High throughput profiling of the cotton bollworm Helicoverpa armigera immunotranscriptome during the fungal and bacterial infections

Abstract

Background: Innate immunity is essential in defending against invading pathogens in invertebrates. The cotton bollworm, Helicoverpa armigera (Hübner) is one of the most destructive lepidopteran pests, which causes enormous economic losses in agricultural production worldwide. The components of the immune system are largely unknown in this insect. The application of entomopathogens is considered as an alternative to the chemical insecticides for its control. However, few studies have focused on the molecular mechanisms of host-pathogen interactions between pest insects and their pathogens. Here, we investigated the immunotranscriptome of H. armigera larvae and examined gene expression changes after pathogen infections. This study provided insights into the potential immunity-related genes and pathways in H. armigera larvae.

Results: Here, we adopted a high throughput RNA-seq approach to determine the immunotranscriptome of H. armigera larvae injected with buffer, fungal pathogen Beauveria bassiana, or Gram-negative bacterium Enterobacter cloacae. Based on sequence similarity to those homologs known to participate in immune responses in other insects, we identified immunity-related genes encoding pattern recognition receptors, signal modulators, immune effectors, and nearly all members of the Toll, IMD and JAK/STAT pathways. The RNA-seq data indicated that some immunity-related genes were activated in fungus- and bacterium-challenged fat body while others were suppressed in B. bassiana challenged hemocytes, including the putative IMD and JAK-STAT pathway members. Bacterial infection elevated the expression of recognition and modulator genes in the fat body and signal pathway genes in hemocytes. Although fat body and hemocytes both are important organs involved in the immune response, our transcriptome analysis revealed that more immunity-related genes were induced in the fat body than that hemocytes. Furthermore, quantitative real-time PCR analysis confirmed that, consistent with the RNA-seq data, the transcript abundances of putative PGRP-SA1, Serpin1, Toll-14, and Spz2 genes were elevated in fat body upon B. bassiana infection, while the mRNA levels of defensin, moricin1, and gloverin1 were up-regulated in hemocytes.

Conclusions: In this study, a global survey of the host defense against fungal and bacterial infection was performed on the non-model lepidopteran pest species. The comprehensive sequence resource and expression profiles of the immunity-related genes in H. armigera are acquired. This study provided valuable information for future functional investigations as well as development of specific and effective agents to control this pest.

Figure 1

Distinct pathogenicity and transcriptomic change elicited by B. bassiana and E. cloacae. (A) Different phenotypes of the cotton bollworm larvae infected with B. bassiana (Bb) and E. cloacae (Ec). The B. bassiana-infected larvae were covered with white hyphae at 72 h post infection and then became rigid before death. However, E. cloacae-infected larvae turned light red at 12 h post infection and became gradually liquefied before death. (B) Survival curves of H. armigera larvae infected by Bb (red line) and Ec (blue line) in comparison with the control group (black line). Three repeats, p < 0.01. (C) Hierarchical clustering analysis of DETs in fat bodies and hemocytes of H. armigera larvae infected by Bb 48 h post infection and Ec 6 h post infection, respectively. The transcript levels significantly (p < 0.001) increased (>2 fold) and decreased (<0.5 fold) under at least two experimental conditions. The heatmap is divided into five discrete clusters and color coded on the left. Cluster 1 (cyan, induced) and Cluster 2 (magenta, suppressed) in hemocytes from Bb infected larvae; cluster 3 (purple) and 4 (brown) induced in fat body of Bb and Ec infected larvae, respectively; cluster 5 (yellow) genes suppressed in fat body of Bb and Ec infected larvae. FB_Mock, control fat body; FB_Bb, fungus-induced fat body; FB_Ec, bacterium-induced fat body; HC_Mock, control hemocytes; HC_Bb, fungus-induced hemocytes; HC_Ec, bacterium-induced hemocytes. (D) A Venn diagram showing the shared and unique DETs in fat body and hemocytes, induced by the fungal and bacterial infection. The overlapping regions represent genes that are concomitantly regulated in two, three or four samples. The directions of transcript level changes are indicated by upward- and downward-pointing arrows. Blue, green, brown and purple represent Bb-fat body, Ec-fat body, Bb-hemocytes, and Ec-hemocytes, respectively.

Figure 2

Comparative transcriptome analysis and identification of immunity-related genes in H. armigera. (A) Gene ontology (GO) annotation of DETs in the H. armigera transcriptome. Enriched GO analysis (* for p < 0.05) of DETs between B. bassiana (blue) and E. cloacae (brown) infections are performed by pairwise comparison with the corresponding control group of fat body (upper panel) or hemocytes (lower panel). Level 2 GO assignments are made in terms of cellular components, molecular functions, and biological processes. The number of gene transcripts assigned to each GO term is shown on the right y-axis, and its percentage of the total number of transcripts is on the left y-axis. (B) Distribution of KEGG functional groups within up- and down-regulated gene cohorts in fat body and hemocytes from the Bb and Ec-infected larvae. The bar chart corresponds to the matched entries of DETs in their own functional category. (C) Principal component analysis (PCA) analysis of global gene expression in hemocytes and fat body in response to the infections. Six samples were analyzed, including Bb-affected fat body (FB_Bb) and hemocytes (HC_Bb), Ec-affected fat body (FB_Ec) and hemocytes (HC_Ec), control fat body (FB_Mock) and hemocytes (HC_Mock). (D) Distribution of H. armigera immunity-related transcripts in categories of recognition, signaling, regulation, and effectors.

