. The TmIKKγ transcript is 1521 bp that putatively encodes a polypeptide of 506 amino acid residues. TmIKKγ contains a NF-κB essential modulator (NEMO) and a leucine zipper domain of coiled coil region 2 (LZCC2). A phylogenetic analysis confirmed its homology to the red flour beetle, Tribolium castaneum IKKγ (TcIKKγ). The expression of TmIKKγ mRNA showed that it might function in diverse tissues of the insect, with a higher expression in the hemocytes and the fat body of the late-instar larvae. TmIKKγ mRNA expression was induced by Escherichia coli, Staphylococcus aureus, and Candida albicans challenges in the whole larvae and in tissues such as the hemocytes, gut and fat body. The knockdown of TmIKKγ mRNA significantly reduced the survival of the larvae after microbial challenges. Furthermore, we investigated the tissue-specific induction patterns of fourteen T. molitor AMP genes in TmIKKγ mRNA-silenced individuals after microbial challenges. In general, the mRNA expression of TmTenecin1, -2, and -4; TmDefensin1 and -2; TmColeoptericin1 and 2; and TmAttacin1a, 1b, and 2 were found to be downregulated in the hemocytes, gut, and fat body tissues in the TmIKKγ-silenced individuals after microbial challenges. Under similar conditions, TmRelish (NF-κB transcription factor) mRNA was also found to be downregulated. Thus, TmIKKγ is an important factor in the antimicrobial innate immune response of T. molitor.


Introduction
Unlike vertebrates, insects rely on their innate defense mechanisms for pathogen surveillance and immunity. The production of antimicrobial peptides (AMPs) is the most important mechanism The yellow mealworm beetle Tenebrio molitor has emerged as a convenient model for studying insect innate immunity in the last decade. Previously, we have described a three-step proteolytic cascade of the extracellular toll signaling pathway that is elicited against Gram-positive bacteria and fungi [22,30]. In the present study, we have discovered the downstream regulatory components of the toll signaling pathway and their function in AMP induction using RNA interference (RNAi) screens [31][32][33]. TmToll-7 RNAi showed the downregulation of five AMP genes in T. molitor when challenged with E. coli [34]. Moreover, the T. molitor IMD, in the IMD signaling pathway, was found to confer resistance against the Gram-negative bacteria E. coli by regulating the expression of nine AMP genes [35]. Furthermore, the silencing of TmRelish, the downstream NF-κB gene of the IMD pathway, showed the pivotal functional regulation in the immune response against Gram-negative and Gram-positive bacteria [36,37]. In T. molitor, regulatory components of the pathway downstream of IMD are yet to be elucidated. In this study, we have functionally characterized T. molitor IKKγ (TmIKKγ) using an RNAi screen by elucidating its role in regulating T. molitor AMP genes after microbial challenges. Moreover, we have studied the mRNA expression of TmRelish (NF-κB factor in the IMD pathway), TmDorsal isoform X1 (TmDorX1), and TmDorX2 (NF-κB factor in the toll pathway) in TmIKKγ knockdown individuals to understand the crosstalk with NF-κB signaling in the T. molitor toll and IMD pathways.

Gene Organization and cDNA Structure of TmIKKγ
The cDNA sequence of TmIKKγ was determined by a local tblastn search against the T. molitor genome and the expressed sequence tag database using the T. castanuem IKKγ (TcIKKγ) protein sequence (EEZ99267.2) as a query. Homology mapping of TmIKKγ against the T. molitor DNA-Seq database (unpublished) revealed a single exon of 1521 bp, beginning with the translation start codon "ATG" and ending with the stop codon "TAA" (Figure S1A,B). The coding region of TmIKKγ translated to a polypeptide of 506 amino acid residues ( Figure 1).

Domain Architecture and Phylogenetics of TmIKKγ
Domain analysis using InterProScan and a conserved domain search in the NCBI database revealed two conserved domains for TmIKKγ: the NF-κB essential modulator (NEMO) domain (between amino acids 106 and 172) and the leucine zipper domain of coiled coil region 2 (LZCC2) of NEMO (between amino acids 373 and 460) (Figure 1). A multiple sequence alignment of the TmIKKγ amino acid sequence with the IKKγ of insect orthologs showed conservation at the level of the NEMO domain and LZ of domain CC2 of NEMO ( Figure S2A). To explore the evolutionary position of TmIKKγ proteins, we constructed a phylogenetic tree using the IKKγ homologs from different insect species. TmIKKγ was clustered with TcIKKγ ( Figure S2B). The full-length amino acid sequences of TmIKKγ and TcIKKγ showed a high degree of conservation with amino acid sequences from the Asian longhorned beetle Anoplophora glabripennis NEMO isoform X2 (AgIKKγ) and the Colorado potato beetle Leptinotarsa decemlineata NEMO isoform X2 (LdIKKγ). The beetle cluster of IKKγ proteins was related with the lepidopteran IKKγ cluster, which included IKKγ amino acid sequences from Bombyx mori, Spodoptera litura, and Plutella xylostella. The IKKγ sequences from dipterans (including Drosophila melanogaster Kenny; DmKenny) and hymenopterans formed separate clusters. This represents a valid evolutionary lineage for TmIKKγ, clustering within the coleopteran IKKγ amino acid sequences.

