Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Nutritional Geometric Profiles of Insulin/IGF Expression in Drosophila melanogaster

  • Stephanie Post ,

    stephanie_post@brown.edu

    Affiliation Department of Molecular Biology, Cell Biology and Biochemistry, Providence, RI, Brown University, United States of America

  • Marc Tatar

    Affiliations Department of Molecular Biology, Cell Biology and Biochemistry, Providence, RI, Brown University, United States of America, Department of Ecology and Evolutionary Biology, Providence, RI, Brown University, United States of America

Nutritional Geometric Profiles of Insulin/IGF Expression in Drosophila melanogaster

  • Stephanie Post, 
  • Marc Tatar
PLOS
x

Abstract

Insulin/IGF signaling (IIS) in Drosophila melanogaster is propagated by eight Drosophila insulin-like peptides (dilps) and is regulated by nutrition. To understand how dietary protein and sugar affect dilp expression, we followed the analytical concepts of the Nutritional Geometric Framework, feeding Drosophila adults media comprised of seven protein-to-carbohydrate ratios at four caloric concentrations. Transcript levels of all dilps and three IIS-regulated genes were measured. Each dilp presented a unique pattern upon a bivariate plot of sugar and protein. Dilp2 expression was greatest upon diets with low protein-to-carbohydrate ratio regardless of total caloric value. Dilp5 expression was highly expressed at approximately a 1:2 protein-to-carbohydrate ratio and its level increased with diet caloric content. Regression analysis revealed that protein-to-carbohydrate ratio and the interaction between this ratio and caloric content significantly affects dilp expression. The IIS-regulated transcripts 4eBP and InR showed strikingly different responses to diet composition: 4eBP was minimally expressed except when elevated at low caloric diets. InR expression increased with protein level, independent of caloric content. Values of published life history traits measured on similar diets revealed correlations between egg production and the expression of dilp8 4eBP, while low protein-to-carbohydrate ratio diets associated with long lifespan correlated with elevated dilp2. Analyzing how nutient composition associates with dilp expression and IIS reveals that nutritional status is modulated by different combinations of insulin-like peptides, and these features variously correlate to IIS-regulated life history traits.

Introduction

Insulin/Insulin-like Growth Factor signaling (IIS) in the fruit fly Drosophila melanogaster is mediated by eight Drosophila insulin-like peptides (dilps) that signal through a common tyrosine kinase receptor INR (insulin/IGF receptor). Dilps and InR are homologous to insulin, insulin-like growth factor and their respective receptors in mammals [1]. Ligand-activated INR interacts with insulin receptor substrate IRS (chico, homolog of human IRS1-4) to initiate canonical PI3K and Akt signaling, and subsequently repress the forkhead transcription factor dFOXO [2]. The eight dilps are variously expressed across the life cycle, development and tissues [1,3]. Messenger RNA of dilp1, dilp2, dilp3 and dilp5 are predominantly expressed in median neurosecretory cells (MNCs; insulin producing cells, IPC) of the adult brain. Dilp5 mRNA is also produced in adult ovarian follicles and renal tubules while dilp3 is expressed in the midgut [1,4,5]. Dilp6 mRNA is produced in adult and larval fat body, a tissue with adipose and liver-like functions [6,7]. Dilp4 is expressed in embryo mesoderm [3]. Dilp7 is expressed in the larval and adult central nervous system [3] and dilp8 has been observed during pupal development [8,9].

Flies mutant for specific dilp loci have been used to explore their growth and metabolic functions, although interpreting outcomes is complicated by compensatory increase or decrease of various dilps when one locus is mutated. DILP2 peptide is inferred to modulate circulating carbohydrates because dilp2 mutants have elevated hemolymph sugar [10]. A similar phenotype was reported from flies where the MNCs were ablated and rescue was subsequently achieved by exogenous expression of dilp2 [11]. Dilp2 has notably been associated with regulation of longevity. Dilp2 mRNA and peptide are reduced in genetic manipulations that extend adult lifespan [7,12] and lifespan is extended in dilp2 mutant adults [10]. Whether dilp2 directly controls these phenotypes remains somewhat uncertain because mutation of dilp2 simultaneously increases dilp3 and dilp5 expression [10,13]. Mutant flies that lack dilp2, dilp3 and dilp5 together forestall this compensatory expression: homozygote mutants no longer show extended lifespan, although heterozygote animals are slightly long-lived [10].

