Introduction

Innate immunity, which exists in all metazoan organisms, is an evolutionarily conserved system for defending the host against microbial invasion. In Drosophila, the gut epithelium is the first line of protection for the host against microorganismal invasion and proliferation (Hoffmann and Reichhart, 2002). Two types of immune molecules are involved in Drosophila gut defense. First, the production of reactive oxygen species (ROS) and nitric oxide (NO) was demonstrated to kill pathogens in the gut epithelium and to trigger downstream immune responses (Wink et al., 2011). Erwinia carotovora carotovora 15 (Ecc15) infection increases the levels of ROS synthesized by dual oxidase (Duox) in the Drosophila gut. Duox-RNA interference (RNAi) flies showed increased mortality and failed to control Ecc15 proliferation in the gut, suggesting that Duox is the main enzyme inducing ROS during gut infection (Ha et al., 2005b). In Aedes aegypti, midgut epithelial cells generate ROS to control bacterial growth (Oliveira et al., 2011). Excess ROS is toxic to the host and is degraded by immune responsive catalase (IRC) to maintain homeostatic redox balance. IRC-RNAi flies exhibited ROS over-production and increased lethality, indicating that IRC plays an antioxidant role in the host defense system (Ha et al., 2005a). NO is generated by nitric oxide synthase (NOS) enzymes, including NOS1, NOS2, and NOS3. NOS2 is inducible, while the other two are constitutively expressed (Wink et al., 2011). Lipopolysaccharide (LPS) stimulation induced the expression of NOS in Bombyx mori (Imamura et al., 2002).

The second intestinal immune defense is the generation of local antimicrobial peptides (AMPs) via the immune deficiency (IMD) pathway (Tzou et al., 2000). Drosomycin and Diptericin are induced in the gut of Drosophila after Erwinia carotovora infection (Basset et al., 2000). In B. mori, local AMP genes, including Cecropin A1 (CecA1), Gloverin 1 (Glov1), Glov2, Glov3, Glov4, and lysozyme (Lys), are induced by Staphylococcus aureus, whereas the expression of CecA1, Glov3 and Glov4 is sometimes inhibited by Escherichia coli infection (Wu et al., 2010b).

In this study, we demonstrate that intestinal hydrogen peroxide (H2O2) and NO levels are elevated after bacterial infection and that the mRNA transcription levels of ROS-related genes and AMP genes are also up-regulated. These results indicate that ROS and AMP have vital defense roles in the midgut of silkworms.

Materials and methods

Silkworm rearing

Silkworm larvae (Nistari strain) were reared on mulberry leaves at 27 °C, 70% relative humidity, and a 12-h light:12-h dark photoperiod.

Oral infection

Pseudomonas aeruginosa and Bacillus bombysepticus were cultured overnight in Luria-Bertani (LB) medium at 37 °C. The bacterial pellet was collected by centrifugation at 8000g for 15 min and washed three times with 0.85% (8.5 g/L) NaCl. The harvested bacterial cells were suspended in 400 µl 0.85% NaCl to an optical density at 600 nm (OD600 nm) of 40 and used for silkworm oral infection. Fresh mulberry leaves were cut into 1 cm×1 cm pieces and coated with bacterial suspensions. Day 3 fifth instar larvae were starved for 12 h before feeding them bacteria. A group of 20 larvae were used for oral infection. Each larva was fed 20 µl bacteria or 0.85% NaCl as a control. Midguts were collected at different time points (4, 8, 16, and 24 h) after feeding.

Mortality recording and colony forming unit (CFU) assay

A group of 20 larvae were infected as described above to evaluate mortality. The number of surviving larvae was recorded every 24 h. Another nine larvae were infected for a bacterial persistence assay. At 0.5, 12, and 24 h post-infection, larvae were dissected, and the peritrophic membranes and their contents (PMC) were collected. Gut contents from individual larvae were separated from the PMC by centrifugation at 500g for 10 min. The supernatant was diluted 100-fold with fresh LB and incubated on a LB agar plate with ampicillin (100 µg/ml) at 37 °C for 12 h. The numbers of colony forming units (CFUs) were counted. Three larvae were selected for each time point, and the experiment was repeated three times.