Domain Architecture and Phylogenetics of TmIKKγ
Domain analysis using InterProScan and a conserved domain search in the NCBI database revealed two conserved domains for TmIKKγ: the NF-κB essential modulator (NEMO) domain (between amino acids 106 and 172) and the leucine zipper domain of coiled coil region 2 (LZCC2) of NEMO (between amino acids 373 and 460) (Figure 1). A multiple sequence alignment of the TmIKKγ amino acid sequence with the IKKγ of insect orthologs showed conservation at the level of the NEMO domain and LZ of domain CC2 of NEMO ( Figure S2A). To explore the evolutionary position of TmIKKγ proteins, we constructed a phylogenetic tree using the IKKγ homologs from different insect species. TmIKKγ was clustered with TcIKKγ ( Figure S2B). The full-length amino acid sequences of TmIKKγ and TcIKKγ showed a high degree of conservation with amino acid sequences from the Asian longhorned beetle Anoplophora glabripennis NEMO isoform X2 (AgIKKγ) and the Colorado potato beetle Leptinotarsa decemlineata NEMO isoform X2 (LdIKKγ). The beetle cluster of IKKγ proteins was related with the lepidopteran IKKγ cluster, which included IKKγ amino acid sequences from Bombyx mori, Spodoptera litura, and Plutella xylostella. The IKKγ sequences from dipterans (including Drosophila melanogaster Kenny; DmKenny) and hymenopterans formed separate clusters. This represents a valid evolutionary lineage for TmIKKγ, clustering within the coleopteran IKKγ amino acid sequences.

Developmental and Tissue-Specific Expression Patterns of TmIKKγ mRNA
We measured the mRNA expression levels of TmIKKγ at five different developmental stages (egg, larva (early larva [YL] and late larva [LL]), prepupa (PP), pupa (days 1-7 [P1-P7]), and adult (days 1-5 [A1-A5])) using qRT-PCR ( Figure 2A). Our results showed that TmIKKγ mRNA expression levels were consistent in all the developmental stages, with higher expressions in the pupal and adult stages. A slight decline in expression was noticed in the late-larval, prepupal, and late-pupal (days 5-7) stages. Further, we examined the TmIKKγ mRNA levels in the larval and adult tissues of T. molitor. The results indicated that TmIKKγ mRNA was expressed in all the larval tissues examined, with higher expressions in the hemocytes and fat body ( Figure 2B). In T. molitor adults (five days old), the expression levels of TmIKKγ mRNA were greater in the Malpighian tubules, integument, and fat body ( Figure 2C).

Developmental and Tissue-Specific Expression Patterns of TmIKKγ mRNA
We measured the mRNA expression levels of TmIKKγ at five different developmental stages (egg, larva (early larva [YL] and late larva [LL]), prepupa (PP), pupa (days 1-7 [P1-P7]), and adult (days 1-5 [A1-A5])) using qRT-PCR ( Figure 2A). Our results showed that TmIKKγ mRNA expression levels were consistent in all the developmental stages, with higher expressions in the pupal and adult stages. A slight decline in expression was noticed in the late-larval, prepupal, and late-pupal (days 5-7) stages. Further, we examined the TmIKKγ mRNA levels in the larval and adult tissues of T. molitor. The results indicated that TmIKKγ mRNA was expressed in all the larval tissues examined, with higher expressions in the hemocytes and fat body ( Figure 2B). In T. molitor adults (five days old), the expression levels of TmIKKγ mRNA were greater in the Malpighian tubules, integument, and fat body ( Figure 2C).

Temporal Expression of TmIKKγ mRNA after Microbial Infection
We investigated the regulation of TmIKKγ mRNA expression after fungal (C. albicans) and bacterial (E. coli and S. aureus) infections by performing a time-course study, conducted over a period of 48 h. TmIKKγ expression was analyzed by qRT-PCR in the whole body, hemocytes, gut, and fat body tissues of T. molitor at intervals of 3, 6, 9, 12, and 24 h post-microbial infection (Figure 3). In the whole body, E. coli was found to be the strongest elicitor of TmIKKγ mRNA at 9 h post-infection ( Figure 3A). Furthermore, E. coli was the strongest elicitor of TmIKKγ mRNA in the hemocytes ( Figure  3B), gut ( Figure 3C), and fat body tissue ( Figure 3D) of T. molitor larvae, in comparison to S. aureus and C. albicans. Maximum induction of the transcript was recorded at 9 h (16-fold) in the fat body tissue, following E. coli infection.