The functions of dilp6 have been elucidated by analyzing mutants and with over-expression. Dilp6 is critical for larval development and responds to the maturation hormone ecdysone [14]. Dilp6 null mutants have slightly elevated lipid levels, suggesting that dilp6 controls lipid storage and use [10]. In adults, overexpressing dilp6 extends lifespan and increases fat and glycogen [7]. However, whether these effects are directly caused by dilp6 is unknown because MSC production of dilp2 and dilp5 are reduced when dilp6 is overexpressed in fat bodies [7].

Given the complex compensatory expression among dilps upon mutation, here we sought to understand how these peptides are expressed in the physiological context of wildtype animals fed different diets. In early reports, starvation reduced larval dilp3 and dilp5 but not dilp2 [15]. With larvae and adults, dilp5 but not dilp2 was reduced when animals were maintained on yeast-restricted or all-component diluted diets [1618]. On the other hand, starvation increased dilp6 expression in larvae and adults while dilp2 and dilp5 were decreased or unchanged [6,7]. Together, these observations suggest that dilps uniquely mediate distinct metabolic roles: glucose metabolism by dilp2, lipid storage by dilp6, lipid metabolism by dilp3, and response to protein by dilp5 [19].

To explore this perspective, we measured all dilp mRNAs in adults fed diets that varied by protein-to-carbohydrate ratio at four levels of caloric content. This design follows the analytical approach of the Geometric Nutritional Framework to separate the impact of nutrient composition from caloric content upon continuous traits [2023]. In particular, we applied the dietary regimen of Lee et al. [24] where Drosophila adults were fed 28 diets of seven protein-to-carbohydrate ratios at four caloric concentrations. In that report, lifespan was maximized by a relatively low protein-to-carbohydrate (1:16) intake regardless of caloric intake, fecundity was maximized on a higher protein-to-carbohydrate ratio intake (1:2), and fitness was greatest at an intermediate protein-to-carbohydrate intake (1:4).

Although it is widely believed that caloric restriction modulates aging by reducing insulin/IGF signaling, evidence to support this theory is actually sparse [16,25,26]. Foxo mutants robustly extend lifespan when diet restricted [16]. Likewise, dilp mutants have little effect on the ability of DR to slow aging [10,17]. These counter-intuitive outcomes might be caused by the compensatory nature of dilp expression in mutant animals, and this further recommends that we study wildtype adults under different physiological conditions to test the hypothesis that insulin/IGF signaling is minimized upon diets where the ratio of protein to carbohydrate maximizes lifespan. Surprisingly, dilp2 mRNA is most elevated upon diets with low protein-to-carbohydrate ratios, contrary to expectation if reduced dilp2 is associated with longevity assurance. On the other hand, dilp5 expression is reduced on diets with low protein-to-carbohydrate ratios, while dilp8 is highly expressed on all diets except upon those with very low calorie level. Life history traits measured and reported for these diets reveal a positive correlation between egg production and expression of dilp8, a positive association between lifespan and expression of dilp1 and dilp2, and a weak negative correlation between dilp5 expression and lifespan.

