We investigated whether insulin signaling, known to mediate physiological plasticity in response to changes in nutrition, also facilitates discrete phenotypic responses such as polyphenisms. We test the hypothesis that the gene FOXO – which regulates growth arrest under nutrient stress – mediates a nutritional polyphenism in the horned beetle, Onthophagus nigriventris. Male beetles in the genus Onthophagus vary their mating strategy with body size: large males express horns and fight for access to females while small males invest heavily in genitalia and sneak copulations with females. Given that body size and larval nutrition are linked, we predicted that 1) FOXO expression would differentially scale with body size (nutritional status) between males and females, and 2) manipulation of FOXO expression would affect the nutritional polyphenism in horns and genitalia. First, we found that FOXO expression varied with body size in a tissue- and sex-specific manner, being more highly expressed in the abdominal tissue of large (horned) males, in particular in regions associated with genitalia development. Second, we found that knockdown of FOXO through RNA-interference resulted in the growth of relatively larger copulatory organs compared to control-injected individuals and significant, albeit modest, increases in relative horn length. Our results support the hypothesis that FOXO expression in the abdominal tissue limits genitalia growth, and provides limited support for the hypothesis that FOXO regulates relative horn length through direct suppression of horn growth. Both results support the idea that tissue-specific FOXO expression may play a general role in regulating scaling relationships in nutritional polyphenisms by signaling traits to be relatively smaller.
Citation: Snell-Rood EC, Moczek AP (2012) Insulin Signaling as a Mechanism Underlying Developmental Plasticity: The Role of FOXO in a Nutritional Polyphenism. PLoS ONE 7(4): e34857. https://doi.org/10.1371/journal.pone.0034857
Editor: Tim Thomas, The Walter and Eliza Hall of Medical Research, Australia
Received: November 16, 2011; Accepted: March 8, 2012; Published: April 13, 2012
Copyright: © 2012 Snell-Rood, Moczek. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: ESR was supported by NIH NRSA F32GM083830. Additional support was provided by National Science Foundation grant IOS 0820411 and IOS 0718522 to APM. The Center for Genomics and Bioinformatics (where qPCR analyses were run) is funded in part by the METACyt Initiative of Indiana University, which is funded in part through a major grant from the Lilly Endowment. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Variation in the quality and quantity of nutrition is a ubiquitous challenge during development both within and across generations. Organisms have evolved a wide range of behavioral, developmental, and physiological mechanisms to cope with such variation. Central among the developmental and physiological mechanisms is the insulin signaling pathway, which permits major adjustments in growth, development, body size, lifespan and behavior in response to nutrition (reviewed in –). In times of plenty, insulin or insulin-like peptides promote, via the insulin receptor, a variety of cellular processes such as cell division, metabolism, and protein synthesis –. During periods of lower food availability, the decrease in insulin or insulin-like peptides results in the activation of the transcription factor forkhead-box-subgroup O, or FOXO, which subsequently inhibits growth –, and promotes immunity , stress resistance , lifespan –, lipid metabolism , and increased insulin sensitivity through up-regulation of the insulin receptor , .
Insulin signaling is a highly conserved pathway that allows organisms as diverse as mammals, insects and nematodes to cope with fluctuations in diet through proportional changes in growth and life history traits . However, many organisms adopt qualitatively different strategies or grow vastly different traits depending on nutrition. For instance, variation in nutrition results in the adoption of different mating tactics in many species of insects and fish ,  and different reproductive castes in many social insects , . In mammals, diet is hypothesized to play a major role in the development of different metabolic syndromes . This raises the questions whether insulin signaling has been co-opted from a primary role in basic physiology to a mechanism of phenotypic plasticity in response to nutritional variation.
Several lines of evidence suggest that insulin signaling should be a common mediator of plastic responses to variable nutritional environments. First, insulin signaling plays a role in determining the relative size of tissues and organs (, , ; reviewed in , ). Changes in the relative size of tissues across a nutritional gradient are important components of plasticity. Such flexibility and diversity in allometric scaling relationships is thought to be regulated by insulin signaling , , . For example, shallow-sloped correlations between body size and genitalia , in contrast to isometric scaling relationships of many other traits, may be due to lower sensitivity of genitalia to insulin signaling, possibly through changes in insulin receptor density . Additional evidence linking insulin signaling to plasticity comes from gene expression studies. Candidate genes in the insulin signaling pathway show different expression patterns between polyphenic morphs, such as different reproductive morphs of paper wasps , reproductive, worker or nurse castes of honey bees –, predator-induced morphs of Daphnia , or sneaker and fighter morphs of horned beetles .
We sought to test the role of insulin signaling in nutritionally induced phenotypic plasticity using the sneaker-fighter polyphenism in horned beetles. Beetles in the genus Onthophagus construct brood balls (of dung) that support the entire larval development of individual offspring. The body size of an adult is largely influenced by the size and nutritional quality of their brood ball , . Emerging adult males adopt distinctly different reproductive tactics depending on their own body size. Large adult males use horns in aggressive contests with other males over females and their tunnels . In contrast, small males grow only horn rudiments and adopt sneaker tactics, digging side tunnels to access to females, or sneaking copulations with females as horned males fight . Small, hornless males are more maneuverable in tunnels , and instead of growing horns , , they often invest in relatively larger genitalia and/or ejaculates , .
We tested the hypothesis that insulin signaling plays a role in the development of this nutritional polyphenism in mating tactics in horned beetles, using a combination of observation and manipulation of patterns of gene expression. We chose to focus on one important player in the insulin signaling pathway, FOXO, because it has been linked to variation in scaling relationships in a range of systems , , , ,  and hypothesized to be an important regulator of horned beetle polyphenisms . Recent microarray analyses  provided the first empirical support of this hypothesis by documenting elevated FOXO expression levels in developing horn tissue, relative to expression in the abdomen, of small, hornless males compared to large, horned males. Furthermore, recent work in Drosophila suggests that FOXO expression may regulate tissue-specific responses to nutritional variation , an important criterion for a gene involved in a nutritional polyphenism.
We sought to test two hypotheses for the potential role of FOXO in this nutritional polyphenism. First, we investigated the hypothesis that FOXO regulates horn size relative to body size. Mechanistically, in this case, the activation of FOXO in the developing horn tissue of small, nutritionally stressed males is expected to repress horn growth. This hypothesis predicts that FOXO expression would be elevated in the developing horn rudiments of small males, and that knockdown of FOXO would result in larger horn length for a given body size. Second, we address the hypothesis that FOXO regulates genitalia investment relative to body size. Past research has suggested that costly horns ,  may tradeoff with investment in genitalia and ejaculates , . Mechanistically, the growth of a large horn in a beetle may trigger FOXO expression in other developing regions, such as the genitalia, resulting in the repression of their growth. This hypothesis predicts that FOXO expression will be elevated in developing abdominal and genitalia tissues of large, horned males and that knockdown of FOXO would result in larger genitalia size for a given body size. Our results suggest that FOXO plays an important role in nutritional polyphenisms, negatively regulating relative trait size for both horns and genitalia.
