Transcriptomic Analysis of Musca domestica to Reveal Key Genes of the Prophenoloxidase-Activating System

Abstract

The proPO system regulates melanization in arthropods. However, the genes that are involved in the proPO system in housefly Musca domestica remain unclear. Thus, this study analyzed the combined transcriptome obtained from M. domestica larvae, pupae, and adults that were either normal or bacteria-challenged by an Escherichia coli and Staphylococcus aureus mixture. A total of 54,821,138 clean reads (4.93 Gb) were yielded by Illumina sequencing, which were de novo assembled into 89,842 unigenes. Of the 89,842 unigenes, based on a similarity search with known genes in other insects, 24 putative genes related to the proPO system were identified. Eight of the identified genes encoded for peptidoglycan recognition receptors, two encoded for prophenoloxidases, three encoded for prophenoloxidase-activating enzymes, and 11 encoded for serine proteinase inhibitors. The expression levels of these identified genes were investigated by qRT-PCR assay, which were consistent with expected activation process of the proPO system, and their activation functions were confirmed by the measurement of phenoloxidase activity in bacteria-infected larvae after proPO antibody blockage, suggesting these candidate genes might have potentially different roles in the activation of proPO system. Collectively, this study has provided the comprehensive transcriptomic data of an insect and some fundamental basis toward achieving understanding of the activation mechanisms and immune functions of the proPO system in M. domestica.

Keywords: Musca domestica; gene identification; genetics of immunity; prophenoloxidase-activating system; transcriptomic analysis.

Figure 1

Comparison between different M. domestica transcriptomes. (A-D) comparisons of samples, clean reads, contigs, and unigenes of three M. domestica transcriptomes marked for 1, 2, and 3 on the abscissa, in which 1 stands for the housefly transcriptome published in 2012 (Liu et al. 2012), 2 stands for the transcriptome published in 2014 (Tang et al. 2014), and 3 stands for the transcriptome in this study.

Figure 2

Length distribution of contigs, unigenes, and BLAST CDSs in M. domestica transcriptomic library.

Figure 3

COG annotations of putative proteins. All putative proteins were classified functionally into 25 categories marked as A-W and Y-Z based on their COG annotations. Of them, the R group was the largest category and accounted for 37.76%, followed by groups G (20.76%) and K (20.42%). In contrast, those marked as Y (0.097%), W (0.99%), and A (1.35%) belonged to the smallest groups.

Figure 4

GO classification of predicted genes. Three GO categories of genes including biological process, cellular component, and molecular function were presented. The numbers of genes in a given functional group are shown to the right of the y-axis and the percentages are marked to the left of the y-axis.

Figure 5

Phylogenetic analysis of mdPGRP unigenes. The amino acid sequences from 21 mdPGRP unigenes of M. domestica and 14 PGRP genes of D. melanogaster including Dm-PGRP SAb (AHN59598), Dm-PGRP SB1 (AAF49420), Dm-PGRP SB2b (AGB94663), Dm-PGRP SC2 (AAF59051), Dm-PGRP SD (AAF50530), Dm-PGRP LAf (AAF50303), Dm-PGRP LBd (AFH06370), Dm-PGRP LCa (AAF50302), Dm-PGRP LCg (ACZ94669), Dm-PGRP LDc (AAO41277), Dm-PGRP LDd (ACL83248), Dm-PGRP LEa (AAF48519), Dm-PGRP LEb (AFH07408), and Dm-PGRP LF(AAF50301) were used to build the NJ phylogenetic tree by MEGA 5.0 with 1000 bootstraps. The 21 mdPGRP unigenes were clustered into eight groups, with each group containing at least one specific Dm-PGRP gene (marked diamond block).

Figure 6

Phylogenetic analysis of mdproPO unigenes. The amino acid sequences from eight unigenes of mdproPO in M. domestica and 21 proPO genes in other species including M. domestica (Md, 1), Sarcophaga bullata (Sb, 2), Drosophila melanogaster (Dm, 3), Anopheles gambiae (Ag, 2), Anopheles stephensi (As,1), Armigeres subalbatus (Ars, 3), Aedes aegypti (Aa, 1), Manduca sexta (Ms, 2), Bombyx mori (Bm, 2), Tenebrio molitor (Tm, 1), Holotrichia diomphalia (Hd, 2), and Eriocheir sinensis (Es, 1) were used to build the NJ phylogenetic tree by MEGA 5.0 with 1000 bootstraps. Six unigenes (circle) were clustered with Sb-proPO1 (disc) into group 1, and the other two unigenes (triangles) were clustered with Sb-proPO2 (triangle block) into group 2.

