Evolutionary divergence of core and post-translational circadian clock genes in the pitcher-plant mosquito, Wyeomyia smithii

Abstract

Background: Internal circadian (circa, about; dies, day) clocks enable organisms to maintain adaptive timing of their daily behavioral activities and physiological functions. Eukaryotic clocks consist of core transcription-translation feedback loops that generate a cycle and post-translational modifiers that maintain that cycle at about 24 h. We use the pitcher-plant mosquito, Wyeomyia smithii (subfamily Culicini, tribe Sabethini), to test whether evolutionary divergence of the circadian clock genes in this species, relative to other insects, has involved primarily genes in the core feedback loops or the post-translational modifiers. Heretofore, there is no reference transcriptome or genome sequence for any mosquito in the tribe Sabethini, which includes over 375 mainly circumtropical species.

Methods: We sequenced, assembled and annotated the transcriptome of W. smithii containing nearly 95 % of conserved single-copy orthologs in animal genomes. We used the translated contigs and singletons to determine the average rates of circadian clock-gene divergence in W. smithii relative to three other mosquito genera, to Drosophila, to the butterfly, Danaus, and to the wasp, Nasonia.

Results: Over 1.08 million cDNA sequence reads were obtained consisting of 432.5 million nucleotides. Their assembly produced 25,904 contigs and 54,418 singletons of which 62 % and 28 % are annotated as protein-coding genes, respectively, sharing homology with other animal proteomes.

Discussion: The W. smithii transcriptome includes all nine circadian transcription-translation feedback-loop genes and all eight post-translational modifier genes we sought to identify (Fig. 1). After aligning translated W. smithii contigs and singletons from this transcriptome with other insects, we determined that there was no significant difference in the average divergence of W. smithii from the six other taxa between the core feedback-loop genes and post-translational modifiers.

Conclusions: The characterized transcriptome is sufficiently complete and of sufficient quality to have uncovered all of the insect circadian clock genes we sought to identify (Fig. 1). Relative divergence does not differ between core feedback-loop genes and post-translational modifiers of those genes in a Sabethine species (W. smithii) that has experienced a continual northward dispersal into temperate regions of progressively longer summer day lengths as compared with six other insect taxa. An associated microarray platform derived from this work will enable the investigation of functional genomics of circadian rhythmicity, photoperiodic time measurement, and diapause along a photic and seasonal geographic gradient.

Fig. 1

Functional clockworks of the genes listed in Table 2. Pink: TTFL genes, the core transcription-translation feedback loop consists of positive-acting CLK and CYC and negative-acting CRY2, PER, and TIM; their cycling is affected by “stabilizing” loops involving CWO, KAYα, VRI, and PDP1. Blue: PTM genes, the duration of the circadian cycle is then altered by a number of post-translational modifiers, mainly kinases and phosphatases. Yellow: Entrainment of the circadian clock by external day and night is achieved via the blue-light receptor CRY1. Clear dashed boxes: phosphorylation or ubiquitination leading to ultimate protein degradation. Solid arrows: enhancing transcription or PP2A-B’ reversing phosphorylation of PER. Dashed lines: inhibiting transcription or promoting phosphorylation. Upper case Roman, proteins; lower case Italic, transcripts promoted by CLK and CYC. Solid black circles: phosphate groups (compiled from [17, 25, 26, 30, 123, 125, 127])

Fig. 2

Flow diagram of assigning contigs or singletons to specific circadian clock genes. The functional circadian clock gene was identified in Drosophila melanogaster through Flybase. The Drosophila melanogaster protein sequence was blasted against OrthoDB7 using the most recent common ancestor of all seven species as the search node. The orthologous genes were then taken from the resulting OrthoDB group, with the ortholog of A. aegypti, W. smithii’s most closely related species, and used in a local BLAST against the contigs and singletons from the W. smithii transcriptome. If the lowest E-value from that BLAST identified a single contig or singleton, that contig or singleton was assigned to the respective D. melanogaster gene function in the OrthoDB group. If the lowest E-value from the BLAST identified a multi-gene family, maximum likelihood trees were used to identify the orthologs of various genes in that family (Figs. 3 and 4)

Fig. 3

Assigning W. smithii orthologs to cry1, cry2, and phr6-4 (6–4 photolyase). The maximum likelihood tree identified a single W. smithii contig (bold) within each of the three monophyletic clades in the tree. Gene number abbreviations: AE, Aedes aegypti; AG, Anopheles gambiae; CP, Culex pipiens; DP, Danaus plexippus; FB, Drosophila melanogaster; NV, Nasonia vitripennis; WSc, W. smithii contigs

Fig. 4

Assigning W. smithii orthologs of (a) PP2A-B’ and widerborst (wdb) and (b) Casein kinase 1α (Ck1α) and doubltime (dbt). In a, wdb emerges as a clade within PP2A-B’ and the W. smithii Contig WSc04554 was assigned to PP2A-B’. In b, W. smithii Contigs WSc08154 and WSc08862 (bold) were assigned to Ck1α and dbt, respectively. Gene number abbreviations as in Fig. 3

Fig. 5

Rates of amino acid divergence in circadian clock genes of Wyeomyia smithii relative to other insects (Table 4). a Relative rates (±2SE) of divergence in the core transcription-translation feedback loop (TTFL) and of post-transcriptional modifiers (PTM). b relationship between relative rates of amino acid divergence and the number of nucleotides in the contigs or singletons upon which the rates were based. TTFL (red) and the PTM (blue). c deviations from regression (residuals) in 5b. The residuals essentially factor out any differences in relative rates due to the number of nucleotides upon which amino acid divergence was based

Fig. 6

Phylogenetic relationships of insects used in this study. The nodes indicate approximate time since the most recent common ancestor of a given branch. Orders and families (top) based on [128]; genera within the family Culicidae (bottom) based on [129]