We sequenced a 1204 bp region of the FOXO transcript (912 bp of coding sequence) in Onthophagus nigriventris. Using BLASTp, this region significantly matched FOXO from Tribolium castaneum (e value = 3e−116) with 79% matching amino acid identity. Additionally, we sequenced FOXO in several species of Onthophagus and aligned the protein sequences. We found that Onthophagus beetles, relative to Tribolium, have a large insertion (5–9 amino acids) in the coding region of the gene (Fig. S1). For subsequent analyses, we focused on a region of the gene past the conserved forkhead domain (see Fig. S1 and methods for more details).
FOXO Expression in Untreated Individuals
Contrary to our expectations, FOXO expression, relative to that of a control gene (actin) was not correlated with body size in the developing horn region (the thoracic epidermis) of either males or females sacrificed as first day pupae (Table 1, Fig. 1). However, FOXO expression was positively correlated with body size in the abdominal epidermis of untreated males: larger males had consistently and significantly higher relative FOXO expression (Table 1, Fig. 1). Because this result was in contrast to our expectation of higher FOXO expression in the prothorax of small, hornless males, yet revealed an unexpected elevation of FOXO expression in the abdomen of large males, we wished to further investigate location and level of FOXO expression in the abdomen of males.
Real-time quantitative PCR was used to measure expression of FOXO in 21 male and 18 female untreated individuals sacrificed as first-day pupae. Gene expression was measured in both the thoracic and abdominal epidermis. We used ANOVAs to test for effects of body size on FOXO expression, relative to the expression of the control gene (actin). Separate ANOVAs were run for each tissue and sex (see Table 1). Shown are leverage plots from an ANOVA that included body size (thorax width) and expression of a control gene (actin Ct) as independent variables (FOXO Ct was the dependent variable); thus, relative FOXO expression is plotted as the dependent variable. The body size range included small, hornless males to large, horned males.
In situ-hybridizations were performed on males to investigate spatial patterns of gene expression. In line with the qPCR results reported above, we observed FOXO expression in several locations in the abdomen of large, horned males, but not in the comparable location in small, hornless males (observed in three replicates for each morph; Fig. 2, 3). Based on the location and shape of these clusters of expression (2–4 oval clusters in the ventral posterior abdomen, adjacent to a single rectangular structure), we can putatively identify FOXO expression in the developing testes and copulatory organs of large, horned males. A DAPI stain of the putative testes region showed a cell-rich area at the anterior end of the oval shaped structure (Figure 2E), consistent with the densely clustered, mitotically dividing cells of the anterior testes (“germinal proliferating center:” ). While the region of FOXO expression was consistent with expression in the genitalia, we cannot rule out additional expression occurring in the fat body. There was little detectable (if any) FOXO expression in the developing horns or prothorax of any of these samples (Fig. 2, 3).
Shown are in situ hybridization images of a representative sagittal-medial sections of large, horned males, obtained from O. nigriventris pupae during the first day of the pupal stage. A, B) FOXO expression in the posterior-ventral region of the abdomen. (C, D) A close-up of FOXO expression (dotted rectangle in (A)) in the putative developing genitalia, including the putative testes (solid arrow) and copulatory organ (dashed arrow). This may also include expression in the fat bodies. The coloring of the cuticle is a normal artifact. (E, F) DAPI stain of (C) showing putative testes, including a region of densely packed cells, the putative germinal proliferating center.
Shown (right panels) is a representative in situ hybridization showing FOXO expression in small, hornless male pupae O. nigriventris (day 1). (B) shows a close-up of the square area highlighted in (A). (C) and (D) label the different body regions shown in the section.
Taken together, our expression results suggest that FOXO is expressed in the developing genitalia (and possibly the fat body and/or adjacent areas in the abdomen) of large, horned males, but much less abundant in the corresponding tissue regions in small, hornless males. Contrary to our initial expectations, FOXO expression is indistinguishable in developing pupal horn tissues of large and small males.
FOXO RNAi Knockdown
FOXO RNAi had significant effects on body size and development time in males, but not females. Relative to control-injected individuals, RNAi males had significantly larger pupal thorax widths and pupal weights, extended length of their third instar, and marginally significantly larger adult thorax widths (Fig. 4, Table 2). In contrast, none of these variables differed between control-injected and RNAi females. The body size response in males was proportional to the degree of FOXO knockdown as approximated by the amount of dsRNA injected (see Table S1). However, the degree of FOXO knockdown did not affect relative horn length (F1,41 = 0.58, P = 0.45) or genitalia scaling relationships and thus was not included in subsequent analyses.
Shown is mean and standard error of pupal mass for beetles were injected with at least 1 ug ds FOXO RNA relative to control-injected individuals. There was a significant effect of FOXO knockdown on body size in males, but not females (see Table 2).
We also detected a moderate effect of RNAi on horn expression. Relative to control-injected individuals, RNAi male adults had significantly longer horns relative to their body size (Fig. 5, F1,52 = 5.75, P = 0.02). Residual horn length did not differ between control and treatment males in the pupal stage (F1,62 = 1.67, P = 0.20). In females, which express a small horn as pupae and a ridge as adults, we observed no differences in horn-body size relationships between control-injected and RNAi individuals for either the pupal (F1,57 = 0.59, P = 0.44) or adult stage (F1,44 = 0.28, P = 0.59).
Allometry between body size (thorax width) and thoracic horn length in control-injected and FOXO knockdown individuals (treated with 1.0, 1.75, or 2.5 ug dsRNA). The lower panel shows the difference in relative horn length between control-injected and FOXO knockdown individuals – the difference was modest, but significant (P = 0.02).
FOXO knockdown also had significant effects on the relative size of the male copulatory organ. Overall, FOXO-knockdown individuals had larger copulatory organs for a given body size relative to control-injected individuals (F1,50 = 3.93, P = 0.05; mean (se), control: 1.58 (0.009), FOXO: 1.61 (0.009), see also Fig. 6). There was a marginally significant interaction between body size and treatment (F1,50 = 3.23, P = 0.07; size effect: F1,50 = 1,28, P = 0.26; see also Fig. 6). Combined, these RNAi results suggest that FOXO regulates not only body size, but also the relative growth of traits important in a nutritional polyphenism – horns and genitalia.
Allometry between body size (thorax width) and copulatory organ length (paramere) in control-injected and FOXO knockdown individuals (treated with 1.0, 1.75, or 2.5 ug dsRNA). The body size range included small, hornless males to large, horned males (see Fig. 5).