H2O2 level measurement in the midgut

After oral infection, five larvae were dissected to collect the PMC at different time points. Gut contents were separated from the PMC by centrifugation at 13 000g for 10 min. The supernatants were transferred to Amicon Ultra 10K filters (Millipore, Billerica, MA, USA) and centrifuged at 13 000g for 5 min. The flow-through samples were used in H2O2 assays using the Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol.

Gene expression analysis using real-time quantitative PCR

Real-time quantitative polymerase chain reaction (qPCR) was used to evaluate the expression levels of ROS-related genes and AMP genes. Total RNA was extracted from the midguts of silkworms after various treatments and purified using the Direct-zol™ RNA MiniPrep Kit (Zymo, Irvine, CA, USA). First strand complementary DNA (cDNA) was synthesized using SuperScript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. B. mori initiation factor 4α (IF4α) was used as an internal control to normalize the expression of target genes (Wu et al., 2010a). All specific primers for qPCR are listed in Table 1. qPCR was performed using a FastStart Essential DNA Green Master mix (Roche, Indianapolis, IN, USA) with the CFX96 Real-Time PCR Detection System (Bio-Rad, Hercules, California, USA). qPCR was performed using an initial denaturation at 95 °C for 10 min, followed by 39 cycles of amplification (95 °C for 10 s, 55 °C for 20 s, and 72 °C for 30 s), and the melting curve analysis was performed from 65 to 95 °C. The relative expression levels of target genes were analyzed using the 2−ΔΔCT method (Livak and Schmittgen, 2001). All experiments were repeated independently three times.

Table 1 Primers for qPCR
Full size table

Statistical analysis

All data are presented as mean±SD. The unpaired Student’s t-test was used to compare expression differences between control and infection conditions. Bonferroni’s correction was used to determine the critical significance level. The log-rank test was used to analyze the survival rate using GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA).

Results

B. bombysepticus infection in the silkworm

When silkworms were infected with P. aeruginosa or B. bombysepticus, death rarely occurred during the first 5 d. Afterwards, the survival rate after B. bombysepticus infection was reduced remarkably, reaching approximately 20% by Day 8, while only 10% mortality was seen by Day 8 after P. aeruginosa infection (Fig. 1). These results suggest that B. bombysepticus is more pathogenic to silkworms.

Fig. 1
figure 1

Survival rate of silkworm larvae after bacterial feeding

NaCl: control; P.a: Pseudomonas aeruginosa; B.b: Bacillus bombyseptieus. Twenty larvae were used for each treatment

Full size image

Bacterial numbers in infected larvae

To investigate the persistence of ingested bacteria in infected larvae, we examined the viability of bacteria in the gut. At 0.5 h post oral infection, compared with the control, P. aeruginosa was more persistent, at 100-fold CFUs higher, than B. bombysepticus, but by 12 h and 24 h, no P. aeruginosa cells were growing on the plates. In contrast, B. bombysepticus persisted and proliferated in the gut with CFUs increasing gradually and reaching a maximum at 24 h (Fig. 2). These findings suggest that the silkworm employs different mechanisms to combat different invading bacteria. Alternatively, B. bombysepticus and P. aeruginosa may have different abilities to strive against silkworm immune responses.

Fig. 2
figure 2

CFU changes after the bacterial feeding

NaCl: control; P.a: Pseudomonas aeruginosa; B.b: Bacillus bombyseptieus. Each value is given as the mean±SD of three replicates. *, **, and *** indicate statistical significance at P<0.05, P<0.01, and P<0.001, respectively. Three larvae were used at each time point, and the experiment was repeated three times

Full size image

H2O2 levels in the gut and expression of genes involved in H2O2 metabolism after bacterial infection

We next measured the H2O2 concentration in the gut after bacterial infection. Both P. aeruginosa and B. bombysepticus infection significantly increased intestinal H2O2 levels at 8 and 16 h (Fig. 3a). The Duox and catalase (CAT) transcription levels were highly up-regulated by P. aeruginosa infection and down-regulated by B. bombysepticus infection at 8 h. At other post-infection times, the expression levels of these two genes were lower than those of the control (Figs. 3b and 3c). These data suggest that H2O2 is an important defense molecule in response to bacterial challenge.