Temporal Expression of TmIKKγ mRNA after Microbial Infection
We investigated the regulation of TmIKKγ mRNA expression after fungal (C. albicans) and bacterial (E. coli and S. aureus) infections by performing a time-course study, conducted over a period of 48 h. TmIKKγ expression was analyzed by qRT-PCR in the whole body, hemocytes, gut, and fat body tissues of T. molitor at intervals of 3, 6, 9, 12, and 24 h post-microbial infection (Figure 3). In the whole body, E. coli was found to be the strongest elicitor of TmIKKγ mRNA at 9 h post-infection ( Figure 3A). Furthermore, E. coli was the strongest elicitor of TmIKKγ mRNA in the hemocytes ( Figure 3B . Asterisk denotes significant differences from the mock control at 95% confidence level (Student's t-test).

TmIKKγ RNAi and Larval Mortality Assay
As TmIKKγ mRNA expression was induced after infection with E. coli, S. aureus, and C. albicans, we hypothesized that TmIKKγ could be a candidate for conferring resistance against bacterial and fungal infections. To test this hypothesis, we investigated the survival of T. molitor larvae after silencing TmIKKγ mRNA, followed by a microbial challenge ( Figure 4). The RNAi efficiency of TmIKKγ mRNA was compared to that treated with the control double-stranded (ds)RNA (ds enhanced green fluorescent protein (EGFP)-treated group). We confirmed an efficient knockdown of TmIKKγ mRNA (about 99%) ( Figure 4A). After confirming the knockdown of TmIKKγ mRNA, we challenged the dsTmIKKγ and dsEGFP-treated larvae with bacteria (E. coli and S. aureus) and fungi (C. albicans). The mortality of T. molitor larvae (dsTmIKKγ and dsEGFP-treated groups) in response to the microbial challenge was monitored for 10 days. We found that the dsTmIKKγ-treated larvae showed increased susceptibility to microbial infections. A clear trend was observed in E. coli-infected dsTmIKKγ-treated larvae, with 50% mortality compared with dsEGFP-treated larvae ( Figure 4B). With the S. aureus infection, the mortality trend was unclear initially, but, at day 6 post-infection, there was a significant increase in mortality in dsTmIKKγ-treated larvae as compared with dsEGFPtreated larvae ( Figure 4C). The mortality observed in TmIKKγ-silenced larvae with a S. aureus infection was considered significant, even though a high mortality rate in the dsEGFP control was perplexing and might be related to the overall health status of the larvae included in the group. Close to 50% mortality was observed in dsTmIKKγ-treated and C. albicans-infected larvae ( Figure 4D). The results suggest the requirement of TmIKKγ mRNA against bacterial and fungal challenges in T. molitor larvae. . Asterisk denotes significant differences from the mock control at 95% confidence level (Student's t-test).

TmIKKγ RNAi and Larval Mortality Assay
As TmIKKγ mRNA expression was induced after infection with E. coli, S. aureus, and C. albicans, we hypothesized that TmIKKγ could be a candidate for conferring resistance against bacterial and fungal infections. To test this hypothesis, we investigated the survival of T. molitor larvae after silencing TmIKKγ mRNA, followed by a microbial challenge ( Figure 4). The RNAi efficiency of TmIKKγ mRNA was compared to that treated with the control double-stranded (ds)RNA (ds enhanced green fluorescent protein (EGFP)-treated group). We confirmed an efficient knockdown of TmIKKγ mRNA (about 99%) ( Figure 4A). After confirming the knockdown of TmIKKγ mRNA, we challenged the dsTmIKKγ and dsEGFP-treated larvae with bacteria (E. coli and S. aureus) and fungi (C. albicans). The mortality of T. molitor larvae (dsTmIKKγ and dsEGFP-treated groups) in response to the microbial challenge was monitored for 10 days. We found that the dsTmIKKγ-treated larvae showed increased susceptibility to microbial infections. A clear trend was observed in E. coli-infected dsTmIKKγ-treated larvae, with 50% mortality compared with dsEGFP-treated larvae ( Figure 4B). With the S. aureus infection, the mortality trend was unclear initially, but, at day 6 post-infection, there was a significant increase in mortality in dsTmIKKγ-treated larvae as compared with dsEGFP-treated larvae ( Figure 4C). The mortality observed in TmIKKγ-silenced larvae with a S. aureus infection was considered significant, even though a high mortality rate in the dsEGFP control was perplexing and might be related to the overall health status of the larvae included in the group. Close to 50% mortality was observed in dsTmIKKγ-treated and C. albicans-infected larvae ( Figure 4D). The results suggest the requirement of TmIKKγ mRNA against bacterial and fungal challenges in T. molitor larvae.