Materials and Methods

Fly husbandry and nutritional geometry design

Outbred flies of the stock ywR were maintained and reared at 25°C, 40% relative humidity and 12h light/dark cycle. Flies were reared on agar-based diet with cornmeal (5.2%), sucrose (11.0%), autolyzed yeast (2.5%; SAF brand) and agar (0.79%) (w/v in 100 mL water) with 0.2% Tegosept (methyl4-hydroxybenzoate, Sigma, St Louis, MO, USA) as an antifungal agent. After eclosion, flies were mated for two days after which females were separated and placed on a series of 28 diets (day 0) (Table 1). These diets used the ratios and energetic content as reported by Lee et al. [24], but here the nutrients were provided in solid agar-based media rather than liquid food. Yeast extract (MP Biomedical) and sucrose were combined with agar (0.79%) in the amounts detailed in Table 1. Ten females were put in each vial, three vials per diet. Flies were transferred to new vials at day 2 and day 4. The three biological replicates were pooled on day 5 and flies were homogenized in Trizol reagent using a TissueLyser (Qiagen). Gene expression from pooled biological samples is estimated to average the gene expression of the separate biological samples, according to the biological averaging assumption [27,28].

Quantitative RT-PCR

Total RNA was extracted from whole flies (30 per sample) in Trizol (Invitrogen, Grand Island, NY, USA) and treated with Turbo DNase (Invitrogen). RNA was quantified with a NanoDrop ND-1000 (Thermo Fisher Scientific Inc., Wilmington, DE, USA) and reverse-transcribed with iScript cDNA synthesis (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Quantitative RT-PCR was conducted with SYBR Green PCR master mix (Applied Biosystems, Carlsbad, CA, USA) and measured on an ABI prism 7300 Sequence Detection System (Applied Biosystems). mRNA abundance was calculated by comparative CT relative to ribosomal protein 49 (RP49). Although there are concerns about choice of reference gene in diet studies, RP49/RPL32 is reported as the most stable option [29], and we verified that RP49 is not sensitive to diet by viewing and analyzing the raw CT values (data not shown). We therefore chose RP49 as the reference in all relative gene expression analyses. Primer sequences are listed in S3 Table.

Data Analysis

Expression data for each gene was normalized to its value on diet #7. Normalized values were visualized on landscapes using nonparametric thin plate splines in R using the “fields” package as performed previously [30]. For quantitative analysis, gene expression was treated as response variables in multivariate multiple regression conducted in R using lm and Manova functions, evaluating linear models (protein-to-carbohydrate ratio or caloric content) and nonlinear models (P:C ratio-by-calorie interaction, and ratio-by-ratio and calorie-by-calorie quadratic functions). Coefficients were compared among dilps by ANOVA (http://statpages.org/anova1sm.html). Post hoc pairwise tests were conducted by the Holm-Sidak student’s t-test using mean and standard error (R tsum.test function, “BSDA” package). Data for lifespan and egg production reported by Lee et al. [24] for females upon the same matrix of diets were compared to our current measures of gene expression by Pearson’s product moment correlation coefficient, adjusting t for sample size by t = (r-ρ)/√[(1-r2)/(n-2) to calculate significance with a two-tailed students t-test.

Results

Each dilp produces a distinct expression pattern across the diet landscape (Fig 1). Several cases show maximum expression along specific vectors of dietary protein-to-carbohydrate ratio, where the highest ridge of expression aligns with a particular ratio, denoted by gray lines. Notably, dilp2 mRNA is greatest at low protein diets where the protein-to-carbohydrate ratio is approximately 1:16 (Fig 1B). While the absolute levels of dilp1 mRNA are lower than dilp2 (S1 Table and S1 Fig), these two dilps show similar patterns across diets (Fig 1A, Table 2). In contrast, dilp5 (Fig 1E) is highly expressed at a protein-to-carbohydrate ratio of approximately 1:2, and upon diets of high caloric value, creating a rising ridge contour. Dilp3 (Fig 1C) is maximized on diets with low caloric content at a protein-to-carbohydrate ratio of approximately 1:8, and its pattern is moderately correlated to dilp2 expression (Table 2). Dilp6 mRNA (Fig 1F) is maximized at low protein, high calorie diets and generally decreases with increasing protein-to-carbohydrate ratio.

thumbnail
Fig 1. Dilp gene expression across the nutritional geometric framework surface.