We used quantitative real-time PCR to validate the RNAi knockdown. We harvested thoracic and abdominal tissue from 20 male and female first-day pupae (within 24 hours of pupation) of control-injected (N = 11) and FOXO-injected (1.0 ug dsRNA, N = 9) individuals. We tested for differences in relative FOXO expression in an ANOVA that included expression of the control gene actin (thorax: F1,16 = 2.33, P = 0.15; abdomen: F1,14 = 0.24, P = 0.63), and a measure of body size (pupal mass: thorax: F1,16 = 10.4, P = 0.005; abdomen: F1,14 = 0.63, P = 0.44). We found a significant effect of dsRNA injection on FOXO expression (Figure S2, thorax: F1,16 = 5.89, P = 0.02; abdomen: F1,14 = 6.83, P = 0.02). Using the methods of Pfaffl , this corresponded roughly to a 0.80 and 1.08 fold reduction in FOXO expression, relative to control-injected individuals, for the abdomen and thorax, respectively. Because our validation was on individuals injected with 1.0 ug FOXO dsRNA, all analyses focused on individuals injected with at least 1.0 ug dsRNA.
We present evidence that suggests insulin signaling plays a role in nutrition-induced phenotypic plasticity in horned beetles. We found only modest support for a role of FOXO in the direct regulation of relative horn size in males. However, our data provide support for the hypothesis that FOXO regulates the relative size of male genitalia, thus playing an important role in trait integration across polyphenic morphs. These results are consistent with recent observations that FOXO expression regulates tissue-specific responses to nutritional variation in Drosophila . We discuss our findings in the context of what is known about FOXO and insulin signaling, and future directions for investigating FOXO as a player in nutritional plasticity.
Conservation of the Insulin Signaling Pathway
Our data are consistent with previous studies on insulin signaling and FOXO in particular. We found that FOXO knockdown affected both development time and body size (Table 2). Previous studies on Drosophila have shown that manipulations of insulin signaling prior to the larval critical weight affect development time, and manipulations that occur after the larval critical weight affect body size . While the critical weight is unknown in this species, our knockdown likely occurred just prior to, and affected animals for several days following, critical weight, at least as defined in Drosophila (mass at which 50% of animals will survive to pupation without further feeding, ). We injected beetles between day 6 and 10 of the third larval instar, just before peak mass occurs (between day 10 and 12; ) and effects of the knockdown were measurable using qPCR in first day pupae. Thus it is likely that we affected development periods both before and following critical weight.
We found that FOXO knockdown increased adult body size (Table 2, Fig. 4). Previous empirical work has shown that increased insulin signaling (through an increase in expression of insulin-like peptides or the insulin receptor) causes an increase in body size , , while decreased insulin signaling causes a decrease in body size , , , , . Similarly, a range of studies in Drosophila have found that over-activation of the growth inhibitor FOXO results in decreased body size –. Thus, we should expect decreased FOXO expression to result in an increase in body size, as found in our study (Table 2, Fig. 4). In contrast to this expectation, one previous study in Drosophila found that a FOXO null mutation had no effect on adult body size , though it is possible that the expected increase in body size is dependent upon sufficient nutrition. In support of this idea, we also observed a significant increase in development time in males (Table 2). Because we manipulated FOXO prior to the critical weight, it is possible that larger body size could have been achieved through a flexible increase in development time (and resulting total nutrient pool). Variation in nutrient availability or timing of knockdown may result in variation among studies in the effect of FOXO knockdown on body size.
We also found that the effect of FOXO on body size in the present study was specific to males (Table 2, Fig. 4), recalling previous work that found sex-specific effects of FOXO knockdown . It is unclear exactly why FOXO knock-down would have sex-specific effects on body size. However, the result recalls the idea that sexually selected traits generally evolve greater degrees of condition dependence , resulting in greater degrees of nutritional sensitivity in many traits in males relative to females . Given that body size is under sexual selection in male beetles (due to its link with contest outcome, reviewed in ), our results are consistent with this idea.
Our results also suggest a role of FOXO in the regulation and establishment of scaling relationships. Previous studies have shown that changes in insulin signaling in specific tissues, including FOXO activity, can result in organ-specific changes in size , , , , , . Such effects are what make the insulin signaling pathway a prime candidate for determining diversity in trait allometries, or differences in shape within and between species and nutritional environments. For example, a tissue-specific change in FOXO expression (as suggested in ) or insulin receptor density (as suggested in ) has the potential to change the relative size of an organ. Indeed, target gene responses to FOXO activity vary with tissue type , . Consistent with these expectations, and as we discuss in more detail below, we found that FOXO expression varied with tissue types (Fig. 1) and knockdown resulted in changes in trait-body size relationships for both thoracic horns (Fig. 5) and genitalia (Fig. 6).
Overall, these data suggest that FOXO may partially mediate its effects on allometry through differential expression. While the classical view of FOXO action considers changes in activity through protein phosphorylation (rather than differential expression) (reviewed in , ), recent work suggests a more complex picture , , . The present research provides further support for the emerging idea that changes in FOXO expression may be just as important as changes in FOXO activity in generating changes in size and shape . However, future work will be necessary to determine the relative importance of FOXO expression versus activation in the development of polyphenisms.
FOXO Expression Is Linked to a Nutritional Polyphenism
Several lines of evidence implicate FOXO in the development of the fighter-sneaker polyphenism in horned beetles in the genus Onthophagus. First, we found a correlation between FOXO expression and body size (Fig. 1) – a result of larval nutritional conditions. This relationship was restricted to males, the sex with pronounced nutritional polyphenism. Specifically, we found that FOXO expression in the abdominal epidermis was positively related to body size in males. FOXO expression in the developing horn tissue of the thoracic epidermis, however, was not. This result confirmed previous microarray analyses which indicated that FOXO expression in the thoracic tissue, relative to the abdominal tissue, was higher in small, hornless males than in large, horned males . However, contrary to initial expectations this relative difference was not due to higher expression of FOXO in the thorax of small compared to large males; instead it was due to higher FOXO expression in the abdomen of large compared to small males. We sought to clarify the spatial pattern of expression of FOXO in the abdomen using in situ hybridization.
In situ hybridizations suggested that FOXO expression was most pronounced in the putative developing genitalia (testes and copulatory organ) of horned males (Fig. 2, 3), although it is possible expression was also present in the fat body. This result is consistent with the qPCR result showing increasing FOXO expression with body size in the abdomen of males. However, given that the tissue used in the qPCR analyses was dorsal abdominal tissue, while the genitalia are located more ventrally, it suggests that the qPCR results may have been driven by FOXO expression in a range of abdominal tissues such as the fat body, genitalia and/or other tissues adjacent to the dissection location (first day pupae have indistinct borders between internal tissues).
