SEARCH

SEARCH BY CITATION

Keywords:

  • Aphids;
  • epigenetic mechanisms;
  • neuroendocrine control;
  • microRNAs;
  • photoperiodism;
  • transcriptomic analysis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Acknowledgements
  5. References

Aphids are major crop pests and show a high level of phenotypic plasticity. They display a seasonal, photoperiodically-controlled polyphenism during their life cycle. In spring and summer, they reproduce efficiently by parthenogenesis. At the end of summer, parthenogenetic individuals detect the transition from short nights to long nights, which initiates the production of males and oviparous females within their offspring. These are the morphs associated with the autumn season. Deciphering the physiological and molecular events associated with this switch in reproductive mode in response to photoperiodic conditions is thus of key interest for understanding and explaining the remarkable capacity of aphids to adapt to fluctuations in their environment. The present review aims to compile earlier physiological studies, focussing on the neuroendocrine control of seasonal photoperiodism, as well as a series of large-scale transcriptomic approaches made possible by the recent development of genomic resources for the model aphid species: the pea aphid Acyrthosiphon pisum. These analyses identify genetic programmes putatively involved in the control of the initial steps of detection and transduction of the photoperiodic signal, as well as in the regulation of the switch between asexual and sexual oogenesis within embryonic ovaries. The contribution of small RNAs pathways (and especially microRNAs) in the post-transcriptional control of gene expression, as well as the role of epigenetic mechanisms in the regulation of genome expression associated with the photoperiodic response, is also summarized.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Acknowledgements
  5. References

Photoperiodism in the living world

Demographic success and the survival of most organisms is highly dependent on their ability to cope with and respond to a variety of environmental factors that can be either biotic or abiotic. Biotic factors mainly correspond to pathogens, parasites or predators that limit the development of a given species. Abiotic factors mainly include temperature, humidity and photoperiod, and their combined and continuous fluctuations across seasons can have a strong impact on the fitness of a wide range of organisms. To face the constant modifications of environmental factors, organisms have developed strategies enabling their long-term adaptation to season alternation. The most common and perhaps reliable mechanism is known as photoperiodism, where organisms can detect variations of day length that occur during the year and use them as signals to trigger the establishment of phenotypic/behavioural modifications, allowing their adaptation to seasons. Manifestations of photoperiodism are widespread amongst a variety of organisms, such as fungi, plants and animals. Plants synchronize their life cycle with season alternation. Arabidopsis thaliana is a facultative long-day plant because flowering is promoted by long days and delayed under short-day conditions. By contrast, flowering time in rice is induced by short days (Yanovsky & Kay, 2003). In numerous species of birds, reproduction and migration timing are controlled by endogenous circannual rhythmicity. The expression (or not) of such rhythms depend on the photoperiod and its fluctuation across seasons (Gwinner, 2003). Humans are also sensitive to photoperiod changes because afflictions such as seasonal affective disorder can be diagnosed at the arrival of autumn and winter (Davis & Levitan, 2005). In invertebrates, insects are striking examples of organisms displaying a photoperiodic response. Cold tolerance, migration and growth rate regulation are common responses of insects to day-length changes. However, the two most striking examples of photoperiodism within the insects are diapause and the appearance of seasonal morphs. Diapause is an arrest of development during the life cycle of the insect that allows the anticipation of adverse environmental conditions (drought or cold) and is often triggered by photoperiod (Saunders et al., 2004). The production of seasonal morphs in aphids is historically documented (Tagu et al., 2008): the detection of short days induces a shift from clonal, viviparous reproduction (parthenogenesis) to sexual, oviparous reproduction. A switch from viviparous to oviparous embryogenesis thus occurs within the individuals that detect changes in day length (Tagu et al., 2005). At the population scale, lineages that are able to respond to photoperiodic cues coexist with lineages that have lost this ability and reproduce asexually during their life cycle (Simon et al., 2011). Depending on the type of organism, photoperiodic response can either result in a behavioural change or in the expression of a plasticity of the phenotype that will be more suited to the future environmental conditions. In this context, aphids represent an extreme case of phenotypic plasticity because the sexual morphs produced by asexual individuals that experience photoperiod changes correspond to different and contrasting phenotypes. This discrete phenotypic plasticity is also called polyphenism (Simpson et al., 2011). In aphids, photoperiodism is thus achieved by a plasticity of the reproductive mode. Understanding the molecular basis of this phenomenon offers the possibility not only to decipher the molecular mechanisms involved in the detection and transduction of the photoperiodic signal, but also to understand the molecular events governing the transition from an asexual to a sexual reproductive mode and embryogenesis. Aphids represent an ideal model for understanding the direct phenotypic consequences of the modification of photoperiod. The present review first introduces the aphid model and then focusses on the physiological and transcriptomic bases of key steps of this phenomenon such as the detection and transduction of the photoperiodic signal and the switch from asexual to sexual oogenesis within embryonic ovaries. The second part of this review focusses on the putative role of post-transcriptomic and epigenetic mechanisms associated with the establishment of phenotypic plasticity in response to the photoperiodic changes in aphids.