Effects of TmIKKγ RNAi on the Expression of AMPs
As TmIKKγ mRNA knockdown resulted in increased susceptibility of the larvae to bacterial and fungal infections, we wanted to understand if TmIKKγ is one of the important factors regulating AMP production. The transcriptional activation of 14 different AMP genes was studied in dsTmIKKγtreated and microbial-challenged larvae compared to the control (dsEGFP-treated and microbialchallenged larvae). The results suggest that, of the 14 AMPs tested, the expression of ten AMP genes, including TmTene1, 2 and 4; TmDef1 and 2; TmCole1 and 2; and TmAtta1a, 1b, and 2, were significantly downregulated by TmIKKγ RNAi in hemocytes after being challenged with E. coli, S. aureus, and C. albicans ( Figure 5A,B,D-H,J-L). This suggests a positive regulatory effect of TmIKKγ on the ten AMP genes. Similarly, in the gut tissue, the expression of ten AMP genes, including TmTene1, 2, and 4; TmDef1 and 2; TmCole1 and 2; and TmAtta1a, 1b, and 2, significantly decreased in dsTmIKKγ-treated larvae when compared to dsEGFP-treated larvae after the microbial challenge ( Figure 6A,B,D-H,J-L). The same repertoire of AMP genes was also positively regulated in the fat body tissue of dsTmIKKγ-treated larvae after the challenge with E. coli, S. aureus, and C. albicans ( Figure 7A

Effects of TmIKKγ RNAi on the Expression of AMPs
As TmIKKγ mRNA knockdown resulted in increased susceptibility of the larvae to bacterial and fungal infections, we wanted to understand if TmIKKγ is one of the important factors regulating AMP production. The transcriptional activation of 14 different AMP genes was studied in dsTmIKKγ-treated and microbial-challenged larvae compared to the control (dsEGFP-treated and microbial-challenged larvae). The results suggest that, of the 14 AMPs tested, the expression of ten AMP genes, including TmTene1, 2 and 4; TmDef1 and 2; TmCole1 and 2; and TmAtta1a, 1b, and 2, were significantly downregulated by TmIKKγ RNAi in hemocytes after being challenged with E. coli, S. aureus, and C. albicans ( Figure 5A,B,D-H,J-L). This suggests a positive regulatory effect of TmIKKγ on the ten AMP genes. Similarly, in the gut tissue, the expression of ten AMP genes, including TmTene1, 2, and 4; TmDef1 and 2; TmCole1 and 2; and TmAtta1a, 1b, and 2, significantly decreased in dsTmIKKγ-treated larvae when compared to dsEGFP-treated larvae after the microbial challenge ( Figure 6A,B,D-H,J-L). The same repertoire of AMP genes was also positively regulated in the fat body tissue of dsTmIKKγ-treated larvae after the challenge with E. coli, S. aureus, and C. albicans ( Figure 7A  . The experimental samples were divided into three groups (E. coli, S. aureus, or C. albicans-challenged groups) and a wounded control (PBS group). AMP expression was studied 24 h after microbial challenge. dsEGFP was used as a negative control, and TmL27a served as an internal control. The numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.
numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.  . The experimental samples were divided into three groups (E. coli, S. aureus, or C. albicans-challenged groups) and a wounded control (PBS group). AMP expression was studied 24 h after microbial challenge. dsEGFP was used as a negative control, and TmL27a served as an internal control. The numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.
challenge. dsEGFP was used as a negative control, and TmL27a served as an internal control. The numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.  dsEGFP was used as a negative control, and TmL27a served as an internal control. The numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.

Effects of TmIKKγ RNAi on NF-κB Genes
After discovering the regulatory effects of TmIKKγ RNAi on the activation of a specific set of Tenebrio AMP genes, we were interested in studying the expression of Tenebrio NF-κB genes in the toll (TmDorX1 and TmDorX2) and IMD (TmRelish) signaling pathways in dsTmIKKγ-treated and microbially challenged larvae. The results showed that the mRNA levels of TmRelish and TmDorX2 were dramatically decreased by TmIKKγ RNAi in the fat body tissue. The expression of TmRelish transcripts was significantly decreased by TmIKKγ RNAi in the hemocytes and gut. Further, TmDorX2 mRNA was decreased by TmIKKγ RNAi in the gut ( Figure 8A-C). groups) and a wounded control (PBS group). AMP expression was studied 24 h after the microbial challenge. dsEGFP was used as a negative control, and TmL27a served as an internal control. The numbers above the bars represent the AMP transcription levels. All experiments were repeated three times, and similar results were obtained. Asterisks denote significance at 95% confidence levels.