Taken together, the results of the qPCR and in situ hybridization run counter to the initial hypothesis that small, nutritionally stressed males would express FOXO in the thoracic tissue, resulting in the repression of horn growth . Instead, these observations support the hypothesis that horn growth in large, horned males may trigger nutrient stress and FOXO expression in other body regions which signals organs within those tissues to grow proportionally smaller. We sought to test this hypothesis further via knockdown of FOXO expression.
We used RNAinterference to test the degree to which FOXO affected horn and genitalia size, relative to body size. We found that FOXO knockdown had significant effects on genitalia size relative to body size (Fig. 6): specifically, FOXO knockdown individuals had relatively larger genitalia relative to control-injected individuals. This observation is consistent with the hypothesis that FOXO may regulate relative genitalia size and that FOXO expression in the abdomens of large males lowers genitalia investment. We also observed that the slopes of the body size-genitalia relationship were different between the two treatments (although this interaction was only marginally significant) such that relative differences in genitalia size were most pronounced in smaller sized individuals. It is possible that growth of horns in large males prevented genitalia growth by means of nutrient limitation despite the signal for continued growth stimulated through depressed FOXO expression, although future research is necessary to test this mechanism. FOXO RNAi also significantly affected the size of thoracic horns relative to body size, although the effect was modest overall (Fig. 5). Relative to control-injected individuals, RNAi individuals expressed larger horns for a given body size (Fig. 5), with the effect being most pronounced in several intermediate-sized and large individuals. This is consistent with the initial hypothesis that FOXO expression in the developing thoracic epidermis may regulate horn growth. However, given that FOXO expression (in first day pupae) was weak in the prothorax of small, hornless males at the first day pupae stage (Fig. 1, 3), this suggests that a corresponding window of nutrition-mediated differential FOXO expression must come earlier in development, for instance during the pre-pupal stage when horn tissue is proliferating. Furthermore, the RNAi effect was most pronounced for intermediate and large males, while the original hypothesis predicted the effects to be most pronounced in small, nutritionally stressed males. Regardless, both the horn and genitalia results taken together suggest that FOXO may generally regulate relative trait size by suppressing growth.
Conclusions and Future Directions
In summary, our results suggest that FOXO plays a role in the development of the Onthophagus polyphenism. Our data provide limited support for the initial hypothesis that FOXO regulates relative horn length through suppression of horn growth. However, both expression and knockdown data suggest that FOXO regulates relative genitalia size. Both results show that FOXO expression negatively affects relative trait size, thus integrating the responses of different tissues and organs to nutritional variation. Small, hornless males generally invest more heavily in testes, ejaculate composition and copulatory organs due to increased sperm competition and the reduced costs of growing horn tissue –, . Our results suggest that FOXO may play a role in mediating the growth response of genitalia to nutrition and horn growth. In particular, our results suggest that FOXO expression may be triggered in the abdomen, which in turn may inhibit the growth of genitalia. This role of FOXO in mediating horn-genitalia tradeoffs parallels FOXOs role in mediating life history tradeoffs –, , and is consistent with recent studies demonstrating a role of FOXO in organ-specific responses to nutritional variation in Drosophila .
This work contributes to a growing body of work suggesting insulin signaling as an important mediator of phenotypic plasticity. For instance, the insulin receptor and insulin itself (or insulin-like peptides), show significant differences in expression between polyphenic morphs in a range of systems , –. This study is the first to show that FOXO, a key growth inhibitor within the insulin pathway, additionally plays a role in the regulation of polyphenic development.
The results of this work suggest several promising areas of future research. A key question for future studies is to investigate whether FOXO expression in one body region is directly triggered by the growth of tissue in another body region (such as horns) or alternatively, is regulated by overall body size. Either scenario suggests different mechanisms by which FOXO results in trait integration, which could be addressed experimentally by manipulating horn size (via ablation) and nutrition (nutrient supplement or starvation) independent of body size. An additional interesting line of inquiry would be to integrate insulin signaling with endocrine signaling, a known mediator of polyphenisms , . Given that both ecdysone and juvenile hormone interact with insulin signaling in insects (e.g., , ), this would be a promising future approach to better understand and integrate the mechanism of insulin signaling-mediated growth regulation in polyphenisms. Finally, it would be interesting to examine insulin signaling across species that have diverged in body size and the degree of the male polyphenism .
Onthophagus nigriventris were collected from populations in Waimea, Hawaii. No specific permits were required for the described field work (this species is not an agricultural pest and is an introduced species) and we had owner permission for collecting on private land. Beetles were maintained in laboratory colonies using established methods . We collected offspring from females setup in separate low-density breeding containers . We transferred larvae in their first or second larval instar from their maternal brood ball to fresh dung in 12-well cell culture plates . Dung in transfer plates had been processed to be similar to maternally processed dung by pressing out as much water as possible using a cheesecloth and paper towels. Once in the 12-well plates, larvae were monitored daily and their transition to third instar recorded for subsequent aging.
FOXO Cloning and Sequencing
A previous EST library for Onthophagus taurus had identified a 704 base pair fragment that matched to FOXO (; GenBank accession FG540767.1). We used this sequence to design primers to first clone larger fragments of FOXO in O. taurus and then to clone FOXO in our focal species, O. nigriventris. Based on these two sequences, we designed a series of primers (see Table S2) that were able to clone fragments of FOXO from several species available for comparison (O. taurus, nigriventris, binodis, sagittarius, hecate, and pennsylvanicus). PCR fragments were cloned into a pSC-A vector with the Strataclone PCR Cloning kit (Stratagene/Agilent; Santa Clara, CA). Clones were sequenced using M13 primers and BigDye PCR reactions (see ). Partial sequences of all species considered (N = 6) have been submitted to GenBank (accession HQ605917-23).
FOXO Expression in Untreated Individuals
A previous microarray experiment  suggested that FOXO was differentially expressed between the thoracic and abdominal epidermis in male O. nigriventris. We wished to validate this result using real-time quantitative PCR. In particular, we were interested in the tissue-specific scaling of FOXO gene expression with body size, in both males and females. Paralleling the methods of previous microarray analyses, we harvested tissue from first-day pupae of O. nigriventris. Tissue was harvested from the prothoracic epidermis, which includes the developing horn and the surrounding prothorax. Tissue was also collected from the dorsal abdominal epidermis, avoiding the small lateral projections common in these pupae. The abdomen sample, while containing mostly epidermal tissue, also likely contained minimal amounts of connected adjacent tissue, including muscle and fat body cells.