Aphids: major crop pests remarkably adapted to their environment

Aphids are phloem sap-feeding hemipterous insects that can cause significant economic losses on various crops such as wheat or maize. In temperate and continental regions, most aphid species reproduce quickly and efficiently by viviparous parthenogenesis during spring and summer. At the arrival of autumn, parthenogenetic individuals detect short days. Once sensed, this signal is transmitted to the embryos, which, in turn, direct their development towards becoming sexual adults. The sexual individuals produced (i.e. males and oviparous females) mate and females lay cold-resistant eggs that can withstand potentially adverse winter conditions (Fig. 1). The aphid genome is thus highly ‘plastic’ in the sense that it is able to predict and respond to environmental parameters (seasons) that can be strongly limiting for their survival and general fitness. Aphids are insects that are remarkably adapted to their environment by being able to respond to its fluctuation, explaining their success as one of the major crop pests. Understanding the molecular events regulating the photoperiodic response and more generally phenotypic plasticity in aphids is of major fundamental and agronomical interest for developing sustainable crop pest management strategies. The development of genomic resources within the aphid scientific community in the early years of the 21st Century has allowed significant progress. The pea aphid Acyrthosiphon pisum genome has recently been sequenced and annotated (Richards et al., 2010), which constitutes an absolute pre-requisite for a wide range of genetic and genomic analyses.

image

Figure 1. Pea aphid life cycle and the production of seasonal morphs. In sexual lineages, aphids reproduce efficiently and quickly by parthenogenesis during spring and summer. At the end of summer, parthenogenetic individuals detect short days and initiate the production of sexual individuals in their offspring. Such individuals that produce sexual forms are also called ‘sexuparae’. Autumn morphs (i.e. sexual females and males) are thus produced and will mate to produce cold-resistant eggs that can overcome winter.

Download figure to PowerPoint

Physiological and transcriptomic bases of photoperiodism in aphids

Photoperiodic signal, photoperiodic clock and photoreceptors

Studies on photoperiodism in aphids started to emerge in the second part of the 20th Century, when it was demonstrated that aphids could measure scotophase (night length). For each species, there is a minimum length of scotophase (i.e. critical night length) above which the induction of sexual morphs is effective (Lees, 1973). A minimum number of consecutive long nights is also necessary to trigger the reproductive mode switch under controlled conditions. It can vary between species, although an average of at least ten consecutive long nights is sufficient to trigger the reproductive mode switch. This might be because day length is interpreted as an adaptive strategy to limit a too rapid switch that could be induced by a short period of exposure to long-nights. Complementary studies showed that temperature could also modulate the photoperiodic response (Lees, 1989). The nature of the photoperiodic clock involved in detecting short days has nevertheless not been clearly stated. The involvement of the circadian clock in the photoperiodic response in aphids has been questioned for a long-time. Three main theoretical models for the mechanism of insect photoperiodic clocks have been proposed, two of which suggest an involvement of the circadian clock (Internal or External Coincidence Model), whereas the third (‘Hourglass’ model) does not involve any circadian component. These models are at their essence theoretical and are still largely debated (Danks, 2005; Saunders, 2005). The accumulation of molecular evidence would thus clearly help in discriminating these different models. A recent study (Cortés et al., 2010) revealed the presence in the pea aphid genome of orthologues for several well-known Drosophila circadian clock genes such as period, timeless, Clock, vrille and Pdp1. Expression analyses confirmed a circadian rhythmicity for some of those genes, as well as a significant effect of photoperiod on the amplitude of oscillations. Nevertheless, the exact contribution of the circadian clock to the photoperiodic response remains unknown. The nature and localization of putative photoperiodic photoreceptors has also been investigated. Antibodies directed against a wide range of opsins and other phototransduction proteins were tested and shown to be localized in the ventral anterior region of the protocerebrum, suggesting that the photoperiodic photoreceptors could be located in this area of the brain (Gao et al., 1999). The molecular nature and the precise function of these receptors in the photoperiodic response remain unknown.

