TmDorX2 positively regulates antimicrobial peptides in Tenebrio molitor gut, fat body, and hemocytes in response to bacterial and fungal infection

Dorsal, a member of the nuclear factor-kappa B (NF-κB) family of transcription factors, is a critical downstream component of the Toll pathway that regulates the expression of antimicrobial peptides (AMPs) against pathogen invasion. In this study, the full-length ORF of Dorsal was identified from the RNA-seq database of the mealworm beetle Tenebrio molitor (TmDorX2). The ORF of TmDorX2 was 1,482 bp in length, encoding a polypeptide of 493 amino acid residues. TmDorX2 contains a conserved Rel homology domain (RHD) and an immunoglobulin-like, plexins, and transcription factors (IPT) domain. TmDorX2 mRNA was detected in all developmental stages, with the highest levels observed in 3-day-old adults. TmDorX2 transcripts were highly expressed in the adult Malpighian tubules (MT) and the larval fat body and MT tissues. After challenging the larvae with Staphylococcus aureus and Escherichia coli, the TmDorX2 mRNA levels were upregulated 6 and 9 h post infection in the whole body, fat body, and hemocytes. Upon Candida albicans challenge, the TmDorX2 mRNA expression were found highest at 9 h post-infection in the fat body. In addition, TmDorX2-knockdown larvae exposed to E. coli, S. aureus, or C. albicans challenge showed a significantly increased mortality rate. Furthermore, the expression of 11 AMP genes was downregulated in the gut and fat body of dsTmDorX2-injected larvae upon E. coli challenge. After C. albicans and S. aureus challenge of dsTmDorX2-injected larvae, the expression of 11 and 10 AMPs was downregulated in the gut and fat body, respectively. Intriguingly, the expression of antifungal transcripts TmTenecin-3 and TmThaumatin-like protein-1 and -2 was greatly decreased in TmDorX2-silenced larvae in response to C. albicans challenge, suggesting that TmDorX2 regulates antifungal AMPs in the gut in response to C. albicans infection. The AMP expression profiles in the fat body, hemocytes, gut, and MTs suggest that TmDorX2 might have an important role in promoting the survival of T. molitor larvae against all mentioned pathogens.


Gene expression analysis of TmDorX2 during different developmental stages and in various tissues.
To elucidate the developmental pattern of TmDorX2 mRNA expression, samples were randomly collected from eggs (EG), young larvae (YL; 10 th -12 th instar), late instar larvae (LL; 19 th -20 th instar), pre-pupae (PP), 1 to 7-day-old pupae (P1-P7), and 1-to 5-day-old adults (n = 20 for each stage). To elucidate the tissue-specific pattern of TmDorX2 mRNA expression, larval and adult tissues of T. molitor including the fat body, Malpighian tubules (MTs), gut, integument, hemocytes, ovary, and testis, were dissected. Subsequently, total RNA was extracted from the collected samples following the LogSpin RNA isolation method with minor modifications 42 . Briefly, the tissue samples were homogenized in guanidine thiocyanate based RNA lysis buffer (20 mM EDTA, 20 mM MES buffer, 3 M guanidine thiocyanate, 200 mM sodium chloride, 40 μM phenol red, 0.05% Tween-80, 0.5% glacial acetic acid (pH 5.5), and 1% isoamyl alcohol in 50 ml) using a bead-based homogenizer (Bertin Technologies, France). After incubation at room temperature (approximately 25 °C) for 1 min, the samples were centrifuged at 15,000 rpm for 5 min at 4 °C. Subsequently, 100 μl from the supernatants were diluted in 200 μl of RNA lysis buffer added to 300 μl of 99.9% ethanol. and were centrifuged at 15,000 rpm for 30 s at 4 °C, using silica spin columns (Bioneer, Korea, KA-0133-1). The aqueous phase was discarded, and the genomic DNA was digested using DNase (Promega, USA, M6101) for 15 min at 37 °C. Subsequently, the silica spin columns were washed with 450 μl of 3 M sodium acetate buffer by centrifugation at 15,000 rpm for 30 s at 4 °C. Next, 500 μl of 80% ethanol was added to the spin columns and the samples were centrifuged again. After drying the spin column for 1 min, total RNA was eluted in 30 µl of distilled water for cDNA synthesis and other downstream applications.