Tissue was harvested from 21 males and 19 females. All dissections were performed in 1× RNase-free PBS (Ambion/Applied Biosystems, Austin, TX), under RNase-free conditions: all dissecting tools were treated with RNase-Zap (Ambion/Applied Biosystems, Austin, TX). Immediately after removal, tissue was placed in 350 µl Buffer RLT 1% v/v BME (RNeasy Mini Kit, Qiagen, Valencia, CA). Tissue was ground using a sterile, RNase-free pestle fit to the 1.5 µl microcentrifuge tube (Kontes grinders, Kimble Chase, VWR, West Chester, PA) and immediately frozen in liquid nitrogen before being stored at −70 C until RNA extraction.
RNA extraction and reverse transcription.
RNA was extracted from individual tissue samples using the RNeasy Mini Kit (Qiagen, Valencia, CA), following standard kit protocols. The optional on-column DNA digest was performed (using the Qiagen RNase-free DNase set) to reduce any genomic DNA contamination during the qPCR analyses. Total RNA was eluted in 25 µl RNase-free water (Qiagen, Valencia, CA), and quantified using a Nanodrop (Thermo Scientific, Franklin, MA). Extracted RNA had an average yield of 7.07 µg, and an average purity of 260/230 = 1.94, 260/280 = 2.14. Total RNA was stored at −70°C until further processing for qPCR analyses.
Total RNA was converted to cDNA using the Quantitect Reverse Transcription Kit (Qiagen; Valencia, CA) and standard kit protocols. Briefly, 250 ng total RNA was incubated with 2 ul gDNA wipeout buffer at 42°C for 2 minutes before being combined with Quantiscript reverse transcriptase, 5× buffer, and primer mix and incubated at 42°C for 15 minutes, followed by 85°C for 3 minutes. cDNA was quantified using a Nanodrop 1000 Spectrophotometer (Thermo Scientific, Franklin, MA) and stored at −20 until the qPCR was performed. cDNA had an average yield of 1022 ng/ul, 260/280 ratio of 1.81 and 260/230 ratio of 2.24.
Real-time quantitative PCR (qPCR) was performed using SYBR green PCR kit (Qiagen, Valencia, CA) and standard kit protocols. Prior to running our samples, we executed a minimum of five preliminary qPCRs to test a range of primers, control genes and test conditions. We designed primers for qPCR using Oligoanalyzer and PrimerQuest/Primer3 (Integrated DNA Technologies, Coralville, IA). Primers were tested at two concentrations over a four-fold dilution series of cDNA (1000 ng/ul, 100 ng/ul, 10 ng/ul, 1 ng/ul). Based on the Ct data for these standard curves and the dissociation curves from these runs, we chose one primer pair for FOXO (for primer sequences, see Table S2).
We used published microarray data  to identify possible control genes for our qPCR analyses. Both actin and GAPDH showed low variability in expression across tissues, sexes, and male sizes (mean (SD) in expression (A) was 9.0 (0.28) and 12.5 (0.31) for GAPDH and actin, respectively). We tested three primer pairs each (at two concentrations) for both actin and GAPDH, across a four-fold dilution series. We chose to focus on actin (as opposed to GAPDH) as a control gene because it showed the most robust amplification across the four-fold dilution series (R2 of standard curve generally 0.99) and consistent dissociation curves. Final primers for actin were based off of O. nigriventris actin sequences (GenBank accession HQ605924, cloned using primers designed off of the O. taurus actin sequence GenBank accession FG541406.1). The appropriateness of actin as a control gene was further verified by microarrays that directly hybridized tissues of male Onthophagus taurus of varying size . We observed no differential expression of actin, thus meeting a critical requirement for being able to compare FOXO expression across individuals of different sizes.
All samples of a particular tissue from a particular sex were run simultaneously (and analyzed together, e.g., female thorax and male abdomen) over four qPCR runs. Based on previous primer optimizations, we ran 1.2 ul of 10 uM actin primers with 250 ng amount of cDNA while 0.6 ul of 10 uM FOXO primers were run with 250 ng of cDNA. Standard curves (over a 4-fold dilution), no-template controls and no-reverse transcription controls were included for each primer pair for each qPCR run. Quantitative real time PCR was performed with a Stratagene MX3000P system (Stratagene/Agilent; Santa Clara, CA); SYBR green and ROX (the reference dye) fluorescent measurements were taken every cycle for 45 cycles. PCR settings were as follows: 95°C for 15 min; 45 cycles of 94°C for 15 second, 57°C for 30 second and 72°C for 30 seconds; followed by a 55°C–95°C dissociation curve. A small subset of samples (3 female abdomen samples) failed to amplify and were dropped from the final analysis.
qPCR data analysis.
We focused on the crossing threshold (referred to as Ct from here-on), for all qPCR analyses. Because we hypothesized FOXO to vary with body size and nutrition, we included body size in all analyses. We controlled for the amount of tissue harvested by using a control housekeeping gene (actin) in all analyses . We used an ANOVA to measure the effects of body size (thorax width) and actin expression (both independent variables) on FOXO expression. This represents an extension of the multiple regression methods developed by Yuan et al  where we are focused on the effect of an additional variable (body size) which is hypothesized to affect gene expression (similar to other model-based approaches to qPCR data that can control for a range of variables, e.g., ).
In situ hybridization.
In situ-hybridizations were performed as described previously . DIG-labeled probes were constructed for sense and antisense strands of the same non-homeodomain FOXO fragment cloned for RNAinterference, described below (see primers used for cloning in Table S2). Probes were synthesized using using MEGAscript High Yield Transcription Kits (Ambion/Applied Biosystems, Austin, TX). Hybridizations were performed using sagittal sections of pupae. Unlike larval sections, high quality pupal sections are exceedingly difficult to obtain in pupal Onthophagus due to the great disparity in densities across tissue types, and we thus regard in situs as supplemental to other data. We identified putative tissue types through comparisons across sections and using DAPI staining.
FOXO RNAi Knockdown
We manipulated FOXO expression using RNA interference. While it would be ideal to test the effects of both downregulation and upregulation of FOXO, current tools available in Onthophagus allow permit the former approach.
FOXO probe generation.
Based on the sequence data generated for several Onthophagus species, we tested two primer pairs that would amplify a 250–400 bp sequence outside of the conserved forkhead domain. We chose a pair that spanned a 380 bp range following the conserved forkhead domain (see Table S2 for primer information).