Neuroendocrine control

In insects, both endocrine glands and neurosecretory cells can release hormonal components. Steel & Lees (1977) showed that one of the five groups of neurosecretory cells from the protocerebrum (Cell Group I) was involved in the photoperiodic response because micro-cauterization of those cells abolished the response. These cells have long axons spreading into the abdomen of the aphid. Steel & Lees (1977) suggested that secretions (hormones or neuropeptides) from these cells are transported all along the axon and released at specific sites close to the ovarioles, which are the target tissues of the photoperiodic signal, although this has never been demonstrated. The nature of these neurosecretory molecules remains unknown. A recent combination of bioinformatics analyses, brain peptidomics and cDNA analyses allowed the establishment of a catalogue of pea aphid neuropeptides and neurohormones. Forty-two genes encoding neuropeptides and neurohormones were identified. The neuropeptides accumulated in the Group I of neurosecretory cells are probably rich in cysteine (because they respond to fuchsin staining). By correlating the type of neuropeptides rich in cysteine present in the pea aphid genome, and also the knowledge of neuropeptides secreted in other insects, it appears that insulins could represent good candidates for neuropeptides involved in the regulation of photoperiodism (Huybrechts et al., 2010). This hypothesis appears to be realistic considering the results obtained from recent transcriptomic analyses of the photoperiodic response (Le Trionnaire et al., 2009), which show the differential expression of transcripts involved in the insulin signalling pathway (see below).

The involvement of Juvenile Hormones (JH) (known to regulate a wide range of developmental processes in insects) in the control of photoperiodism has also been studied. Topical application of JH or Kinoprene (a JH analogue) on the abdomen of viviparous aphids producing sexual individuals resulted in the reversion of the response to production of asexual individuals (Hardie & Lees, 1985). JH thus appears to play a role in the transduction of the photoperiodic signal. The role of melatonin in the photoperiodic response has also been investigated. In insects, this hormone is involved in the regulation of the visual system and displays a circadian rhythm of expression in head tissues (Bloch et al., 2012). Long-day, parthenogenetic aphids treated with this hormone produce sexual individuals in their offspring instead of asexual individuals (Gao & Hardie, 1997). These results indicate that melatonin might also play a role in the transduction of the photoperiodic signal. To further elucidate the molecular bases of the photoperiodic response, a fine analysis of the genetic programmes set up during this process was needed.

Genetic programmes associated with photoperiodic signal detection and transduction

Initial studies used methods such as the differential display reverse transcriptase-polymerase chain reaction (DD-RT-PCR) or suppression subtractive hybridization to identify transcripts differentially expressed between aphids reared under long days (producers of parthenogenetic progeny) and short days (producers of sexually-reproducing offspring). A transcript homologous to an amino acid transporter within GABAergic neurones was first identified by DD-RT-PCR as being over-expressed in short-day, sexual-offspring-producing individuals (Ramos et al., 2003). A putative role for this transcript in the transduction of the photoperiodic signal was proposed. Suppression subtractive hybridization approaches coupled with quantitative RT-PCR then allowed the identification of transcripts coding cuticular proteins and a β-tubulin that could play a role in hormone responses (Cortés et al., 2008). The precise function of these candidate genes in the regulation of photoperiodism is nevertheless unknown.