Effects of TmIKKγ RNAi on NF-κB Genes
After discovering the regulatory effects of TmIKKγ RNAi on the activation of a specific set of Tenebrio AMP genes, we were interested in studying the expression of Tenebrio NF-κB genes in the toll (TmDorX1 and TmDorX2) and IMD (TmRelish) signaling pathways in dsTmIKKγ-treated and microbially challenged larvae. The results showed that the mRNA levels of TmRelish and TmDorX2 were dramatically decreased by TmIKKγ RNAi in the fat body tissue. The expression of TmRelish transcripts was significantly decreased by TmIKKγ RNAi in the hemocytes and gut. Further, TmDorX2 mRNA was decreased by TmIKKγ RNAi in the gut ( Figure 8A-C). and TmDorX2 (C). T. molitor larvae injected with PBS acted as the mock control group. dsEGFP was used as a negative control, and TmL27a served as an internal control. Asterisk denotes significant differences from the mock control at 95% confidence levels (Student's t-test).

Discussion
The inhibitor of NF-κB (IκB) interacts with the Rel-homology domain of NF-κB and masks the nuclear localization signal of NF-κB. This prevents the translocation of NF-κB to the nucleus, retaining the inactive protein in the cytoplasm [38,39]. IκB kinases (IKKs) phosphorylate the IκB protein in response to various stimuli, leading to its ubiquitination and proteasomal degradation. This releases NF-κB factor to the nucleus, where it is needed for the transcriptional activation of AMPs [40]. IKKs are a part of the high molecular weight signalosome complex, where IKKα and IKKβ (also known as IKK1/CHUK and IKK2, respectively) are the two IκB kinases, and IKKγ (NEMO/NF-κB essential modulator/IKKAP-1/FIP-3) acts as the non-catalytic regulatory subunit [41]. In fact, IKKs serve as an adaptor between the receptor/associated protein complexes and the catalytic IKKα and IKKβ subunits [42].
In this study, we identified an IKKγ homolog in the yellow mealworm beetle T. molitor (TmIKKγ) and characterized it at the molecular level. The mammalian homologs of IKKγ have been known in D. melanogaster (DmKenny) and Crassostrea gigas (CgNEMO), among invertebrates [43,44]. The structural characteristics of TmIKKγ include the NEMO domain and the LZCC2 of NEMO. Similar to mammals, an IKK binding region, NEMO ubiquitin binding domain, and zinc finger domain have been reported in CgNEMO [38,45].
Ubiquitination of IKKγ/ NEMO at specific lysine residues is also found to be conserved, and site-directed mutagenesis studies have indicated their direct role in NF-κB activation [46]. DmKenny comprises the ubiquitin binding in the ABIN and NEMO domain (UBAN) and a LC3-interacting region (LIR) or Atg8-interacting protein (AIM) motif. The LIR motif is responsible for the interaction between Kenny and Atg8a and the selective degradation of Kenny by autophagy [47,48]. TmIKKγ also displayed the relaxed LIR motif (xLIR) from amino acid positions 27 to 32 (ESFVVL pattern) and from amino acid positions 103 to 108 (SSYEEI pattern). The iLIR server (http://ilir.warwick.ac.uk/) also predicted the conventional LIR motif (WxxL) in TmIKKγ from amino acid positions 5 to 10 (DPFVKV pattern), 174 to 179 (KLFNEL pattern), and 487 to 492 (NNYDSL pattern). This indicates that TmIKKγ could interact with the autophagosomal membrane protein microtubule-associated protein 1'light chain 3 (Atg8/LC3).
The alignment of TmIKKγ with IKKγ from orthologous species showed conservation at the domain level, indicating the presence of phosphorylation and ubiquitination sites, but the details need to be investigated in the future. Moreover, for the IKK complex, post-translational modifications at the level of catalytic subunits IKKα and IKKβ would also be crucial to understanding the positive or negative activation of the NF-κB signaling cascade in T. molitor. For the present study, we screened IKKα and IKKβ homologs from the T. molitor RNA-Seq database using T. castaneum sequences as the query. We were interested in understanding the involvement of TmIKKγ in the regulation of NF-κB genes, which are involved in the toll and IMD signaling pathways in Tenebrio, and the subsequent transcriptional control of AMP genes.
The induction of TmIKKγ mRNA expression in the whole body, hemocytes, gut, and fat body tissues of T. molitor after microbial challenges was an early indicator of the putative involvement of TmIKKγ in innate immunity. Further, RNAi experiments revealed that TmIKKγ is required for larval survival against the Gram-negative bacteria E. coli, the fungus C. albicans, and the Gram-positive bacteria S. aureus. This suggested that impairment in the larval survivability in the absence of TmIKKγ is possibly due to the susceptibility of the larvae to microbial infections. It is known that NF-κB activation downstream of IKKs is significant in mounting a humoral response against a plethora of biotic stimuli, including a diverse group of pathogens and parasites. In the Drosophila IMD pathway, TAK1 kinase complex phosphorylates the IKK complex that eventually phosphorylates Relish (NF-κB) to release the Rel homology domain to the nucleus and leads to the transcriptional activation of AMP genes [49,50].