Double-stranded RNA was produced by first using the RNAi FOXO primers to amplify and clone the relevant section of FOXO into a vector using the pSC-A vector with a Strataclone PCR Cloning kit (Stratagene/Agilent, Santa Clara, CA, USA). M13 primers (Integrated DNA Technologies, Coralville, IA) were used to amplify the relevant fragment, which was gel-extracted, using the Qiaquick Gel Extraction kit (Qiagen, Valencia, CA). We used the MEGAscript kit (and associated protocol; Ambion/Applied Biosystems, Austin, TX) to synthesize RNA. Briefly, an in-vitro transcription reaction was run overnight at 37°C using 1 ug of PCR product, NTP mix, reaction buffer and either the T7 or T3 enzyme mix. Control dsRNA was generated using the same methodology applied to vectors with no inserted PCR product. Template DNA was removed by adding 1 ul Turbo DNase (37°C for 15 minutes); RNA was recovered by precipitating with 30 ul LiCl (and 30 ul water and chilling at −20°C for 4 hours). Concentration of product was verified using a Nanodrop (average = 3.57 ug.ul). Complementary RNA strands were annealed by mixing equal amounts of RNA (from with T7 and T3 reactions), heating to 80°C for 5 minutes and then slowly cooling to room temperature over 4 hours in an insulated beaker. dsRNA was brought to 3 ul with injection buffer (5 mM KCl, 1 mM KPO4 pH 6.9) and stored at −70°C until injection.
Treatment individuals were injected with either 0.5, 1.0, 1.75, or 2.5 ug of double stranded FOXO RNA, suspended in 3 ul of injection buffer. Control individuals were injected with 1.0 ug of double-stranded vector RNA suspended in 3 ul of injection buffer. Beetles were injected between day 6 and 10 of the third instar (mean (SD) = 8.31 (1.47)); variation in time of injection did not affect body size or horn-body size scaling. Injections occurred under RNase-free conditions using a gas-tight 1801 Hamilton syringe with a 32-gauge needle. Larvae were injected in the anterior, dorsal abdomen, just under the cuticle.
We measured several phenotypic traits of larvae, pupae and adults. We used pupal thorax width, pupal mass and adult thorax width as measures of body size. The length of the third ( = last) instar, measured in days, was used as an estimate of development time. The third instar constitutes the dominant feeding stage of onthophagine larvae. We measured thoracic horn length (in pupae and adults) as described in . Genitalia investment of males was estimated by first dissecting out the copulatory organ (aedeagus) and then measuring the length (in lateral view) of the paramere (following description in Supplementary material of ). We also measured larval mass in a subset of individuals at three time points prior to injection to ensure that any observed effects of knockdown on size were not due to differences in establishment of treatment groups: no differences between control and treatment animals in average larval mass prior to injection were found (F1,39 = 0.02, P = 0.88).
We validated our RNAi knockdown using real-time quantitative PCR. Dissection and qPCR methods were identical to those described above for untreated individuals. We compared gene expression in the prothoracic epidermis and dorsal abdominal epidermis of 11 control-injected and 9 FOXO-knockdown individuals (all 11 individuals were 1 ug dsRNA dosage). We used an ANOVA to test for significant differences between control and knock-down individuals, which allowed us to control for differences in body size among individuals. We also estimated the degree of knockdown (fold change) using the methods of Pfaffl  by first size-matching control and treatment individuals, and then accounting for differences in efficiency between the control and treatment genes (e.g., for the abdominal tissue run, Eactin = 0.91; EFOXO = 1.16).
We fit a sigmoidal curve based on a 4-parameter Hill equation to male body size-horn scaling data using Sigma Plot 2001. This equation was used to calculate expected pupal or adult horn length for a given body size (thorax width). Residual horn length was calculated as the difference between observed and expected horn length for a given body size. For body size-genitalia scaling relationships we used standard ANOVAs instead of reduced major axis methods (errors in variable statistics) sometimes used in allometries . We chose this method because we were interested in relative differences in slopes rather than absolute slope values. Furthermore, the measurement error associated with body size (the x variable) is negligible as thorax width is a highly repeatable measure, making the use of RMA unnecessary .
FOXO protein sequences of five Onthophagus species compared to Tribolium castaneum. Shown are the first 300 amino acids of the gene. The conserved forkhead domain lies approximately between amino acids 80 and 170 (highlighted with the light dotted line). The gene region that made up the RNAi and in situ probe follows the conserved domain (highlighted in heavy dotted line). Darker grays indicate more conserved amino acids.
Validation of FOXO RNAi knockdown. Shown are least square means from an ANOVA that included expression of the control gene (actin) and body size (pupal mass) of control-injected and knockdown individuals harvested as first-day pupae (N = 20). Gene expression was measured in the prothoracic epidermis and the abdominal epidermis.
Effect of degree of FOXO knockdown on body size. Shown are results of linear regressions considering the effect of FOXO knockdown (0.5, 1.0, 1.75, or 2.5 ug of dsRNA injected) on body size. In this analysis, control individuals were treated as “0.”
Primers. Shown are primers used to clone multiple fragments of FOXO from several species of Onthophagus; primers used in qPCR analyses; and primers used to clone a non-homeodomain fragment of FOXO from O. nigriventris to be used in RNAi and in situs. Base pair regions are with reference to this original FOXO sequence (GenBank accession FG540767.1; the coding region starts at base pair 366) and actin sequence (GenBank accession FG541406.1). A series of primers (sequences available upon request) were used to clone larger fragments of FOXO in the original species (taurus) and the focal species (nigriventris) before designing the primers reported below, some of which (in particular, the cloning primers) worked in a range of species to clone out fragments of FOXO.
We thank H. Parzer, W. Anderson, E. Yoder, D. Rose, and B. Wasik for assistance with aspects of data collection and beetle care. A. Cash, B. Blackman and C. LeBlanc provided assistance with experimental design. Comments from several anonymous reviewers helped to improve earlier drafts of this manuscript. We are grateful to the Center for Genomics and Bioinformatics at Indiana University and its staff, especially John Colbourne and Zhao Lai, for use of qPCR facilities.
Conceived and designed the experiments: ESR APM. Performed the experiments: ESR. Analyzed the data: ESR APM. Contributed reagents/materials/analysis tools: APM. Wrote the paper: ESR APM.
- 1. Nijhout H (2003) The control of body size in insects. Dev Biol 261: 1–9.
- 2. Barthel A, Schmoll D, Unterman TG (2005) FOXO proteins in insulin action and metabolism. Trends Endocrin Met 16: 183–189.
- 3. Edgar BA (2006) How flies get their size: genetics meets physiology. Nat Rev Genet 7: 907–916.
- 4. Mirth CK, Riddiford LM (2007) Size assessment and growth control: how adult size is determined in insects. Bioessays 29: 344–355.
- 5. Brogiolo W, Stocker H, Ikeya T, Rintelen F, Fernandez R, et al. (2001) An evolutionarily conserved function of the Drosophila insulin receptor and insulin-like peptides in growth control. Curr Biol 11: 213–221.