Analysis of TmDorX2 mRNA expression after microbial challenge. In order to determine TmDorX2 mRNA expression upon microbial challenge, T. molitor larvae (10 th -12 th instar larvae) were experimentally challenged by injecting 1 µl of E. coli (1 × 10 6 cells/µl), S. aureus (1 × 10 6 cells/µl), and/or C. albicans (5 × 10 4 cells/µl) into separate groups of larvae. Immune tissues, including the fat body, hemocytes, gut, and MTs, were collected from each of the microbe-infected group and from PBS-injected groups acting as a wounding control at 3, 6, 9, 12, and 24 h post-infection. Subsequently, total RNA extraction, cDNA synthesis, and qRT-PCR were carried out as described above. www.nature.com/scientificreports www.nature.com/scientificreports/ TmDorX2 gene silencing. To prepare double-stranded RNA against TmDorX2 (dsTmDorX2), we designed forward and reverse primers containing the T7 promoter sequence at their 5′ ends using the SnapDragon-Long dsRNA design software (https://www.flyrnai.org/cgi-bin/RNAi_find_primers.pl) ( Table 2) (Fig. S1). The 480-bp PCR product was amplified in AccuPower ® Pfu PCR PreMix with the TmDorX2_Temp_Fw and TmDorX2_Temp_ Rv (Table 2) (Fig. S1) at 95 °C for 2 min, followed by 30 cycles of denaturation at 95 °C for 20 s, annealing at 56 °C for 30 s, and extension at 72 °C for 5 min. The procedure was followed using the same PCR conditions, which led to production of a 388-bp PCR product containing the T7 promoter sequence ( Table 2). The synthesized dsTmDorX2 was purified, using an AccuPrep ® PCR Purification Kit (Bioneer, Korea), precipitated with 5 M ammonium acetate, and washed with 70% ethanol. Subsequently, the PCR product was used as a template to synthesize dsTmDorX2 in vitro using an EZ TM T7 High Yield in vitro Transcription Kit (Enzynomics, Korea) as per the manufacturer's instructions. Briefly, 1 μg of the final PCR product was mixed with 4 μl of 5X Transcription Buffer, 2 μl of 10X MgCl 2 , 2 μl of DTT (100 mM), 1 μl of RNase Inhibitor (40 U/μl), 1 μl of rATP (100 mM), 1 μl of rGTP (100 mM), 1 μl of rCTP (100 mM), 1 μl of rUTP (100 mM), and 1 μl of T7 RNA polymerase. Subsequently, the mixture was incubated at 37 °C for 3 h and 25 °C for 1 h. The synthesized dsTmDorX2 was mixed with one volume of 5 M ammonium acetate, incubated on ice for 15 min, and washed three times using 70%, 80% and 99.9% ethanol, respectively. Finally, after drying, the pellet was resuspended in 30 μl distilled water (Sigma, USA, W4502-1L). A 546 bp PCR product of the Enhanced Green Fluorescent Protein (EGFP) gene (derived from the plasmid EGFP-C1) was used as a template to synthesize double-stranded EGFP (dsEGFP) acting as negative control.

Statistical analysis.
All experiments were carried out in triplicate and data were subjected to one-way analysis of variance (ANOVA). In order to evaluate the difference between groups (p < 0.05), the Tukey's multiple range test was performed. The results for the mortality assay were analyzed using the Kaplan-Meier plot (log-rank Chi-square test) in Excel (http://www.real-statistics.com/survival-analysis/kaplan-meier-procedure/ real-statistics-kaplan-meier/).

Results
Gene organization, cDNA analysis, and phylogenetic tree. A local tblastn search using the T. castaneum Dorsal 2 protein sequence (GenBank: EFA02885.1) as the query and the T. molitor RNAseq and EST libraries as the subject was sufficient to identify the Dorsal homologue from T. molitor (Designated as TmDorX2; accession number: MN056348). The genomic organization of TmDorX2 showed that it contains five exons and four introns (Fig. S2). The TmDorX2 full-length ORF consists of 1,482 bp, encoding a polypeptide of 493 amino acid residues (Fig. 1). According to InterProScan analysis and the NCBI conserved domain database, the TmDorX2 amino acid sequence comprises a RHD (P 24 to N 193 ), an IPT (E 199 to P 302 ), and a nuclear localization signal (NLS; P 307 GALKRKREKY 317 ). Four putative NF-κB signature sequences (I 28 to P 46 ; G 215 to F 235 ; F 251 to Y 269 ; H 292 to K 306 ) were found at the N-terminus of the RHD and the N-and C-terminusof the IPT domain. To evaluate the evolutionary position of TmDorX2 using percentage identity and phylogenetic analysis, we retrieved orthologous sequences from 22 insect species. Furthermore, the conserved RHD and IPT domain in TmDorX2 were compared at the amino acid level using ClustalX 2.1 multiple sequence alignment ( Fig. 2A).