Fig. 3
figure 3

H 2 O 2 concentrations (a) in the gut after bacterial feeding and Duox (b) and CAT (c) expression levels after bacterial feeding

In (b) and (c), shown are the relative expression levels of Duox and CAT compared to IF4α. NaCl: control; P.a: Pseudomonas aeruginosa; B.b: Bacillus bombyseptieus. Each value is given as the mean±SD of three replicates. *, **, and *** indicate statistical significance at P<0.05, P<0.01, and P<0.001, respectively. Three larvae were used at each time point, and the experiment was repeated three times

Full size image

Changes in the level of NO in the gut and the transcription of genes related to NO metabolism after bacterial infection1

We also measured intestinal NO levels after bacterial infection. After P. aeruginosa infection, the NO concentration was lower than that of the control at 8 h, followed by a significant increase at 16 h and then a decrease at 24 h; whereas after B. bombysepticus infection, the NO level was slightly elevated only at 24 h (Fig. 4a). qPCR results showed that the NOS1 gene was down-regulated by both P. aeruginosa and B. bombysepticus infection from 4 to 24 h (Fig. 4b). The NOS2 expression level was up-regulated at 8 h by B. bombysepticus infection and at 16 h by P. aeruginosa infection (Fig. 4c). These results imply that NO may play an important role in host defense and that NOS2 is involved in NO generation.

Fig. 4
figure 4

NO concentrations (a) in the gut content after bacterial feeding and NOS1 (b) and NOS2 (c) expression levels after bacterial feeding

In (b) and (c), shown are the relative expression levels of NOS1 and NOS2 compared to IF4α. NaCl: control; P.a: Pseudomonas aeruginosa; B.b: Bacillus bombyseptieus. Each value is given as the mean±SD of three replicates. *, **, and *** indicate statistical significance at P<0.05, P<0.01, and P<0.001, respectively. Three larvae were used at each time point, and the experiment was repeated three times

Full size image

Expression changes in AMP genes in response to bacterial challenge

We measured the expression levels of six AMP genes (Attacin 2 (Att2), Cecropin B6 (CecB6), Cecropin D (CecD), Glovorin 2 (Glov2), Glovorin 3 (Glov3), and Morricin (Mor)) in the midgut using qPCR. All AMP genes were highly up-regulated from 4 to 24 h post-infection except for Mor (Fig. 5). Glov2 and Glov3 were the two most induced genes. At 4 h after P. aeruginosa and B. bombysepticus infection, the expression of Glov2 mRNA increased approximately 90-fold compared with the control, while the transcription level of Glov3 induced 500- and 650-fold by P. aeruginosa and B. bombysepticus infection, respectively. The transcription level profile of Mor differed from the other AMP genes. It was only significantly up-regulated at 8 and 24 h by P. aeruginosa infection. At other time points, both P. aeruginosa and B. bombysepticus infection caused a reduction in the Mor expression. At 24 h, only Att2 and Glov3 were induced at higher levels (15- and 100-fold, respectively) by P. aeruginosa and B. bombysepticus infection. These findings suggest that Glov2 and Glov3 are the major AMPs used to cope with early stage bacterial invasion.

Fig. 5
figure 5

mRNA expression levels of AMP genes after bacterial feeding

Shown are the relative expression levels of AMP genes in relation to IF4α. Each value is given as the mean±SD of three replicates. *, **, and *** indicate statistical significance at P<0.05, P<0.01, and P<0.001, respectively. Three larvae were used at each time point, and the experiment was repeated three times

Full size image

Discussion

To defend against ingested harmful microorganisms, the insect gut has evolved an effective immune system depending on the local production of AMPs and ROS (Lemaitre and Hoffmann, 2007).