Genomic resources such as expressed sequence tag libraries from various aphid tissues were generated (Sabater-Muñoz et al., 2006). These libraries were used to build two generations of cDNA microarrays containing, respectively, 1700 (Le Trionnaire et al., 2007) and 7000 transcripts (Le Trionnaire et al., 2009, 2012). Heads of aphids reared under long-day or short-day photoperiods were collected at five stages of development during the process of sexual morph induction. By focusing on heads and cerebral tissues, the aim was to capture the genetic programmes set up during the initial steps of photoperiodic signal detection and transduction (Le Trionnaire et al., 2007, 2009). Microarray hybridizations combined with proteomics approaches (two dimensional differential in gel electrophoresis) revealed the differential expression of a significant number of transcripts (10% of spotted cDNAs) and peptides within the heads of aphids in response to short photoperiods, allowing the identification of several genetic programmes that could be associated with the photoperiodic response (Fig. 2). Among these, a subset of transcripts showed homologies with Drosophila melanogaster genes involved in the visual system such as Arrestin and Calnexin, known to play a role in rhodopsin phototransduction and maturation. This confirmed an earlier study showing that antibodies against a vertebrate arrestin strongly labelled the putative brain photoperiodic photoreceptors (Gao et al., 1999). Another set of transcripts were related to the nervous system, with several transcripts differentially expressed displaying homologies with Drosophila genes involved in axon guidance (Rho I, NLaz, Capulet and Wunen) and neurotransmission (Kinesin, Dunc 10-4A, Dunc 13-4A and a DEP-containing domain protein), strongly suggesting an involvement of the nervous system in the transduction of the photoperiodic signal. Insulin signalling might also play a role because one transcript encoding an insulin-degrading enzyme and another one coding for an insulin receptor were found to be differentially expressed in response to short photoperiods. Unexpectedly, a large number (n = 38) of cuticular protein transcripts appeared to be regulated. Most of them (n = 25) contained a RR domain (RR1 or RR2) that allows chitin-cuticular protein linkage (Gallot et al., 2010). Most of these transcripts were down-regulated under short-day photoperiods, suggesting a putative relaxing of the chitin-cuticular protein network in response to short days. Cuticle also contains N-β alanyl dopamine (NBAD) that allows linkage between cuticular proteins to produce hard-cuticle or sclerotization. NBAD is made of dopamine and β-alanine and the enzyme responsible for this conjugation is coded by the ebony gene. β-Alanine is synthesized from aspartate by the action of an enzyme coded by black gene. Transcriptomic analyses revealed that ebony and black transcripts were down-regulated in short-day reared aphids. Consequently, it can be hypothesized that less NBAD is synthesized under short-day conditions. This suggests that short photoperiods could result in the reduction of sclerotization level in the aphid heads, thereby modifying cuticle structure. These observations also raise the question of the level of dopamine in aphid heads under short-day conditions. Indeed, if less NBAD is synthesized, is the general level of dopamine affected? Dopamine synthesis involves two main enzymes: tyrosine hydroxylase (th), which metabolizes tyrosine into l-3,4-dihydroxyphenylalanine (l-DOPA), and dopa-decarboxylase (ddc), which metabolizes l-DOPA into dopamine. RT-PCR experiments showed that th and ddc transcripts were down-regulated in short-day reared aphid heads, suggesting that short photoperiods could result in a diminution of dopamine synthesis within aphid brains (Gallot et al., 2010). Because dopamine is a neurotransmitter (and a neurohormone), it is tempting to speculate that this molecule might be involved in the transduction of the photoperiodic signal. A recent study in Locusta migratoria demonstrated that the dopamine synthesis pathway was involved in the transition from the solitary to the gregarious phase (Ma et al., 2011). More precisely, the data showed that th (tyrosine hydroxylase), henna and vat1 (vesicle amino-acid transporter), three genes coding for enzymes involved in dopamine biosynthesis and synaptic release, were significantly down-regulated during the solitary phase. Functional and pharmacological analyses confirmed that the dopamine pathway was clearly involved in the behavioural transition (Ma et al., 2011). Because such a behavioural change in the locust is a case of phase polyphenism (but not triggered by day length changes), a clear parallel with reproductive polyphenism (triggered by photoperiod shortening) can be made and the dopamine biosynthesis pathway might also be involved in the transition from asexual to sexual reproduction in response to short days in aphids. To address this, the level of expression, the localization and the functional characterization of pale, vat1 and henna transcripts in both long- and short-day reared aphids all have to be investigated. It is striking to emphasize that some of these transcriptomic modifications observed on aphids reared under controlled conditions were also detected in aphids reared outdoor under natural photoperiodic conditions. However, the differential expression of several heat-shock protein transcripts also suggested a strong response of aphids to additional environmental parameters such as temperature (Le Trionnaire et al., 2012).

image

Figure 2. Hypothetical model for regulation of seasonal photoperiodism in aphids. Recent large-scale transcriptomic analyses combined with earlier physiological studies allowed the identification of genetic programmes that might play key roles in the regulation of the photoperiodic response. The initial steps of detection and transduction of the photoperiodic signal appear to be associated with a modification of cuticle structure that could be linked to a reduction in dopamine levels within aphid heads. Visual and brain nervous systems might also play a role in this signalling step. Juvenile Hormones were also shown to play a central role in the endocrine transduction of this signal from the brain to the target tissues displaying the reproductive mode switch in the embryos. Later steps corresponding to a shift from asexual to sexual oogenesis appear to be associated with the differential expression of transcripts involved in germline fate and oogenesis, transcriptional and post-transcriptional control, as well as epigenetic modifications.