In the present study, we tried to understand the transcriptional activation of fourteen Tenebrio AMP genes in the IKKγ-silenced T. molitor larval hemocytes, gut, and fat body tissues. In the hemocytes, gut'and fat body of TmIKKγ-silenced larvae the mRNA expression of ten, nine, and ten AMP genes were downregulated, respectively, when compared with dsEGFP injected larvae (negative control). AMP genes, such as TmTene1, TmTene2, TmTene4, TmDef1, TmDef2, TmCole1, TmCole2, TmAtta1a, TmAtta1b, and TmAtta2, were found to be downregulated in the hemocytes, gut, and fat body of TmIKKγ-silenced T. molitor larvae challenged with E. coli, S. aureus, and C. albicans. In a previous study, knockdown of the T. molitor IMD protein (central adaptor protein in the IMD pathway that receives signals through PGRP-LC and PGRP-LE), led to the downregulation of a similar set of AMPs (excepting TmDef1) after challenge with microbes. The suppression of AMP expression was more striking in the TmIMD-silenced larvae challenged with E. coli, suggesting the requirement of the transcript to mount a defense response against the pathogen [35]. Considering the IKK complex as a downstream protein complex in the IMD signaling pathway of T. molitor that transcribes AMP gene expression upon being microbially challenged, it was obvious to observe the regulation of the same set of AMPs in TmIMD and TmIKKγ knockdown models in separate studies.
Further, the increased susceptibility of T. molitor larvae to E. coli infection in a TmToll knockdown model was attributed to the downregulation of TmTene1, TmDef1, TmDef2, TmCole1, and TmAtta2 mRNA expression [34]. This highlights the diversity of the T. molitor innate immune system in recognizing the polymeric meso-diaminopimelic acid-type peptidoglycan of E. coli, signaling through the NF-κB protein sequences (Dif and Relish orthologs) for the transcriptional activation of a specific set of AMPs [51]. We therefore were interested to understand the NF-κB component that regulates the expression of the ten AMP genes in TmIKKγ knockdown T. molitor tissues. Interestingly, the expression of TmRelish was significantly decreased by TmIKKγ RNAi, and the expression of other NF-κB genes, including TmDorX1 and -X2, were slightly decreased, suggesting that these ten AMP genes were mainly regulated by the TmIKKγ-dependent IMD pathway. In contrast, mRNA levels for nine AMP genes and two NF-κB genes, including TmRelish and TmDorX2, were dramatically decreased by TmIKKγ RNAi in the gut, indicating that TmIKKγ activates both the TmRelish-dependent IMD pathway and TmDorX2-dependent toll pathway in the gut. Interestingly, the expression of TmDorX1 was drastically increased by TmIKKγ RNAi, suggesting that TmIKKγ positively and negatively regulates innate immune responses. In D. melanogaster, a synergistic immune response elicited by the toll and IMD pathways has been reported [14]. Another report stated that both the toll and IMD pathways could be activated by Gram-positive bacteria and fungi in D. melanogaster [52]. Further, AMPs have been clustered in three independent groups in T. castaneum: AMPs regulated by the toll pathway alone, by the IMD pathway alone, and by the toll and IMD pathways [15,53]. In the present study, we observed that TmIKKγ could activate both the TmRelish-dependent IMD pathway and TmDorX2-dependent toll pathway in T. molitor.
Given these observations, it is important to identify the α and β isoforms of IKK in T. molitor and to understand whether they work independently or in a complex to maintain the immune homeostasis. We also need to elucidate the putative role of TmIKKγ/NEMO in selective autophagic degradation of the IκB kinase. In the future, it will be interesting to visualize the crosstalk between the toll and IMD signaling pathways with respect to the immune response of T. molitor. Further, a categorization of Tenebrio AMPs regulated by the toll and IMD pathways independently and by the both toll and IMD pathways in response to various groups of microorganisms could be crucial towards understanding the humoral response.

Insect Rearing
T. molitor larvae were reared at 26 ± 1 • C and 60% ± 5% relative humidity in an environmental chamber established in the laboratory under dark conditions. The larvae were fed with an artificial diet (1.1 g of sorbic acid, 1.1 mL of propionic acid, 20 g of bean powder, 10 g of brewer's yeast powder, and 200 g of wheat bran in 4400 mL of distilled water autoclaved at 121 • C for 15 min). Healthy 10th to 12th instar larvae were used for all experiments.