- 6. Britton JS, Lockwood WK, Li L, Cohen SM, Edgar BA (2002) Drosophila's insulin/P13-kinase pathway coordinates cellular metabolism with nutritional conditions. Dev Cell 2: 239–249.
- 7. Ikeya T, Galic M, Belawat P, Nairz K, Hafen E (2002) Nutrient-dependent expression of insulin-like peptides from neuroendocrine cells in the CNS contributes to growth regulation in Drosophila. Curr Biol 12: 1293–1300.
- 8. Rulifson EJ, Kim SK, Nusse R (2002) Ablation of insulin-producing neurons in flies: Growth and diabetic phenotypes. Science 296: 1118–1120.
- 9. Zhang H, Liu JN, Li CR, Momen B, Kohanski RA, et al. (2009) Deletion of Drosophila insulin-like peptides causes growth defects and metabolic abnormalities. P Natl Acad Sci USA 106: 19617–19622.
- 10. Junger MA, Rintelen F, Stocker H, Wasserman JD, Vegh M, et al. (2003) The Drosophila forkhead transcription factor FOXO mediates the reduction in cell number associated with reduced insulin signaling. J Biol 2: 20.
- 11. Kramer JM, Davidge JT, Lockyer JM, Staveley BE (2003) Expression of Drosophila FOXO regulates growth and can phenocopy starvation. BMC Dev Biol 3: 5.
- 12. Puig O, Marr MT, Ruhf ML, Tjian R (2003) Control of cell number by Drosophila FOXO: downstream and feedback regulation of the insulin receptor pathway. Gene Dev 17: 2006–2020.
- 13. Becker T, Loch G, Beyer M, Zinke I, Aschenbrenner AC, et al. (2010) FOXO-dependent regulation of innate immune homeostasis. Nature 463: 369–373.
- 14. van der Horst A, Burgering BMT (2007) Stressing the role of FOXO proteins in lifespan and disease. Nat Rev Mol Cell Bio 8: 440–450.
- 15. Giannakou ME, Goss M, Junger MA, Hafen E, Leevers SJ, et al. (2004) Long-lived Drosophila with overexpressed dFOXO in adult fat body. Science 305: 361–361.
- 16. Hwangbo DS, Gersham B, Tu MP, Palmer M, Tatar M (2004) Drosophila dFOXO controls lifespan and regulates insulin signalling in brain and fat body. Nature 429: 562–566.
- 17. Flatt T, Min KJ, D'Alterio C, Villa-Cuesta E, Cumbers J, et al. (2008) Drosophila germ-line modulation of insulin signaling and lifespan. P Natl Acad Sci USA 105: 6368–6373.
- 18. DiAngelo JR, Birnbaum MJ (2009) Regulation of fat cell mass by insulin in Drosophila melanogaster. Mol Cell Biol 29: 6341–6352.
- 19. Puig O, Tjian R (2005) Transcriptional feedback control of insulin receptor by dFOXO/FOXO1. Gene Dev 19: 2435–2446.
- 20. Teleman AA (2010) Molecular mechanisms of metabolic regulation by insulin in Drosophila. Biochem J 425: 13–26.
- 21. Moczek AP, Emlen DJ (2000) Male horn dimorphism in the scarab beetle, Onthophagus taurus: do alternative reproductive tactics favor alternative phenotypes? Anim Behav 59: 459–466.
- 22. Aubin-Horth N, Dodson JJ (2004) Influence of individual body size and variable thresholds on the incidence of a sneaker male reproductive tactic in Atlantic salmon. Evolution 58: 136–144.
- 23. O'Donnell S (1998) Reproductive caste determination in eusocial wasps (Hymenoptera: Vespidae). Annu Rev Entomol 43: 323–346.
- 24. Smith CR, Toth AL, Suarez AV, Robinson GE (2008) Genetic and genomic analyses of the division of labour in insect societies. Nat Rev Genet 9: 735–748.
- 25. Gilbert SF, Epel D (2009) Ecological developmental biology: integrating epigenetics, medicine, and evolution. Sunderland, Mass: Sinauer Associates. 459 p.
- 26. Nijhout HF, Grunert LW (2002) Bombyxin is a growth factor for wing imaginal disks in Lepidoptera. P Natl Acad Sci USA 99: 15446–15450.
- 27. Shingleton AW, Frankino WA, Flatt T, Nijhout HF, Emlen DJ (2007) Size and shape: the developmental regulation of static allometry in insects. Bioessays 29: 536–548.
- 28. Hietakangas V, Cohen SM (2009) Regulation of tissue growth through nutrient sensing. Annu Rev Genet 43: 389–410.
- 29. Emlen DJ, Szafran Q, Corley LS, Dworkin I (2006) Insulin signaling and limb-patterning: candidate pathways for the origin and evolutionary diversification of beetle ‘horns’. Heredity 97: 179–191.
- 30. Tang H, Smith-Caldas M, Driscoll MV, Salhadar S, Shingleton AW (2011) FOXO regulates organ-specific phenotypic plasticity in Drosophila. Plos Genet 7(11): e1002373.
- 31. Eberhard WG (2009) Static allometry and animal genitalia. Evolution 63: 48–66.
- 32. Shingleton AW, Das J, Vinicius L, Stern DL (2005) The temporal requirements for insulin signaling during development in Drosophila. Plos Biol 3: 1607–1617.
- 33. Toth AL, Varala K, Newman TC, Miguez FE, Hutchison SK, et al. (2007) Wasp gene expression supports an evolutionary link between maternal behavior and eusociality. Science 318: 441–444.
- 34. Corona M, Velarde RA, Remolina S, Moran-Lauter A, Wang Y, et al. (2007) Vitellogenin, juvenile hormone, insulin signaling, and queen honey bee longevity. P Natl Acad Sci USA 104: 7128–7133.
- 35. Ament SA, Corona M, Pollock HS, Robinson GE (2008) Insulin signaling is involved in the regulation of worker division of labor in honey bee colonies. P Natl Acad Sci USA 105: 4226–4231.
- 36. de Azevedo SV, Hartfelder K (2008) The insulin signaling pathway in honey bee (Apis mellifera) caste development - differential expression of insulin-like peptides and insulin receptors in queen and worker larvae. J Insect Physiol 54: 1064–1071.
- 37. Miyakawa H, Imai M, Sugimoto N, Ishikawa Y, Ishikawa A, et al. (2010) Gene up-regulation in response to predator kairomones in the water flea, Daphnia pulex. BMC Dev Biol 10: 45.
- 38. Emlen DJ (1994) Environmental control of horn length dimorphism in the beetle Onthophagus acuminatus (Coleoptera, Scarabaeidae). P Roy Soc B – Biol Sci 256: 131–136.
