The role of neuropeptide F in a transition to parental care

Abstract

The genetics of complex social behaviour can be dissected by examining the genetic influences of component pathways, which can be predicted based on expected evolutionary precursors. Here, we examine how gene expression in a pathway that influences the motivation to eat is altered during parental care that involves direct feeding of larvae. We examine the expression of neuropeptide F, and its receptor, in the burying beetle Nicrophorus vespilloides, which feeds pre-digested carrion to its begging larvae. We found that the npf receptor was greatly reduced during active care. Our research provides evidence that feeding behaviour was a likely target during the evolution of parental care in N. vespilloides. Moreover, dissecting complex behaviours into ethologically distinct sub-behaviours is a productive way to begin to target the genetic mechanisms involved in the evolution of complex behaviours.

1. Introduction

Parental care is a complex social phenotype that integrates many behavioural pathways [1–4], which can be predicted based on behavioural precursors that must exist before parenting can evolve [1]. For example, parenting often evolves in species with pre-existing defence of mates and food resources [1]. Other predicted behavioural precursors are changes in timing of reproductive and feeding behaviour. Predictable behavioural precursors allow us to examine mechanisms, including genetics, that change to allow parenting.

A candidate gene screen is a powerful approach for examining molecular genetics underlying complex behaviour [5]. Predictions of candidate genes are based on the assumption that the action of a gene is conserved across taxa during the expression of homologous/analogous behaviours. However, newer evidence points to an important role for conserved pathways rather than individual genes influencing behaviours [6,7]. Thus, a modification to the classic candidate gene screen is to predict pathways expected to underlie behavioural evolution.

Here, we test the hypothesis that gene expression influencing feeding behaviour is modified during parental care. Feeding pathways have altered gene expression in parenting insects, fish and mammals [8–10]. We characterize the expression of a pathway that heavily influences the motivation to eat, neuropeptide F (neuropeptide Y in vertebrates [11]), during parental care in the subsocial burying beetle Nicrophorus vespilloides, an insect with complex parental care [12,13]. Burying beetles require a vertebrate carcass for reproduction, which serves as a food resource for both the parents and the offspring. The parents provide indirect care for offspring by preparing a carcass and maintaining it to retard microbial growth. Parents also provide direct care by regurgitating pre-digested food to begging larvae. There is temporal kin recognition in N. vespilloides, and transition into willing parents (caring for larvae rather than eating them) between the second and third day after beginning to prepare the carcass, prior to the hatching of larvae [14]. Care lasts for around 72 h, at which time larvae are completely self-feeding and independent [15]. Parents disperse around 90 h and larvae disperse about 120 h after hatching [16], at which time the carcass is typically completely consumed. Larvae do not eat again until they are adults.

Because parental care requires regurgitation, parents cannot eat and feed offspring simultaneously and must partition their time between feeding themselves and their offspring. Therefore, the regulatory mechanisms of eating and care should overlap [17]. We predicted that npf expression will covary with the behavioural transition to direct parental care. Specifically, we predicted that npf signalling and, therefore, motivation to feed oneself would be reduced during direct parental care. To test this, we assessed gene expression of npf and its receptor in five behavioural states that reflect a transition from a non-parenting state to a parenting state and back (figure 1). We predicted that npf signalling should be repressed while parents directly care for larvae but should return to pre-breeding levels after parenting is complete.

Figure 1.

Simplified timeline of transitions in parenting and larval development of N. vespilloides. We have included time points when samples were collected. Virgins were collected directly from housing boxes. Note all samples were age-matched.

2. Material and methods

(a). Experimental design

We collected 10 individuals in five treatments: virgin (non-social), a female and male paired for 48 h without a mouse (social, no resource, no parenting), a female and male paired for 48 h with a mouse (social, with resource, parenting, no larvae), uniparental females and males during peak direct care (approximately 24 h after larvae arrive; non-social, with resource, larvae present, direct parenting) and post-caring females and males (approximately 5 days after larvae arrive, non-social, no resource, no parenting). Adult beetles were sexually mature, outbred and fed ad libitum with mealworms prior to any treatments. Samples were collected when individuals were 21 days post-adult eclosion for all treatments.

(b). Gene sequence validation

We used the available protein sequences of Drosophila melanogaster and Tribolium castaneum to search the genome of N. vespilloides [18] for the sequences of npf and npfr using tBLASTn (v. 2.2.25+ [19]), validated by searching against non-redundant insect proteins at NCBI using BLASTx with default settings. We obtained single copy, full-length orthologues for both npf and npfr.

(c). Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) protocols and analysis

Sample collection (whole heads), RNA extraction and cDNA synthesis are described in [20]. Primer sequences were generated and validated as in [21]. We ran samples on a Roche LightCycler 480 [21] with tbp as the endogenous control [20]. Analyses were performed with the MCMC.qpcr R package [22] using an ‘informed’ MCMC qpcr model with default settings and both sex and behavioural state as factors. Results are reported as natural log fold changes from the a priori comparison state of virgin and female with p-values calculated using the posterior distribution tested with a z-score against a normal distribution and corrected for multiple testing.