Phylogenetic analysis revealed that the Dorsal isoforms from representative insect species clustered under separate insect orders (Fig. 2B). The Dorsal X2 isoforms from T. molitor and T. castaneum were clustered together, as confirmed by bootstrap replications. The Dorsal isoforms from the order Coleoptera were clustered separately. Similarly, the Dorsal isoforms from insect orders Lepidoptera, Hemiptera and Hymenoptera were classified into separate independent clusters. Moreover, species belonging to the order Hymenoptera (including ants and bees) formed two distinct clusters, one formed by ants [PbDorX3 (Pogonomyrmex barbatus), TcDorX3 (Trachymyrmex cornetzi), and MpDorX2 (Monomorium pharaonis)] and another by bees [BtDor (Bombus terrestris), AdDorX5 (Apis dorsata), AfDor (Apis florea), and AmDor (A. mellifera)]. Whereas Drosophila Dorsal isoforms showed relatedness, the Dorsal isoforms belonging to other species in the order Diptera were clustered separately.
Percent identity calculated based on specific domain analysis showed that TmDorX2 has the highest similarity with TcDor and AvDor (64% identity), followed by 59% and 57% identity with TcDor2 and NvDor, respectively. In addition, the maximum and minimum identities of TmDorX2 within the Hymenoptera (56-57%), Diptera (51-54%), and Lepidoptera (46-49%) orders were calculated and are presented in Fig. S3.

Temporal and spatial expression patterns of TmDorX2. qRT-PCR was employed to investigate
TmDorX2 mRNA expression during different developmental stages of the insect relative to the levels of TmL27a as an internal control (Fig. 3A). The TmDorX2 mRNA was expressed during all developmental stages of T. molitor. The mRNA levels were upregulated in the late larval stage, followed by a decline in expression in early pupae. This was followed by an increase in mRNA expression in 2-day old pupae and a consistent decline in late pupal stages. The expression of TmDorX2 was higher during adult stages, with the highest expression observed in 3-day-old adults (Fig. 3A). The TmDorX2 mRNA was detected in all T. molitor larval tissues with the highest expression level observed in the fat body and MTs, followed by that in the integuments and hemocytes, while the lowest expression was observed in the gut (Fig. 3B). Likewise, the transcription of TmDorX2 in 5-old-day adults was significantly higher in MTs, followed by that in the integument and hemocytes. The lowest expression of TmDorX2 was found in reproductive tissues (Fig. 3C).

Induction profile of TmDorX2 upon microbial insult.
To elucidate the participation of TmDorX2 in T. molitor innate immunity upon microbial infection, we evaluated the levels of TmDorX2 mRNA expression by qRT-PCR at different time points (3,6,9,12, and 24 h) in the fat body, hemocytes, gut, MTs, and the whole body of T. molitor larvae (Fig. 4). In Tenebrio larvae injected with Gram-negative bacteria E. coli, the whole-body levels of TmDorX2 expression increased 9 h post-injection (hpi), followed by a gradual decline at 12 hpi and 48 www.nature.com/scientificreports www.nature.com/scientificreports/ hpi (Fig. 4A). Significant increase in whole-body TmDorX2 expression was also observed after S. aureus infection, at all time points tested, when compared with that in the mock control (p < 0.05). In the fat body tissue, the fold-increase in TmDorX2 expression upon infection with E. coli and the fungus C. albicans reached its highest www.nature.com/scientificreports www.nature.com/scientificreports/ level at 9 hpi. S. aureus infection also caused a significant increase in TmDorX2 expression in the fat body when compared with that in the mock-infected larvae (p < 0.05) (Fig. 4B). In hemocytes, E. coli infection led to a dramatic increase in TmDorX2 expression at early stages of infection followed by a large decline at later time points (Fig. 4C). Similar TmDorX2 expression profiles were observed in the gut after E. coli infection. The increase in TmDorX2 expression in the gut was significant when compared with the mock control (p < 0.05) (Fig. 4D). In the MTs, an early increase in the expression of TmDorX2 mRNA was observed after C. albicans infection, which was statistically significant (p < 0.05) (Fig. 4E).