Figure 7

Phylogenetic analysis of mdPAP unigenes. The amino acid sequences from 12 mdPAP unigenes in M. domestica and 12 known genes of prophenoloxidase activating factor in other species including Drosophila melanogaster (Dm-easter AAF55170; Dm-MP1 AAF52151.3; Dm-MP2 AAF54143.1), Holotrichia diomphalia (Hd-PPAE I precursor BAA34642.1; Hd-PPAF III BAC15604.1), Tenebrio molitor (Tm-SPE BAG14262.2), Bombyx mori (Bm-BAEE NP_001036844.1; Bm-PPAE NP_001036832.1), Manduca sexta (Ms-PAP1 AAX18636.1; Ms-PAP2 AAL76085.1; Ms-PAP3 AAO74570.1), and Tachypleus tridentatus (Tt-PCE AAA30094.1) were used to build the NJ phylogenetic tree by MEGA 5.0 with 1000 bootstraps. The circle-marked five unigenes and Dm-easter and Dm-MP1 were clustered into group 1. Two unigenes and one Dm-MP2 (diamond) were divided to group 2. The other five unigenes and Tt-PCE (triangle) were clustered into group 3.

Figure 8

The interspecific phylogenetic analyses of mdSerpin unigenes. The amino acid sequences from 22 mdSerpin unigenes in M. domestica and seven Dm-Sps (Dm-Sp1 to Dm-Sp7: FBpp008812, FBpp0079094, FBpp0079171, FBpp0079243, FBpp0080979, FBpp0289586, FBpp0110138) of Drosophila melanogaster and 18 Rm-Sps (Rm-Sp1 to Rm-Sp18: AHC98652.1, AHC98653.1, AHC98654.1, AHC98655.1, AHC98656.1, AHC98657.1, AHC98658.1, AHC98659.1, AHC98660.1, AHC98661.1, AHC98662.1, AHC98663.1, AHC98664.1, AHC98665.1, AHC98666.1, AHC98667.1, AHC98668.1, AHC98669.1) of Rhipicephalus microplus were used to build the NJ phylogenetic tree by MEGA 5.0 with 1000 bootstraps. Eleven representative unigenes (triangle) were clustered into 11 groups together with known genes of D. melanogaster and R. microplus.

Figure 9

Simplified schemes of the activation of proPO system in Musca domestica. The proPO system might be triggered by pattern-recognition proteins such as PGRP that have bound to pathogen-associated molecular patterns like PGN. Then, a proteolytic cascade is activated in which the upstream proteinases are activated by the complex of PGRP and PGN, culminating in the activation of pro-forms of prophenoloxidase-activating enzymes (promdPAPs), which are cleaved into active mdPAPs; some mdPAPs are capable of directly cleaving mdproPOs into active mdPOs and the mdPOs catalyze phenols into quinones that are converted into melanin spontaneously. The activating process may be suppressed by several serine proteinase inhibitors (mdSerpins) that specifically block different steps of the activation cascade.

Figure 10

Expression profiles of candidate genes. The expression profiles of selected genes were detected by qRT-PCR using the third instar larvae at different time intervals (0, 2, 4, 6, 12, 24 hr) after challenge by E. coli (A) or S. aureus (B), in which actin acted as the quantity and quality control to normalize interest gene expression level. The error bars represent the mean ± SD of three repeat amplifications. The asterisks represent significant differences from the control (unpaired t-test, ***P < 0.001, **P < 0.01, *P < 0.05).

Figure 11

Measurement of PO activity. Phenoloxidase (PO) activity of hemolymph samples were assayed with 96-well microplates method using L-DOPA as a substrate. The PO level in hemolymph from normal larvae remained constant, as marked in the control (blank column). Compared with controls, four sample groups marked 1, 2, 3, and 4 (decorative columns) showed the variable increases in PO activity in the hemolymph. The most significant increase in PO activity appeared in sample 3 from larvae challenged by S. aureus and E. coli for 30 min. The lowest increase in PO activity existed in sample 2 from larvae injected with mdproPO 1 antibodies. Importantly, the marked increase in PO activity in bacteria-infected larvae was reverse of the quietly low level after blocking with mdproPO 1 antibodies in sample 4, which was close to the PO level in sample 1 from PBS challenged larvae. The significant variation between control and tested samples was calculated by t-test (***P < 0.001, **P < 0.01, *P < 0.05).