Gene expression was plotted against protein and carbohydrate content of 28 diets by nonparametric thin plate splines in R. Gray lines represent the seven protein-to-carbohydrate ratios. Heat maps from blue to red represent the normalized level of gene expression from lowest to highest. A) Dilp1, B) Dilp2, C) Dilp3, D) Dilp4, E) Dilp5, F) Dilp6, G) Dilp7 and H) Dilp8.

https://doi.org/10.1371/journal.pone.0155628.g001

thumbnail
Table 2. Correlation coefficients among gene expression and life history traits.

https://doi.org/10.1371/journal.pone.0155628.t002

The expression patterns of other dilps are maximized or minimized in small regions of the dietary space. Dilp4 is greatest on high sugar, low protein and high caloric diets and uniformly expressed at low levels on other diets (Fig 1D). Dilp7 is expressed on most diets, except when it is reduced on very low calorie food (Fig 1G). The absolute level of dilp8 transcription was much higher than any other dilp gene (S1 Table) but showed a bifurcated pattern of expression similar to dilp4 when protein levels are low, and similar to dilp5 when protein levels are high (Fig 1H); overall, dilp8 is most correlated to dilp7 expression (Table 2).

Multivariate multiple regression analysis was used to evaluate the overall effect of individual diet variables (P:C ratio, caloric content), and their interaction (P:C ratio-by-caloric content) upon net, collective dilp expression (Table 3). Dilps and related insulin/IGF signaling factors (4eBP, InR and Upd2) were significantly affected by protein-to-carbohydrate ratio (linear and quadratic models) and by the P:C ratio-by-caloric content interaction (S2 Table; p = 0.026, p = 0.012 and p = 0.011 respectively). Caloric content was significantly associated with the overall expression of dilps in a quadratic model (S2 Table; p = 0.043). Considering the expression of each gene individually, protein-to-carbohydrate ratio linearly associated with dilp1, dilp2, dilp3 and Upd2, while caloric content affected dilp4, dilp7, dilp8, Upd2 and 4eBP (Table 3, p<0.05). Caloric content significantly affected expression of dilp5, dilp8 and 4eBP when fitted to a quadratic model (C×C, Table 3, p<0.05). Dilp1, dilp2, dilp8 and Upd2 were significantly affected by the P:C ratio-by-caloric content interaction (Table 3, p<0.05).

thumbnail
Table 3. Estimated model effects of protein-to-carbohydrate ratio and caloric content on insulin/IGF signaling gene expression.

https://doi.org/10.1371/journal.pone.0155628.t003

Among dilps, estimated parameters for protein-to-carbohydrate ratio, caloric content, and the interaction between P:C ratio and caloric content varied significantly (Fig 2). High protein-to-carbohydrate ratio (top panel) reduces dilp1, dilp2 and dilp3 expression, but increases expression of dilp5, dilp6, dilp7 and dilp8. The P:C ratio-by-caloric content interaction (bottom panel) has similar clusters: dilp1, dilp2 and dilp3 have a positive coefficient, and dilp5, dilp6, dilp7 and dilp8 have negative or negligible coefficients.

thumbnail
Fig 2. Protein-to-carbohydrate ratio, caloric content and P:C ratio-by-caloric content interaction affect dilp gene expression.

The multivariate multiple regression functions’ coefficients were compared between dilps by ANOVA. Protein-to-carbohydrate ratio (top panel): p = 0.006 among dilps; significant differences upon pairwise post hoc test for dilp3-dilp7 and dilp3-dilp8. Caloric content (middle panel): p = 0.004 among dilps; significant differences upon pairwise post hoc test for dilp1-dilp4, dilp2-dilp4 and dilp4-dilp6. P:C ratio-by-caloric content interaction (bottom panel): p = 0.003 among dilps; significant difference upon pairwise post hoc test for dilp1-dilp4 and dilp3-dilp4.