Mortality upon TmDorX2 knockdown and microbial challenge. Double-stranded RNA (dsRNA) against TmDorX2 was synthesized and used to assess the involvement of TmDorX2 in the Toll pathway. Subsequently, the transcription of 14 AMP genes was examined to determine their regulation by the Toll signaling pathway. Evaluation of the levels of TmDorX2 mRNA expression after dsRNA injection in comparison to those in the dsEGFP treated negative control demonstrated a 91% knockdown efficiency for TmDorX2. (Fig. 5A).
To examine the effect of TmDorX2 silencing on larval survival after challenging with E. coli, S. aureus, and C. albicans, we performed experiments in two sets of T. molitor larvae treated with dsTmDorX2 and dsEGFP, respectively, and counted the dead insects for 10 days. Larval mortality was compared between the dsEGFP-and dsTm-DorX2-treated groups and a difference at p < 0.05 was considered significant. The results show that the mortality rate resulting from E. coli (48%, Fig. 5B) and S. aureus (44%, Fig. 5C) infections in the dsTmDorX2-treated groups were significantly different from that in the corresponding dsEGFP groups, respectively. The percent mortality of C. albicans challenged larvae was 55%, and was significantly different than that of the dsEGFP-treated larvae (p < 0.05) (Fig. 5D).

Role of TmDorX2 in T. molitor AMP gene expression.
Given that the stimulation of immune signaling cascades leads to the production of antibacterial or antifungal AMPs to combat the invading pathogens, we investigated the expression of 14 T. molitor AMP genes, namely TmTene1, −2, −3, and −4, TmAtt1a, −1b, and −2, TmDef1 and −2, TmCole1 and −2, TmCec2, and TmTLP1 and −2. We hypothesized that the significant mortality observed in the dsTmDorX2-treated group after microbial challenge was due to the downregulation of AMP gene expression. To confirm our hypothesis, experiments were performed with two groups of T. molitor larvae, injected with 1 μl (1 μg) of TmDorX2 dsRNA and dsEGFP, respectively. After confirming TmDorX2 silencing 2 days after dsRNA injection, larvae were injected with E. coli, S. aureus, C. albicans, or PBS. One-day post-microbial infection, the AMP gene expression profile was studied by qRT-PCR in dissected immune tissues, such as the fat body, hemocytes, gut, and MTs.
In general, after infection with all chosen microbes (E. coli, S. aureus, and C. albicans), the expression of several AMP genes was significantly decreased in TmDorX2-silenced larvae in comparison with dsEGFP-injected cohorts. Our present data demonstrate that TmAtt1a (Figs 7E and 8E) and TmAtt 2 (Figs 7G and 8G) were downregulated in hemocytes and the gut of TmDorX2-silenced larvae. Moreover, TmTene3 (Figs 6C and 8C), TmDef2 (Figs 6I, 8I and 9I) and TmCec2 (Figs 6L and 8L) were highly downregulated in the fat body, gut and MTs of dsTmDorX2 groups. Finally, in the hemocytes and gut, expression of TmCole2 (Figs 7K and 8K) and TmAtt1b (Figs 7F and 8F) in response to E. coli, S. aureus, and C. albicans was considerably decreased after TmDorX2 knockdown.