- 39. Moczek AP, Emlen DJ (1999) Proximate determination of male horn dimorphism in the beetle Onthophagus taurus (Coleoptera: Scarabaeidae). J Evolution Biol 12: 27–37.
- 40. Emlen DJ (1997) Alternative reproductive tactics and male-dimorphism in the horned beetle Onthophagus acuminatus (Coleoptera:Scarabaeidae). Behav Ecol Sociobiol 41: 335–341.
- 41. Emlen DJ (2001) Costs and the diversification of exaggerated animal structures. Science 291: 1534–1536.
- 42. Moczek AP, Nijhout HF (2004) Trade-offs during the development of primary and secondary sexual traits in a horned beetle. Am Nat 163: 184–191.
- 43. Simmons LW, Emlen DJ (2006) Evolutionary trade-off between weapons and testes. P Natl Acad Sci USA 103: 16346–16351.
- 44. Simmons LW, Tomkins JL, Hunt J (1999) Sperm competition games played by dimorphic male beetles. P Roy Soc B – Biol Sci 266: 145–150.
- 45. Bohni R, Riesgo-Escovar J, Oldham S, Brogiolo W, Stocker H, et al. (1999) Autonomous control of cell and organ size by CHICO, a Drosophila homolog of vertebrate IRS1–4. Cell 97: 865–875.
- 46. Snell-Rood EC, Cash A, Han MV, Kijimoto T, Andrews A, et al. (2011) Developmental decoupling of alternative phenotypes: insights from the transcriptomes of horn-polyphenic beetles. Evolution 65: 231–245.
- 47. Hardy RW, Tokuyasu KT, Lindsley DL, Garavito M (1979) Germinal proliferation center in the testis of Drosophila melanogaster. J Ultra Mol Struct R 69: 180–190.
- 48. Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29: 6.
- 49. De Moed GH, Kruitwagen C, De Jong G, Scharloo W (1999) Critical weight for the induction of pupariation in Drosophila melanogaster: genetic and environmental variation. J Evolution Biol 12: 852–858.
- 50. Moczek AP, Nijhout HF (2002) Developmental mechanisms of threshold evolution in a polyphenic beetle. Evol Dev 4: 252–264.
- 51. Rowe L, Houle D (1996) The lek paradox and the capture of genetic variance by condition dependent traits. P Roy Soc B – Biol Sci 263: 1415–1421.
- 52. Bonduriansky R (2007) The evolution of condition-dependent sexual dimorphism. Am Nat 169: 9–19.
- 53. Snell-Rood E, Moczek A (2012) Horns and the role of development in the evolution of beetle contests. In: Hardy ICW, Briffa M, editors. Animal Contests. Cambridge: Cambridge University Press. In press.
- 54. Gershman B, Puig O, Hang L, Peitzsch RM, Tatar M, et al. (2007) High-resolution dynamics of the transcriptional response to nutrition in Drosophila: a key role for dFOXO. Physiol Genomics 29: 24–34.
- 55. Teleman AA, Hietakangas V, Sayadian AC, Cohen SM (2008) Nutritional control of protein biosynthetic capacity by insulin via myc in Drosophila. Cell Metab 7: 21–32.
- 56. van der Heide LP, Hoekman MFM, Smidt MP (2004) The ins and outs of FOXO shuttling: mechanisms of FOXO translocation and transcriptional regulation. Biochem J 380: 297–309.
- 57. Harvey KF, Mattila J, Sofer A, Bennett FC, Ramsey MR, et al. (2008) FOXO-regulated transcription restricts overgrowth of Tsc mutant organs. J Cell Biol 180: 691–696.
- 58. Parzer HF, Moczek AP (2008) Rapid antagonistic coevolution between primary and secondary sexual characters in horned beetles. Evolution 62: 2423–2428.
- 59. Emlen DJ, Nijhout HF (1999) Hormonal control of male horn length dimorphism in the dung beetle Onthophagus taurus (Coleoptera: Scarabaeidae). J Insect Physiol 45: 45–53.
- 60. Oostra V, de Jong MA, Invergo BM, Kesbeke F, Wende F, et al. (2011) Translating environmental gradients into discontinuous reaction norms via hormone signalling in a polyphenic butterfly. P Roy Soc B – Biol Sci 278: 789–797.
- 61. Parthasarathy R, Palli SR (2011) Molecular analysis of nutritional and hormonal regulation of female reproduction in the red flour beetle, Tribolium castaneum. Insect Biochem Molec 41: 294–305.
- 62. Iga M, Smagghe G (2011) Relationship between larval-pupal metamorphosis and transcript expression of insulin-like peptides and insulin receptor in Spodoptera littoralis. Peptides 32: 531–538.
- 63. Sutter NB, Bustamante CD, Chase K, Gray MM, Zhao KY, et al. (2007) A single IGF1 allele is a major determinant of small size in dogs. Science 316: 112–115.
- 64. Moczek AP, Nijhout HF (2003) Rapid evolution of a polyphenic threshold. Evol Dev 5: 259–268.
- 65. Moczek AP, Nagy LM (2005) Diverse developmental mechanisms contribute to different levels of diversity in horned beetles. Evol Dev 7: 175–185.
- 66. Shafiei M, Moczek AP, Nijhout HF (2001) Food availability controls the onset of metamorphosis in the dung beetle Onthophagus taurus (Coleoptera: Scarabaeidae). Physiol Entomol 26: 173–180.
- 67. Kijimoto T, Costello J, Tang Z, Moczek AP, Andrews J (2009) EST and microarray analysis of horn development in Onthophagus beetles. BMC Genomics 10: 504.
- 68. Moczek AP, Rose , DJ (2009) Differential recruitment of limb patterning genes during development and diversification of beetle horns. P Natl Acad Sci USA 106: 8992–8997.
- 69. Yuan JS, Reed A, Chen F, Stewart CN (2006) Statistical analysis of real-time PCR data. BMC Bioinformatics 7: 85.
- 70. McCairns RJS, Bernatchez L (2010) Adaptive divergence between freshwater and marine sticklebacks: insights into the role of phenotypic plasticity from an integrated analysis of candidate gene expression. Evolution 64: 1029–1047.
- 71. Moczek AP, Rose D, Sewell W, Kesselring BR (2006) Conservation, innovation, and the evolution of horned beetle diversity. Dev Genes Evol 216: 655–665.
- 72. Moczek AP (2006) Pupal remodeling and the development and evolution of sexual dimorphism in horned beetles. Am Nat 168: 711–729.
- 73. LaBarbera M (1989) Analyzing body size as a factor in ecology and evolution. Annu Rev Ecol Syst 20: 97–117.
- 74. McArdle BH (1988) The structural relationship-regression in biology. Can J Zoolog 66: 2329–2339.