Preparation of Microorganisms
The Gram-negative bacterium E. coli (strain K12), Gram-positive bacterium Staphylococcus aureus (strain RN4220), and the fungus Candida albicans were used for the immune challenge experiments. While E. coli and S. aureus were cultured in Luria-Bertani broth (MB cell, Seoul, Korea) at 37 • C, C. albicans was cultured in Sabouraud dextrose broth (MB cell, Seoul, Korea) at 37 • C. The microorganisms were harvested, washed twice in 1X phosphate-buffered saline (PBS pH 7.0), centrifuged at 1146 × g for 10 min, and resuspended in 1X PBS. Thereafter, concentrations were measured at 600 nm (OD 600 ) using a spectrophotometer (Eppendorf, Hamburg, Germany). Finally, 1 × 10 6 cells/µL of E. coli and S. aureus and 5 × 10 4 cells/µL of C. albicans were used for the immune challenge studies.

Identification and in Silico Analysis of TmIKKγ
TmIKKγ sequence was retrieved from the T. molitor RNA-Seq (unpublished) and expressed sequence tag database. Local tblastn analysis was conducted using Tribolium castanuem IKKγ amino acid sequence (EEZ99267.2) as a query. For the gene structure of TmIKKγ, a local tblastn analysis was performed with TmIKKγ as the query and the T. molitor DNA-Seq database (unpublished) as the subject. Full-length cDNA and deduced amino acid sequences of TmIKKγ were analyzed using the BLASTx and BLASTp algorithms at NCBI (https://blast.ncbi.nlm.nih.gov/Blast.cgi), respectively. The full-length open reading frame (ORF) sequence was formatted using UltraEdit64-bit text-editor (http://www.ultraedit. com). The gene-finding software FGENESH was used for the prediction of the TmIKKγ open reading frame (http://www.softberry.com/berry.phtml?topic=fgenesh&group=programs&subgroup=gfind). The domain architecture of the protein sequences was retrieved using the InterProScan domain analysis program (https://www.ebi.ac.uk/interpro/search/sequence-search). Multiple sequence alignment, percentage identity, and percentage distance of the TmIKKγ amino acid sequence with representative IKKγ from other insects were analyzed by the ClustalX 2.1 program (www.clustal.org) [54]. Phylogenetic tree was constructed based on the amino acid sequence alignments using the neighbor-joining method (bootstrap trial set to 1000). The phylogram was analyzed using Tree Explorer view under the Molecular Evolutionary Genetics Analysis (MEGA) version 7.0 (https://www.megasoftware.net/) program [55].
For the immune induction of TmIKKγ mRNA, healthy T. molitor larvae were intra-abdominally injected with 1-µL suspension containing either 1 × 10 6 cells of E. coli or S. aureus or 5 × 10 4 cells of C. albicans. A similar volume of PBS was administered into a separate group of larvae, which served as the wounded control. Samples (whole body, hemocytes, gut, and fat body tissue) were collected at 3, 6, 9, 12, and 24 h post-injection. Total RNA isolation and cDNA synthesis were done as described before.
The expression of TmIKKγ mRNA was evaluated by performing relative quantitative PCR using AccuPower ® 2X GreenStar TM qPCR Master Mix (Bioneer, Daejeon, Korea), synthesized cDNAs, and specific primers for TmIKKγ ( Table 1). The cycling conditions included an initial denaturation at 95 • C for 5 min, followed by 40 cycles of denaturation at 95 • C for 15 s, and annealing and extension at 60 • C for 30 s. Relative quantitative PCR assays were performed on an AriaMx Real-Time PCR System (Agilent Technologies, Santa Clara, CA, USA), and the results were analyzed using AriaMx Real-Time PCR software (www.agilent.com). Each analysis was measured independently at least three times. The 2 −∆∆Ct method [56] was employed to analyze the TmIKKγ mRNA expression levels. The mRNA expression levels were normalized to T. molitor ribosomal protein L27a (TmL27a), which was used as an internal control. The results represent the mean ± SE of three biological replicates. Table 1. Primers used in this study.