https://doi.org/10.1371/journal.pone.0155628.g002

Several genes associated with Drosophila IIS show distinct expression patterns with respect to diet. Unpaired-2 (Upd2) is a cytokine-like signaling molecule produced in Drosophila fat body [31] and midgut [32]. Fat body Upd2 was reported to increase on high sugar diets and to regulate the release of DILP hormone from the brain [31]. Here, Upd2 expression was elevated ~5-fold on low protein-to-carbohydrate diets and especially when caloric content was minimal (Fig 3; Excluding diet #28 as an outlier, S2 Fig, Grubbs/ESD test p<0.05). The Upd2 surface pattern correlates moderately with that of dilp1 and dilp2 (Table 2).

thumbnail
Fig 3. Upd2 expression is increased on low protein:carbohydrate ratio diets.

Gene expression was plotted against the protein and carbohydrate content of 28 diets by nonparametric thin plate splines in R. Gray lines represent the seven protein-to-carbohydrate ratios. Heat maps from blue to red represent the normalized level of gene expression from lowest to highest.

https://doi.org/10.1371/journal.pone.0155628.g003

DILP1-7 are proposed to signal through a common insulin/IGF receptor (INR), and thus repress the transcription factor dFOXO. 4eBP and InR are verified FOXO transcriptional targets [33,34], although 4eBP is also strongly regulated by the TORC1 related transcription factor REPTOR [35]. InR and 4eBP produce strikingly different expression surfaces on the dietary landscape (Fig 4). 4eBP mRNA is greatest on diets with low caloric content and when the protein-to-carbohydrate ratio is approximately 1:2. In contrast, InR mRNA increases with dietary protein, independent of caloric content. 4eBP expression was strongly yet inversely correlated with dilp8 mRNA (Table 2), while InR mRNA was moderately correlated with dilp1 (Table 2).

thumbnail
Fig 4. dFOXO transcriptional targets respond differently to diet composition.

Gene expression was plotted against the protein and carbohydrate content of 28 diets by nonparametric thin plate splines in R. Gray lines represent the seven protein-to-carbohydrate ratios. Heat maps from blue to red represent the normalized level of gene expression from lowest to highest.

https://doi.org/10.1371/journal.pone.0155628.g004

Discussion

Eight dilps are produced and are measurable in adult female Drosophila. To date, the expression of these dilps in response to nutrition has been quantified using diets that differ by a single component (yeast) or by diluting all components. In these studies, dilp5 expression was regulated by protein level while dilp2 and dilp3 mRNA were not affected by nutrients [16,17]. In contrast to single dimensional designs, Geometric Nutritional Framework analysis varies the proportion of nutritional components across defined ranges of caloric content. It provides a multidimensional approach to analyze how diet composition and quantity affect complex phenotypes such as longevity and gene expression. Geometric Framework analyses in Drosophila [24,36] as well as mammals [37] demonstrate that metabolism and longevity are modulated by particular ratios of protein to carbohydrate in the diet rather than by caloric content. The mechanism by which the ratio of protein to carbohydrate affects these traits is unknown, but is widely thought to involve the expression of insulin-like peptides.

We find with Drosophila that each dilp produces a distinctive expression surface relative to dietary protein-to-carbohydrate ratio and caloric content. The topography for dilp2 and dilp5 produce ridges of maximal expression at distinct protein-to-carbohydrate ratios of about 1:16 and 1:2 respectively. This pattern for dilp2 may be consistent with previous reports where dilp2 mRNA levels did not change in response to dietary yeast or total calories [16,17], because those studies used diets where dilp2 is fairly constant. In contrast, dilp5 is greatest on diets with a higher proportion of protein, and somewhat more so on diets with higher caloric content. This pattern may be consistent with studies that measured dilp5 expression when dietary yeast (protein) was increased from low to moderate levels [16,17]. Dilp2 and dilp5 thus respond to nutrients in qualitatively different ways.