Discussion
Insects have evolved a robust tolerance and resistance mechanism against pathogenic infections which enable them to adapt to a wide variety of environmental niches (e.g., endoparasitic lifestyle) 44 . The plasticity of the innate immune defense mechanisms are pivotal towards combating microbial infection 45 . Towards elucidating the biochemical basis of innate immunity, such as the mechanism of pathogen recognition and the ensuing signaling cascades, D. melanogaster and T. castaneum have been used as reliable insect models. T. molitor has recently emerged as an excellent host-pathogen interaction model 24 . Compared to D. melanogaster, which is intolerant of high temperatures (25 °C and 37 °C) 46 , T. molitor exhibits thermal tolerance, making the species suitable for studying host defense mechanism against biotic and abiotic stressors 47,48 . In addition, laboratory rearing of T. molitor is relatively easy 49 and its transcriptome, which represents the largest genetic sequence dataset for insects, has been already reported 50 , providing possibilities for carrying out molecular studies. A comparison of the Toll signaling pathway between T. molitor and Drosophila has revealed commonalities and differences in terms of the immune signaling mechanisms. In Drosophila, Lys-type PGNs of Gram-positive bacteria, β-1,3-glucan of fungi, and DAP-type PGNs of Gram-negative bacteria, activate the Toll signaling pathway 51,52 . Unlike Drosophila, the polymeric DAP-type PGN can also be recognized by the T. molitor PGRP-SA/GNBP1, complex leading to the sequential activation of a three-step proteolytic cascade, similar to that activated by Lys-Type PGN (the Imd pathway) 53,54 . Accordingly, in T. molitor, the recognition of bacterial and fungal PGN initiates Toll and Imd signaling pathways, which induce the expression of AMP genes 23 . Many intracellular proteins are present in the Toll www.nature.com/scientificreports www.nature.com/scientificreports/ signaling pathway. In the present study, we have focused on its final component, Dorsal, a transcription factor downstream the Toll pathway that translocates into the nucleus and binds to appropriate motifs in the promoters of specific AMP genes 55 .
Focusing on the T. molitor Toll pathway, we identified a Dorsal homolog using the T. castaneum Dorsal 2 as a query against the T. molitor RNAseq database. Conserved domain analysis of the full-length TmDorX2 ORF revealed RHD and IPT domains, and an NLS at the C-terminus of the IPT domain. All members of the NF-κB family share the structurally conserved RHD 56 . N-terminal sequences of RHD comprise a recognition loop that is responsible for DNA binding; the C-terminal sequences of RHD are required mainly for dimerization and interaction with inhibitor kappa Kinase (IKK) 57 . Previous studies on NF-κB dimerization found that the IPT domain is crucial for homodimerization, and deleting the IPT domains leads to the degradation of NF-κB precursors 58,59 . TmDorX2 is destined to translocate into the nucleus, hence, it contains an arginine (R)/lysine (K)-rich NLS (N-P 307 GALKRKREKY 317 -C) 60 . Sequence alignment of TmDorX2 RHD with that from other insects showed six conserved cysteine residues. The amino acid cysteine is fundamental for forming disulfide bonds, which is responsible for protein folding and stability.
The TmDorX2 mRNA levels were increased at the late larval and 2-day old pupa stages and reached peak values during adult stages, with the highest expression observed in 3-day-old adults. Prior studies on hormonal regulation of the innate immune response showed that juvenile hormone (JH) and ecdysone, which control development and growth in insects, modulate the expression of immune-induced genes in response to pathogen infection 61 . It is possible that, the fluctuations in the mRNA levels of TmDorX2 mRNA during different developmental stages are related to these versatile hormones. www.nature.com/scientificreports www.nature.com/scientificreports/ TmDorX2 was expressed primarily in MTs and the fat body and less in hemocytes and the gut. Earlier studies in Drosophila have established the fat body (equivalent to the mammalian liver) as the foremost immune responsive organ that synthesizes and secretes AMPs into the hemolymph 62,63 . In addition to the fat body epithelial tissues, including the gut epithelium 64 , reproductive tract, trachea epithelial cells, and MTs (nephridia or kidney analogs) 65,66 , play an important role in immune defense. In the coleopteran model, Zophobas morio, the fat body and MTs are versatile tissues that share pivotal functions, such as immunity, detoxification, nitrogen metabolism, and eye pigmentation 67 . MTs are considered independent epithelial immune-responsive sites in insects. Furthermore, earlier studies have shown that genes involved in the Imd pathway are expressed in Drosophila MTs, and they lead to the induction of AMPs in response to microbial