Download figure to PowerPoint

Transcriptomic modifications associated with the transition from asexual to sexual oogenesis within embryonic ovaries

Once short days/long nights are detected by aphids, this signal has to be transduced to the target tissues, which are the embryos. A recent large-scale transcriptomic approach thus aimed to study the consequences of photoperiodic signal detection and transduction on embryo phenotypic plasticity. Transcriptomes from sexual and asexual embryos along a developmental series were compared using an oligo-nucleotide microarray with approximately 24 000 transcripts (Gallot et al., 2012). Based on previous studies (Corbitt & Hardie, 1985), a perfectly synchronized system was developed to target transcriptomic modifications associated only with the asexual to sexual oogenesis transition in the embryonic ovaries. Aphids reared under short photoperiods contain sexual embryos with a haploid meiotic germline. When Kinoprene (a JH analogue) is applied to the dorsal side of the abdomen, these embryos reverse their reproductive mode and produce asexual embryos containing a diploid non-meiotic germline. Under these conditions, sexual and asexual embryos are perfectly synchronized because the photoperiod does not change. This fine-tuned experimental design was used to compare the transcriptomes of asexual and sexual embryos at three stages of development: 18, 19 and 20 as defined by Miura et al. (2003). These are the final three developmental stages in aphid embryogenesis and correspond to eye differentiation (stage 18), muscle formation (stage 19) and the mature embryo (stage 20). Kinoprene treatment is performed when embryos are at stage 17 (i.e. the latest stage that responds to the hormonal treatment). After that specific stage, embryos are no longer responsive. This developmental window was chosen to study the direct effect of kinoprene on the sexual to asexual oogenesis switch. Statistical analysis of microarrays hybridization results revealed that only a few transcripts (n = 33) were differentially expressed between the two types of embryos. In situ hybridizations confirmed that most of the transcripts were located within germ cells and/or oocytes of asexual and/or sexual ovaries. Regulated transcripts could be assigned to four main functional categories (Fig. 2). Seven of those are involved in oogenesis, with a few playing a role in oocyte axis formation and specification (orb and nudel) or female meiosis chromosome segregation (nanos). Five transcripts play a role in post-transcriptional regulation, such as polyA-tail stabilization (Pop2). Four transcripts are also involved in epigenetic regulations (see below) and three in cell cycle control (cyclin J). These transcripts may therefore determine the aphid clonal or sexual oogenesis. It was thus revealed that JH signalling might control (directly or indirectly) the reproductive fate of aphid embryos.

Combined together, these large-scale transcriptomic approaches allowed the identification of a significant number of candidate transcripts that could play a key role in the detection and transduction of the photoperiodic signal, as well as in the transition from asexual to sexual oogenesis within embryonic ovaries (Fig. 2). The precise function of these different transcripts needs to be tested. The development of stable transgenesis tools remains challenging in aphids, mainly as a result of the complexity of the biological model (telescoping of generations, asexuality with lack of recombination events being predominant during the life cycle). So far, only transitory methods of transcripts silencing (RNA interference) have been developed in aphids with the direct injection of double-stranded RNAs into aphids (Mutti et al., 2006, 2008; Jaubert-Possamai et al., 2007; Shakesby et al., 2009) or by feeding aphids on plants expressing double-stranded RNAs in phloem sap (Pitino et al., 2011; Pitino & Hogenhout, 2013). These technologies displayed various levels of efficiency, mainly depending on the tissue localization of targeted transcripts. Pharmacological approaches (hormone or neurotransmitters injected or topically applied) appear to be a promising alternative for validating the function of specific candidate transcripts or at least signalling/biosynthetic pathways. Nevertheless, strong and efficient methods to modify gene expression in aphids are still missing.