Drosophila showed high lethality under P. aeruginosa infection, which activated both the Toll and IMD pathways (Lau et al., 2003). P. aeruginosa exhibited higher pathogenicity to flies in the hemocoel than in the intestine (Chieda et al., 2005). B. bombysepticus, which produces spores and parasporal crystals, was reported to be a highly pathogenic bacterium in silkworms. B. bombysepticus oral infection could provoke strong host immune responses (Huang et al., 2009). In our study, we rarely found live P. aeruginosa in the gut at 12 h after oral infection (Fig. 2), which is likely caused by the orally invading P. aeruginosa being eliminated by the silkworm intestinal immune system; whereas the proliferation of B. bombysepticus increased gradually from 0.5 to 12 h (Fig. 2). The higher CFU of B. bombysepticus in the gut at 24 h is consistent with the higher silkworm mortality after infection (Fig. 1). Furthermore, our results showed that the expression of Duox and CAT was induced by P. aeruginosa, but not by B. bombysepticus (Figs. 3b and 3c), though both bacteria increased local ROS levels. We found that B. bombysepticus infection induced ROS production only at 8 h (Fig. 3a). Afterwards, ROS levels, Duox and CAT expression were barely increased after 8 h (Figs. 3b and 3c). We speculate that proliferation of B. bombysepticus is able to rapidly overcome the silkworm intestinal immune system and gradually cause epithelial cells damage, and finally block further ROS generation. The high level of ROS at 8 h might not be produced by the gut epithelia, but from the haemolymph instead.

In Drosophila, Attacin, Diptericin, Defensin, and Mechtnikowin genes were induced in the gut after Ecc15 oral infection (Buchon et al., 2009). In our study, among the six AMP genes induced in the midgut by P. aeruginosa and B. bombysepticus infection, Glov2 and Glov3 were significantly up-regulated at 4 h (Fig. 5), while local ROS levels showed no changes between infected and control individuals (Fig. 3a). These results imply that Glov2 and Glov3 may serve as the main intestinal defense molecules during the early infection stage.

Natural gut infection also activates the IMD pathway to induce local AMP gene transcription, which plays complementary roles in combating ROS-resistant microbes (Ryu et al., 2006). In our study, at 24 h, intestinal ROS levels were relatively low after P. aeruginosa and B. bombysepticus infection (Fig. 3a). However, the expression levels of AMP genes were still high. These results indicate that local AMP has more persistent antibacterial activity than ROS.

ROS signaling triggers local NO production by inducing intestinal NOS transcription in Drosophila (Wu et al., 2012). NO is induced and serves as an innate immune signal in response to gram-negative bacteria challenge (Foley and O’Farrell, 2003). NO is generated by inducible NOS2 (Wink et al., 2011). NO is thought to induce AMP gene expression by activating the IMD pathway in bacterially challenged Drosophila (Nappi et al., 2000). In the silkworm, Ceropin D was specifically activated by the IMD pathway after Escherichia coli and Bacillus subtilis infections (Tanaka et al., 2009). In our study, the expression of NOS2 was significantly induced by P. aeruginosa at 16 h, leading to an increase of the NO level in the gut (Figs. 4a and 4c). It is notable that the transcription of CecD was highly up-regulated by P. aeruginosa infection at 16 h (Fig. 5c). It is possible that the elevated NO serves as a signal to induce CecD expression. Given the facts that there was no living P. aeruginosa in the gut after 12 h and the H2O2 level increased earlier than NO after bacterial challenge, we speculate that H2O2 induces NO generation in Bombyx as in Drosophila (Wu et al., 2012), though this hypothesis needs further investigation.

Taken together, we investigated the intestinal immune defenses after oral bacterial infection in the silkworm. Our results show that ROS and AMP play important roles in protecting the host against bacterial infection.

Compliance with ethics guidelines

Lei ZHANG, Yan-wen WANG, and Zhi-qiang LU declare that they have no conflict of interest.

All institutional and national guidelines for the care and use of laboratory animals were followed.