Name
Primer Sequences

TmIKKγ RNAi
For RNAi-based functional characterization, dsRNA fragments of TmIKKγ were synthesized-TmIKKγ DNA fragments were amplified by PCR using gene-specific primers tailed with a T7 promoter sequence ( Table 1). The primers were designed using the SnapDragon software (http://www.flyrnai.org/cgi-bin/RNAi_find_primers.pl) to prevent any cross-silencing effects. The first PCR reaction for TmIKKγ was carried out using AccuPower Pfu-PCR PreMix (Bioneer, Daejeon, Korea) with cDNA and specific primers for TmIKKγ (Table 1). The second PCR reaction was conducted with 100-fold dilution of the first PCR product; the final PCR was performed with primers tailed with the T7 promoter sequence and 100-fold dilution of the second PCR product. The PCR reaction was conducted under the following conditions: initial denaturation at 94 • C for 5 min, followed by 35 cycles of denaturation at 94 • C for 30 s, annealing at 53 • C for 1 min, and extension at 72 • C for 30 s on a MyGenie96 Thermal Block (Bioneer, Daejeon, Korea). The PCR products were purified using the AccuPrep ® PCR Purification Kit (Bioneer, Daejeon, Korea), which were then used to synthesize double-stranded RNA (dsRNA) using the EZ TM T7 High Yield in vitro transcription kit (Enzynomics, Daejeon, Korea). Thereafter, 1 µg of PCR purified product was used as template DNA and mixed with 4 µL of 5× transcription buffer, 2 µL of 10× MgCl 2 , 2 µL of 100-mM DTT, 1 µL of RNase Inhibitor (40 U/µL), 1 µL of 100-mM rATP, 1 µL of 100-mM rGTP, 1 µL of 100-mM rCTP, 1 µL of 100-mM rUTP, and 1 µL of RNA polymerase. Subsequently, the reaction mixture was incubated at 42 • C for 3 h. The dsRNA product was purified using the PCI (phenol:chloroform:isoamyl alcohol) method, precipitated with 5-M ammonium acetate, and washed with 70% and 90% ethanol. The synthesized dsRNA was stored at −20 • C until further use.
For the knockdown of TmIKKγ mRNA, 1 µg of synthesized dsRNA of enhanced green fluorescent protein (EGFP) and TmIKKγ were injected into the T. molitor 10th-12th instar larvae. EGFP dsRNA synthesized from pEGFP-C1 plasmid DNA was used as a negative control for RNAi.

Mortality Assay
To measure mortality, 1 × 10 6 cells/µL of E. coli and S. aureus and 5 × 10 4 cells/µL of C. albicans were injected into T. molitor larvae, which TmIKKγ had silenced. Dead larvae were counted daily up to 10 d post-infection; ten insect larvae were used for each set of the mortality assay, and the experiments were triplicated.

Effects of TmIKKγ RNAi on the Expression of AMP and NF-κB Genes
To further characterize the function of TmIKKγ in the humoral innate immune response, the effect of TmIKKγ silencing by RNAi on the expression levels of fourteen AMP genes in the presence of the microbial challenge were investigated. After treating the 10th to 12th instar T. molitor larvae with TmIKKγ dsRNA, E. coli (1 × 10 6 cells per larva), S. aureus (1 × 10 6 cells per larva), or C. albicans (5 × 10 4 cells per larva) were injected into the larvae, and the samples (from the hemocytes, gut, and fat body) were collected 24 h post-injection. As an injection control, 1× PBS was used. Expression levels of fourteen AMP genes, including TmTenecin 1, 2, 3, and 4 (TmTene1, 2, 3, and 4); TmDefensin 1 and 2 (TmDef1 and 2); TmColeoptericin 1 and 2 (TmCole1 and 2); TmAttacin 1a, 1b, and 2 (TmAtta1a, 1b, and 2); TmCecropin 2 (TmCec2); and TmThaumatin-like protein 1 and 2 (TmTLP1 and 2) were measured by qRT-PCR with AMP gene-specific primers (Table 1). In addition, the expression of Tenebrio NF-κB genes in the toll (TmDorX1 and TmDorX2) and IMD (TmRelish) signaling pathways were examined by qRT-PCR with the NF-κB-specific primers listed in Table 1. Relative quantitative PCR was performed, as mentioned above, with AMP-specific primers.

Statistical Analysis
All experiments were carried out in triplicate, and the data were subjected to one-way analysis of variance. Tukey's multiple range test was used to evaluate the differences between groups (p < 0.05).

Conclusions
The study identified, for the first time, the IKKγ isoform in the beetle T. molitor. TmIKKγ transcript expression was induced in the whole larva and other tissues of T. molitor after E. coli, S. aureus, and C. albicans challenges. The RNAi experiments suggested that TmIKKγ might be required for larval survival against microbial challenges by regulating AMPs through the innate immune signaling cascade mechanisms. The NF-κB factors in the IMD pathway (Relish) and the toll pathway (Dorsal isoform 2) were involved, as their transcripts were downregulated in TmIKKγ-silenced larval tissues after the microbial challenge ( Figure 9). This study advances our understanding of the beetles' immunity in response to pathogenic challenges.

Conflicts of Interest:
The authors declare no conflicts of interest.