Contribution of post-transcriptional and epigenetic mechanisms

The global expression of a genome is the result of a combination of transcriptomic and post-transcriptomic events that contribute to the establishment of a given phenotype. Small noncoding RNAs and especially microRNAs have emerged in the last years as key post-transcriptional regulators of gene expression (Kim et al., 2009). However, the expression of these different molecules (mRNAs and small RNAs) depends on the accessibility of corresponding genomic regions to transcriptional machinery or transcriptional modulators/regulators. This so-called ‘epigenetic’ state of the genome will thus be at the basis of global genome expression and shape phenotypes. A given epigenome can be explained by a combination of DNA methylation patterns and chromatin structure and organization. Integrating post-transcriptional and epigenetic data with already well-identified transcriptomic changes associated with the photoperiodic response should thus allow the fine characterization of genome expression modifications associated with seasonal photoperiodism in aphids.

MicroRNAs and alternative morph production

The first catalogue of pea aphid microRNAs has been recently completed (Legeai et al., 2010). A combination of bioinformatic prediction of putative hairpin structures (typical of pre-microRNAs) on the genome and high-throughput sequencing of small RNAs from the whole bodies of parthenogenetic individuals allowed the identification of 149 microRNAs, including 55 conserved and 94 new microRNAs. The level of expression of candidate microRNAs between different aphid morphs (parthenogenetic females producing asexual progeny, oviparous/sexual females and parthenogenetic females producing sexual offspring, also called sexuparae) was then tested using a dedicated microRNA chip. Statistical analyses allowed the identification of 17 microRNAs (12 mature miRNAs and 5 miR*) displaying morph-specific profiles of expression. Seven microRNAs were differentially expressed between oviparous females and sexuparae, and nine were differentially expressed between oviparous and parthenogenetic females. Interestingly, ap-let-7 and ap-mir-100 were up-regulated in oviparous females compared with parthenogenetic and sexuparae females. Their Drosophila homologues let-7 and miR-100 have been reported to play a role in metamorphosis and the response to ecdysone, a hormone involved in insect development. Ap-miR-34 also showed different expression levels between sexuparae and parthenogenetic females, which differ by the type of embryos they contain (sexual versus asexual). Interestingly, miR-34 is regulated in D. melanogaster by ecdysone as well as by JH. These microRNAs might thus target transcripts that could play key roles in morph specification and, by extension, in the photoperiodic response.

Sequencing and annotation of the pea aphid genome (IAGC, 2010) revealed that it displayed a high rate of gene duplication. For example, it shows an unexpected expansion of the microRNA pathway for genes that are highly conserved and have only a single copy in most organisms (Jaubert-Possamai et al., 2010; Ortiz-Rivas et al., 2012). There are indeed two copies of the microRNAs pathway-specific dcr1 and ago1 genes. One of the two copies (dic1-b and ago1-b) shows accelerated evolution. RT-PCR experiments also showed a morph-biased expression of these genes showing an accelerated evolution (e.g. dic1-b and ago1-b). This observation raises questions about the specific function of these duplicated copies in the microRNAs pathway within specific aphid morphs, especially in morphs displaying the reproductive mode switch. Further functional analysis will be needed to assess the specific roles of these duplicated copies.

However, systems biology could possibly leverage the lack of functional characterization. MicroRNAs and mRNAs work as a network of interactions because thousands of such interactions are usually predicted for one given species and one specific trait. Genes network and graphs methods are currently being developed to answer this question. A graph can integrate different information: microRNAs–mRNAs interactions, their differential level of expression between two conditions, and additional relationships, such as regulation by transcription factors. This integrated graph allows a global view of a given biological phenomenon. The constitution of such networks in the course of asexual to sexual oogenesis within embryonic ovaries might thus help identify new key regulators of photoperiodism in aphids.

DNA methylation in the pea aphid A. pisum

In mammals, DNA methylation is usually associated with promoter regions and highly methylated regions are correlated to low transcription. This methylation pattern is somehow different in insects. Even if some insect species such as beetles and Drosophila appear to have lost DNA methylation (Patalano et al., 2012), pea aphid as well as honey bee Apis mellifera or locust genome annotation confirmed that all the genes from the DNA methylation pathway are present (Walsh et al., 2010; Hunt et al., 2010). Methylation appears to be important in social insects such as honey bees, which also exhibit a phenotypic plasticity (caste morphs). A recent study showed that 550 genes displayed a differential methylation pattern between queens and workers. Strong correlations between methylation patterns and splicing sites were also found. It was proposed that modulation of alternative splicing could be one of the mechanisms by which DNA methylation is linked to gene regulation in the context of phenotypic plasticity (Lyko et al., 2010). In the case of the pea aphid, Walsh et al. (2010) showed that 0.69% of all cytosines were methylated. Methylation appears to be restricted to gene coding sequences at CpG sites. The precise role of DNA methylation in reproductive mode plasticity in response to photoperiod has not been studied yet, although some studies are currently underway aiming to analyse the role of DNA methylation in the regulation of dispersal polyphenism (Srinivasan & Brisson, 2012). It would thus be of great interest to evaluate the contribution of this epigenetic pathway to the regulation of photoperiodism by evaluating in details DNA methylation patterns between morphs.