insults 65,66 . Moreover, Toll-associated transcripts, such as Toll receptors, Spz, Tube, Pelle, and Cactus have been detected in the MTs of Z. morio larvae 68 . MTs have also been regarded as immune sites that respond to ecdysone in the presence or absence of pathogenic microbes 69 . Furthermore, the mRNA levels of TmCactin, (positive regulator of Cactus degradation and mediator of Dorsal trans-nuclear localization) was found to be higher in MTs of T. molitor 35 . The results are indicative of simultaneous expression of TmCactin andTmDorX2 in MTs in response to bacterial challenges To understand the involvement of Dorsal in T. molitor innate immunity, we examined the mRNA profiles of TmDorX2 upon E. coli, S. aureus, and C. albicans challenges. Our observation of increased TmDorX2 transcript levels in the whole-body, fat body, and hemocytes of the host larvae after infection with E. coli, and S. aureus is consistent with the findings in the Chinese shrimp, F. chinensis showing highly upregulated FcDorsal mRNA levels in response to both Gram-positive and Gram-negative bacteria 16 . In our study, the highest expression levels of TmDorX2 were observed 9 h post-infection in the immune tissues (fat body, hemocytes, and gut). Moreover, the increased TmDorX2 transcript levels in the hemocytes and gut of the E. coli-infected group in comparison with those in S. aureus and C. albicans-challenged groups, suggest that exposure to E. coli (Gram-negative bacteria) accelerates TmDorX2-Toll induction. A previous study on the expression of EsDorsal, in response to lipopolysaccharides (LPS) from E. coli, peptidoglycan (PG) from S. aureus, and zymosan (GLU) from Saccharomyces cerevisae, showed similar results wherein the EsDorsal responses to LPS were higher than those to GLU and PG 19 . As explained before in the case of Drosophila, the Toll signaling pathway senses β-1,3-glucan from fungi and lysine-type PGN from Gram-positive bacteria, whereas the Imd pathway is activated in response to DAP-type PGN from Gram-negative bacteria 51 . However, previous studies in T. molitor have shown that polymeric DAP-type PGN can be recognized by PGPRP-SA and GNBP1 of the Toll pathway 23 . Moreover, It has www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ been recently reported that activation of the Toll pathway in T. molitor during E. coli infection can occur through TmToll-7 31 . Based on these results, it is not surprising that TmDorX2 is highly expressed after E. coli challenge in T. molitor larvae.  www.nature.com/scientificreports www.nature.com/scientificreports/ To investigate the functional role of TmDorX2 in mediating humoral immunity through the expression of AMP-encoding genes in T. molitor, we first monitored the survival rates of young larvae treated with dsTmDorX2 upon E. coli, S. aureus, or C. albicans challenge. An RNAi efficiency of 91% was confirmed at day 2 post dsTm-DorX2 injection. Larvae mortality rate was significantly higher after dsTmDorX2 silencing, reaching 48%, 42%, and 55% after 10 days of exposure to E. coli, S. aureus and C. albicans, respectively. This suggests that TmDorX2 has a positive and conserved role in T. molitor innate immunity against S. aureus and C. albicans infection, which is consistent with its role in the Pacific white shrimp, L. vannamei 17 . The highest mortality rate was observed in C. albicans-infected larvae compared to the E. coli-and S. aureus-challenged groups. It is possible that such high mortality rates of larvae after C. albicans insult are due to the absence of TmDorX2, which leads to lower AMP-encoding gene expression AMPs are evolutionarily conserved effectors with bactericidal and antifungal activities, and are produced when free Dorsal translocates into the nucleus. qRT-PCR data from dsRNA-injected groups followed by pathogen infection were interpreted by describing TmDorX2 as a positive regulator when the expression of AMP genes was suppressed in dsTmDorX2-treated groups compared to dsEGFP-treated groups, and as a negative regulator in the opposite scenario. The expression of 11 AMP genes was highly increased in the fat body and gut of dsEGFP-injected larvae challenged with E. coli, whereas levels of all 11 AMP transcripts were significantly decreased in the dsTmDorX2-treated groups after E. coli insult. Among all 14 AMPs, the TmAttacin family 70 , TmTene2 71 , and TmTene4 23 are well-known anti-Gram-negative AMPs. In the T. molitor gut, the expression levels of TmAtt1a, −1b and −2, and TmTene2 and −4 were dramatically downregulated in the dsTmDorX2-treated group. In the fat body and hemocytes of TmDorX2-silenced larvae, the levels of all mentioned anti-Gram-negative AMPs except for that of TmTene2 were downregulated. The expression levels of AMP genes in the fat body and gut of TmDorX2 knockdown larvae after E. coli insult (11 AMPs) were more strongly downregulated than those the expression of AMP genes in hemocytes and MTs (8 AMPs).
