Sequencing and de novo assembly of the western tarnished plant bug (Lygus hesperus) transcriptome

Abstract

Background: Mirid plant bugs (Hemiptera: Miridae) are economically important insect pests of many crops worldwide. The western tarnished plant bug Lygus hesperus Knight is a pest of cotton, alfalfa, fruit and vegetable crops, and potentially of several emerging biofuel and natural product feedstocks in the western US. However, little is known about the underlying molecular genetics, biochemistry, or physiology of L. hesperus, including their ability to survive extreme environmental conditions.

Methodology/principal findings: We used 454 pyrosequencing of a normalized adult cDNA library and de novo assembly to obtain an adult L. hesperus transcriptome consisting of 1,429,818 transcriptomic reads representing 36,131 transcript isoforms (isotigs) that correspond to 19,742 genes. A search of the transcriptome against deposited L. hesperus protein sequences revealed that 86 out of 87 were represented. Comparison with the non-redundant database indicated that 54% of the transcriptome exhibited similarity (e-value ≤ 1(-5)) with known proteins. In addition, Gene Ontology (GO) terms, Kyoto Encyclopedia of Genes and Genomes (KEGG) annotations, and potential Pfam domains were assigned to each transcript isoform. To gain insight into the molecular basis of the L. hesperus thermal stress response we used transcriptomic sequences to identify 52 potential heat shock protein (Hsp) homologs. A subset of these transcripts was sequence verified and their expression response to thermal stress monitored by semi-quantitative PCR. Potential homologs of Hsp70, Hsp40, and 2 small Hsps were found to be upregulated in the heat-challenged adults, suggesting a role in thermotolerance.

Conclusions/significance: The L. hesperus transcriptome advances the underlying molecular understanding of this arthropod pest by significantly increasing the number of known genes, and provides the basis for further exploration and understanding of the fundamental mechanisms of abiotic stress responses.

Figure 1. Summary of mixed sex adult L. hesperus transcriptomic sequences.

(A) Length distribution of contig sequences. (B) Length distribution of isotig sequences. (C) Number of contigs used in the assembly of individual isotigs. (D) Number of isotigs used in the assembly of individual isogroups.

Figure 2. Distribution of BLASTx homology search e-values.

BLAST analysis against the non-redundant database was performed with assembled L. hesperus isotig sequences and an e-value cutoff of 1e ⁻⁵.

Figure 3. Species distribution of the top L. hesperus isotig sequence BLASTx hits.

BLAST analysis against the non-redundant database was performed with an e-value cutoff of 1e ⁻⁵.

Figure 4. Comparative summary of L. hesperus isotig sequences with protein sequences from three insect species.

Amino acid sequence comparisons were performed using BLASTx with predicted protein sequences from the holometabolous insect, Drosophila melanogaster (Diptera), and two representatives of the hemipteran assemblage, Acyrthosiphon pisum (pea aphid) and Pediculus humanus humanus (human body louse).

Figure 5. Classification of L. hesperus isotig sequences based on predicted Gene Ontology

(GO) terms. (A) Biological Process, (B) Cellular Components, and (C) Molecular Function. GO terms were determined using Blast2GO , with an e-value cutoff of 1e ⁻⁵, a 10% initial filter, and sorted based on level 2 classifications.

Figure 6. Distribution of L. hesperus isotig sequences among KEGG

(Kyoto Encyclopedia of Genes and Genomes) pathways. The top 15 most highly represented pathways are shown. Analysis was performed using Blast2GO and the KEGG database , .

Figure 7. Distribution of the top 25 Pfam domains identified in translated L. hesperus sequences.

Analysis was performed using HMMER3 with Pfam database A.

Figure 8. Phylogenetic relationships between predicted L. hesperus heat shock protein

(Hsps) sequences and Hsps from various insect species. Sequences were aligned using MAFFT with default settings. The phylogenetic tree was constructed using the maximum parsimony method implemented in MEGA 5 with default settings. Percentage bootstrap support values above 50%, based on 1000 replicates, are shown. Analyses were performed using only sequences predicted to encode complete ORFs. Non-L. hesperus sequences are indicted by accession numbers with the respective genus and species shown in parentheses. L. hesperus sequences are indicated by the corresponding isogroup identifier. Sequence-confirmed L. hesperus Hsps are shown in bold with the respective accession number in parentheses. Sequences clustered into the six major Hsp protein families (brackets). Asterisk (*) indicates isogroups generated from multiple identical isotig sequences: isogroup00528 = 9 isotigs, isogroup00422 = 9 isotigs, isogroup01416 = 2 isotigs, isogroup03066 = 2 isotigs, isogroup00127 = 12 isotigs, isogroup00701 = 5 isotigs, isogroup01249 = 4 isotigs, and isogroup02441 = 2 isotigs.

Figure 9. Semi-quantitative expression analyses of eight L. hesperus Hsps in response to thermal stress.

PCR was performed using cDNA prepared from 6-day old adult L. hesperus females exposed for 6 hr to either normal conditions (25°C) or thermal stress conditions (39°C). Actin was used as an amplification control. Products were analyzed on 1.5% agarose gels and stained with SYBR Safe. For clarity, the negative images of the gels are shown. Amplification data are representative of three biological replicates.