Chromatin organization and histone modifications

Chromatin is defined as the association between DNA and proteins (histones and nonhistone proteins). Nucleosomes are sub-units of chromatin made of a DNA fragment of 140 bp wrapped around a protein complex of two copies of each histone protein (H2A, H2B, H3 and H4). Nucleosome numbers and organization all along the chromosome can shape accessibility of genomic regions such as promoters to transcription factors or other regions such as enhancers to regulatory molecules. Nucleosome occupancy can be studied by recently developed methods such as FAIRE-seq (formaldehyde-associated isolation of regulatory elements; Kaplan et al., 2008) and MAINE-seq (MNase-mediated purification of mononucleosomes; Simon et al., 2012) that allow the isolation of protein-free DNA and histone-bound DNA, respectively. Such methods are of great interest for identifying genomic regions epigenetically regulated during a given phenomenon. Nucleosomic histones can also be modified post-translationally. Histone residues such as specific lysines (K) can be methylated or acetylated. The combination of different histone modifications will have consequences for the level of DNA accessibility. Different chromatin states can then be defined by a combination of several histone modification marks. For example, genome-wide profiling of a combinatorial pattern of enrichment or depletion for specific histone modification marks has been established for all the chromosomes of Drosophila, allowing the establishment of a nine-state model for Drosophila chromatin (Kharchenko et al., 2011). So far in aphids, only H3K9me mark and HP1 proteins have been localized on heterochromatic regions (Mandrioli & Borsatti, 2007). More recently, it has been shown that the pea aphid genome possesses a complement of metazoan histone-modifying enzymes with greater gene family diversity than that seen in a number of other arthropods. Several genes have undergone recent duplication and divergence, potentially enabling greater combinatorial diversity among the chromatin-remodelling complexes (Rider et al., 2010). The comparison of sexual and asexual aphid transcriptomes (Gallot et al., 2012) demonstrated the differential expression of transcripts coding proteins involved in epigenetic mechanisms, such as Histones H2B.3 and H1, which are known to participate to chromatin assembly and disassembly. Another example concerns Suv4-20H1, which is involved in histone methylation. This fine comparison of sexual and asexual embryos transcriptomes already suggests that some epigenetic regulations involving chromatin structure modifications are occurring during phenotypic plasticity. Depicting the type of histone modifications associated with the reproductive mode switch of embryos in response to photoperiodic cues would thus be of great interest.

Perspectives

The regulation of photoperiodism in aphids and its effects on the embryo phenotypic plasticity has been extensively studied at the transcriptomic level. These large-scale studies have allowed the identification of some of the genetic programmes involved in the photoperiodic signal detection and transduction and in the embryos' reproductive mode switch. These studies have established an extensive catalogue of transcripts, hormones and neurotransmitters (e.g. insulin, dopamine) as candidates for further functional and pharmacological validation experiments. The recent and on-going development of high-throughput sequencing technologies now allows the identification of key post-transcriptional regulators of gene expression (such as microRNAs), as well as the mapping of distinct epigenetic marks (nucleosome occupancy, histone modification marks and DNA methylation patterns). The establishment of alternative phenotypes in response to environmental cues such as photoperiod definitely involves a combination of epigenetic, transcriptomic and post-transcriptomic regulatory events. An integrative view [in accordance with the modENCODE model (Celniker et al., 2009) but for a non-model organism such as aphids] of the contribution of these different mechanisms thus appears to be an ideal approach that should allow the identification of key genomic regions involved in the regulation of phenotypic plasticity, especially in the case of the aphid photoperiodic response.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Acknowledgements
  5. References

Jennifer Brisson (University of Nebraska) is sincerely thanked for her help in reading and correcting this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Acknowledgements
  5. References