TmDorX2 knockdown decreased the survivability of larvae after C. albicans challenge. The increased susceptibility of the larvae can be explained by the fact that 11 AMP genes were markedly decreased in the gut of TmDorX2 dsRNA-treated T. molitor. More specifically, the expression of antifungal AMPs (TmTene3, TmTLP1, and TmTLP2) was strongly upregulated in response to C. albicans, and this response was greatly decreased upon TmDorX2 knockdown. The requirement of TmTene3 as an antifungal AMP is known 72 , and supports the results www.nature.com/scientificreports www.nature.com/scientificreports/ of our study. Furthermore, the mRNA levels of TmTene1 and TmTene2 observed upon TmDorX2 knockdown this study agree with previously reported results in the gut 33 . In the fat body and hemocytes, upon TmDorX2 knockdown, the expression of 10 and 8 AMP genes decreased after S. aureus injection, respectively. Taken together, our results suggest that the gut is a crucial immune tissue for mediating an innate immune response to C. albicans and E. coli, while the fat body is pivotal for conferring defense against S. aureus and E. coli.
Overall, the expression levels of AMP-encoding genes in TmDorX2 knockdown larvae significantly decreased after bacterial and fungal challenge compared with those in dsEGFP-injected groups. In a previous study, TmCactin knockdown led to the downregulation of 7AMP genes namely TmTene1 and −4, TmDef1 and −2, TmCole1 and −2, and TmAtt1b post-E. coli, -S. aureus, and -C. albicans challenges 35 . Upon infection with all the above mentioned microbes, dsTmDorX2-treated larvae showed significant downregulation of 6, 6, 9, and 2 AMP genes in the fat body, hemocytes, gut and MTs, respectively (Fig. 10). We must add that, similar to the depletion of TmCactin and TmToll-7 genes, TmDef2 is significantly decreased in the fat body, gut, and MTs tissues of dsT-mDorX2-treated larvae 31,35 . In addition, we found that the induction of TmAtt2 was suppressed in the fat body, hemocytes, and gut of TmDorX2-silenced larvae; a similar downregulation has been observed in dsTmToll-7 injected larvae 31 . These findings suggest that TmDef2 and TmAtt2 are induced after Toll pathway stimulation mediated by E. coli, S. aureus, and C. albicans exposure.
Interestingly, we found several AMP genes induced in the dsTmDorX2-treated group compared to the dsEGFP-treated group. These results raise the possibility that TmDorX2 acts as a negative regulator of those AMP genes in different tissues, but also that of cross talk between Toll and another immune signaling pathway, such as the IMD pathway 35 .
Finally, the mortality rate of T. molitor larvae upon C. albicans and S. aureus infection was higher that upon exposure to E. coli suggesting that TmDorX2 is required for mounting an innate immune response against S. aureus and C. albicans in the larval gut followed by a response in the fat body and hemocytes.

Conclusions
The Dorsal homologue identified in T. molitor (TmDorX2) was highly expressed in the fat body and MTs, and less in the hemocytes and gut. Upon challenge with E. coli, S. aureus, and C. albicans, the TmDorX2 mRNA levels were highly upregulated in the gut, fat body, and hemocytes of T. molitor larvae. According to mortality assay results, survival of TmDorX2-silenced larvae was remarkably decreased after S. aureus and C. albicans infection to a greater extent than after E. coli challenge, although the effect was significant in all infected groups. A Loss-of-function study of AMP expression revealed that TmDorX2 knockdown affects the induction of 11 AMP genes against E. coli and C. albicans in the larval gut, whereas it downregulates 10 AMP genes in response to S. aureus in the fat body. In summary, TmDorX2 can be considered as a positive regulator against E. coli, S. aureus, and C. albicans in the fat body, hemocytes, gut, and MTs of young T. molitor larvae. As inferred, injection of the microorganisms upregulated 6, 6, 9, and 2 AMP genes in fat body, hemocytes, gut and MTs of dsTmDorX2 larvae, respectively. TmDorX2 knockdown followed by microbial challenge resulted in high mortality of T. molitor larvae due to downregulation of AMPs, suggesting that TmDorX2 plays a key role against bacterial and fungal infections in immune tissues such as the fat body and gut.