3. Results

(a). npf gene expression across different reproductive/social contexts

The post-care treatment showed a statistically significant increase from virgins (ln(fold change) = 0.351, p_MCMC = 0.018; figure 2a). There was no statistically significant effect of sex or of the interaction between sex and behavioural state.

Figure 2.

The graph displays the log₂(abundance) of both components of neuropeptide F signalling, (a) the prepropeptide npf and (b) the npf receptor (npfr), across a breeding cycle in both male and female N. vespilloides. Values are represented as means and 95% credible intervals of the posterior distribution. *Statistically significant.

(b). npfr gene expression across different reproductive/social contexts

There was an overall effect of sex (ln(fold change) males = −0.926, p_MCMC < 0.001; figure 2b). The mated-with-a-mouse and active caring treatments had a statistically significant decrease from virgins (ln(fold change) = −1.167, p_MCMC < 0.001; ln(fold change) = −1.071, p_MCMC < 0.001, respectively). There was not a statistically significant interaction between sex and behavioural state.

4. Discussion

We hypothesized that N. vespilloides parents should have reduced motivation to eat while they feed their young, so gene expression associated with food seeking pathways should be decreased (sensu [17]). We tested this by profiling the gene expression of the neuropeptide F system in parenting and non-parenting states. The npf system is a highly conserved ligand–receptor that heavily influences the motivation to eat in organisms from insects to mammals [11]. We found greatly reduced expression of the npf receptor in all parenting states. This suggests that parents' response to food sources is modified during parenting and supports the hypothesis that feeding pathways are targeted during the evolution of parental care [17].

Changing expression of the npf receptor, rather than the neuropeptide itself, may localize the effect to specific subpopulation of neurons. Neuropeptide F signalling is under many different controls and has different phenotypic influences depending on the specific subpopulation of neurons activated [11]. For example, npf can have sex-specific expression [23] and influences locomotor and circadian activity [24]. npf acting within the larval brain is developmentally controlled and influences the transition to wandering [25]. npf also serves as a neurohormone and influences the production of ecdysteroids [26], and ovarian [27] and testicular physiology [28]. Thus modifying the expression of a receptor, which is always bound to a single neuron, potentially invokes fewer constraints than releasing a ligand into the haemolymph.

Our prediction that the npf system would be involved in parenting was based on its role in motivation to eat. Honeybee workers involved in brood care have reduced npf signalling compared with foraging bees [8] consistent with the decreased expression we see of npfr expression during parenting in N. vespilloides. However, npf signalling may play multiple roles in the transition to parenting as it can also influence social interactions. Decreased npf increases male–male aggression in Drosophila [29]. Reduced npfr expression when parenting may reflect the need to defend the resource and young from conspecifics. Neuropeptide F also influences mating behaviour [23,26,30] and decreased npfr signalling causes increased social aggregation in Drosophila [25] and Caenorhabditis elegans [31]. Thus, the depression of npfr here might facilitate parenting by influencing the tendency of individuals to remain with a carcass and offspring, and to stop mating. It is unlikely that parents have depressed npfr expression because they are satiated from eating the carcass because the first four behavioural states tested had food in excess of physiological needs and the post-caring treatment had only been without a carcass for one day. Additionally, when food-deprived, npf increases expression in male N. vespilloides while npfr is not differentially expressed in either sex (see the electronic supplementary material). Finally, in a cichlid, Grone et al. [9] found that one npfr orthologue was differentially expressed during parenting (mouthbrooding) but not during food deprivation. Thus, a decreased level of npf and its receptor provides a mechanism to affect multiple behavioural changes from solitary individual to caring parent.

We found expression of npf itself increased in the post-care treatment. It is unlikely that this was due to hunger as they can eat on the carcass. Rather, given the ecology of N. vespilloides, it is more likely that increased npf is associated with preparation for another bout of reproduction. In other insects npf is associated with an increased motivation to seek a new reproductive resource or mating opportunities [8,30]. It also acts as a signal to reinitiate full reproductive capabilities [27,28], such as yolking up new eggs, or increased motivation to leave a social group [25,31].

Ethics

This research complies with all local, state and federal laws of the USA regarding research.

Data accessibility

All supporting data are available in Dryad: http://dx.doi.org/10.5061/dryad.330c1.

Authors' contributions

C.B.C. and A.J.M. designed the research. C.B.C., K.V., D.B.K. and E.C.M. conducted the molecular work. C.B.C. analysed the data. C.B.C. and A.J.M. drafted and all authors revised the manuscript. All authors agree to be accountable for the content and approve publication.

Competing interests

We have no competing interests.

Funding

This research was supported by the US NSF IOS 1326900.