Figure 3

The PG recognition proteins of H. armigera. (A) Schematic representations of the H. armigera PGRP domain structures. Lengths of the amino acid sequences are indicated. (B) Phylogenetic analysis of PGRPs. The amino acid sequences of 9 H. armigera (Ha), 7 T. castaneum (Tc), 13 D. melanogastor (Dm), 11 B. mori (Bm), and 4 A. mellifera (Am) PGRPs are compared. Scale bar, 0.1 substitutions per site. The tree shows that HaPGRP and BmPGRPs form good orthologous groups, except for HaPGRP-LD. Red dots at nodes indicate bootstrap values greater than 800 from 1,000 trials. The putative 1:1 or 1:1:1 orthologs were connected by green lines. HaPGRP-LA, −LB, −LC, and -LD contain a transmembrane domain, while HaPGRP-SA1, −SA2, −SB1, −SB2, and -SD have a signal peptide. HaPGRP-SB1,-SB2, and -SD contain the key residues of the amidase activity that hydrolyzes PGs.

Figure 4

The β-1,3-glucanase related proteins (βGRPs) of H. armigera. (A) Schematic representations of the domain structure of H. armigera βGRPs. Lengths of the amino acid sequences are indicated. (B) Phylogenetic relationships of the βGRPs. Sequences of the βGRP family members from 5 H. armigera (Ha), 3 T. castaneum (Tc), 3 D. melanogaster (Dm), 4 B. mori (Bm), 2 A. mellifera (Am) PGRPs, 4 A. aegypti (Aa), and 2 M. sexta (Ms) are aligned to build the tree. Scale bar, 0.1 substitutions per site. Bacillus circulans (Bc) β-1,3-glucanase is included as an out-group.

Figure 5

The C-type lectins (CTLs) and galectins of H. armigera. (A) Phylogenetic analysis of CTLs. The amino acid sequences of 24 H. armigera (Ha), 10 T. castaneum (Tc), 10 D. melanogaster (Dm), 14 B. mori (Bm), 4 M. sexta (Ms), 18 A. aegypti (Aa), 2 A. gambiae (Ag), 2 Acyrthosiphon pisum (Ap) and 8 A. mellifera (Am) CTLs are examined. Scale bar, 0.1 substitutions per site. (B) Phylogenetic analysis of galectins. The amino acid sequences of 3 Ha, 3 Tc, 7 Dm, 4 Bm, 7 Ag and 2 Am galectins are compared. Scale bar, 0.1 substitutions per site.

Figure 6

The clip-domain serine protease-related proteins (CLIPs) and serpins of H. armigera. (A) Phylogenetic analysis of CLIPs. The amino acid sequences of 12 H. armigera (Ha), 9 D. melanogaster (Dm), 8 B. mori (Bm), 3 M. sexta (Ms), 3 H. diomphalia (Hd), 4 A. aegypti (Aa), and 8 A. gambiae (Ag) CLIPs are compared and divided into four groups (A ~ D) based on sequence similarity. Scale bar, 0.1 substitutions per site. (B) Phylogenetic analysis of serpins. The amino acid sequences of 22 Ha, 11 Tc, 9 Dm, 18 Bm, 6 Ms, and 5 Am serpins are examined. The clade showing expansion compared with B. mori is shaded yellow. Scale bar, 0.1 substitutions per site. The red dots at nodes denote bootstrap values greater than 800 from 1,000 trials.

Figure 7

The spätzles and Toll-like receptors of H. armigera. (A) Phylogenetic analysis of spätzle homologs. The amino acid sequences of 6 H. armigera (Ha), 4 T. castaneum (Tc), 6 D. melanogaster (Dm), 3 B. mori (Bm), 1 M. sexta (Ms), 5 A. gambiae (Ag), and 2 A. mellifera (Am) spätzles are compared. The red dots at nodes denote bootstrap values greater than 800 from 1,000 trials. Scale bar, 0.1 substitutions per site. (B) Phylogenetic analysis of Toll-like receptors from seven insect species. The amino acid sequences of 11 Ha, 9 Tc, 9 Dm, 7 Bm, 2 Aa, 6 Ag and 3 Am Toll-like receptors are examined. Scale bar, 0.1 substitutions per site. (C) Schematic representations of the domain structure of H. armigera Tolls. Lengths of the amino acid sequences are indicated.

Figure 8

Putative H. armigera members for major immune signaling pathways. Illustrated were Toll, IMD, JNK, JAK-STAT, and melanization pathways that were revealed in D. melanogaster and M. sexta. Below names were candidate H. armigera members predicted based on sequence similarity. H. armigera gene names are also followed by corresponding Unigene numbers.

Figure 9

Expression patterns of the H. armigera immunity-related genes. (A) Cluster analysis. Expression profiles of immunity-related genes data are organized into three groups: recognition, signaling, and effectors. Fat bodies and hemocytes are extracted from H. armigera larvae infected by Bb 48 post infection and Ec 6 post infection respectively. Six datasets were included: FB_Mock, FB_Bb, FB_Ec, HC_Mock, HC_Bb, HC_Ec. Gene families and functional pathways (Toll, IMD) are categorized within the group. Gene names are shown on the right side. (B) Quantitative real-time PCR analysis of the H. armigera immunity-related gene expression in hemocytes and fat body of the fifth instar larvae after B. bassiana (48 h) and E. cloacae (6 h) injection. H. armigera ribosomal protein S3 (rps3) was used as an internal standard to normalize the templates. The relative mRNA levels are represented as the mean ± S.D. (n = 3). *, p < 0.05; **, p < 0.01. FB_Mock, control fat body; FB_Bb, fungus-induced fat body; FB_Ec, bacterium-induced fat body; HC_Mock, control hemocytes; HC_Bb, fungus-induced hemocytes; HC_Ec, bacterium-induced hemocytes.