These nutrient responses can be related to life history traits measured by Lee et al. [24] upon the same matrix of diets we employ in our current study. Although our wildtype strain (ywR) differs from that of Lee (Canton-S), meta-analysis of various Drosophila wildtype strains from different labs utilizing both liquid- and agar-based food delivery produced similar life history trait patterns mapped upon the Geometric Framework [24]. In Fig 5 we plot the egg production and lifespan data of Lee et al. [24] against the nutrient content of our common diets, noting that Lee originally plotted these traits relative to nutrient intake rather than to diet content. Based on diet content, longevity is maximized along the 1:16 protein-to-carbohydrate rail, as also seen when Lee et al. plotted this trait relative to nutrient intake. Here we now find that longevity and dilp2 expression are correlated in a modest but positive fashion (Table 2, R2 = 0.36).

thumbnail
Fig 5. Reported measures of lifespan and fecundity associate with expression of specific dilps.

Lifespan and fecundity data from Lee et al. [24] are plotted relative to diet nutrient content of 28 diets (used in Lee et al. and in the current study) by nonparametric thin plate splines in R. Gray lines represent seven protein-to-carbohydrate ratios. Heat maps from blue to red represent the phenotype value from lowest to highest.

https://doi.org/10.1371/journal.pone.0155628.g005

One potential way to understand this unexpected association considers how Drosophila use different food sources as adults. Drosophila need yeast-rich rotting fruit and vegetation to support egg production and larval development [38]. Such food sources are patchy in nature, and adults may fuel while searching by feeding upon common carbohydrate biased foods such as nectar [38]. In environments where protein is scarce, elevated dilp2 will occur in a nutrient landscape while adult Drosophila are searching for protein-rich sites, and may thus induce physiology appropriate for longevity assurance while foraging in a patchy environment.

We are aware that the observed positive correlation between longevity and dilp2 contrasts with studies where reduced dilp2 is associated with longevity [10,12,39]. This difference is not readily explained but we note that mutation of any one specific dilp changes the expression of other dilps. Compensatory expression among these genes could affect metabolism and lifespan and confound how we interpret their individual function.

In contrast to dilp2, dilp5 expression is greatest on diets where longevity is minimized and there is no correlation between the phenotypes). This is consistent with studies where reduced dietary yeast extend lifespan and simultaneously reduce dilp5 expression [17]. While current models point to dilp2 as most responsible for the control of aging, our observation suggests that dilp5 accelerates aging under normal physiological conditions.

The expression patterns of dilp1 and dilp2 are strongly correlate (Table 2), suggesting that they may share physiological functions. No functions in the adult fly have yet been attributed to dilp1, possibly because of redundancy with highly expressed dilp2. Both dilp1 and dilp2 correlate moderately with expression of Upd2. Upd2 is thought to non-cell autonomously regulate dilp2 in IPCs in response to nutrition [31]; by extension, Upd2 might also regulate dilp1. Dilp3 expression is greatest at a simple maximum at approximately 1:8 protein-to-carbohydrate in low calorie diets. Thus, the four dilps produced in brain MSC produce three distinct expression patterns with respect to dietary protein and carbohydrate composition.

Other adult dilps are produced outside of the brain. Expression of dilp6 is greatest on low protein-to-carbohydrate, high calorie diets. Previously, Skorupa et al. [36] observed increased triglyceride levels and adipose tissue volume with low protein-to-carbohydrate ratio and high sugar diets. High adiposity may elevate dilp6 expression because dilp6 is produced in fat body [6,7].

Although expressed in adults, dilp4, dilp7 and dilp8 are proposed to function predominantly in development [1,8,40,41]. However, because these dilp transcripts may occur as maternally deposited mRNAs in embryos within adult females, our interpretations of their adult function are tentative. Dilp4 and dilp7 are most abundant at extreme nutritional states, such as diets with very high protein-to-carbohydrate ratio or very high caloric content. These results suggest that dilp4 and dilp7 are not broadly sensitive to nutritional status.

In pupae, dilp8 delays metamorphosis through control of ecdysone during adverse conditions such as injured discs [9]. Here, adult dilp8 mRNA was surprisingly abundant on many diets (S1 Table). Notably, dilp8 expression was negatively correlated with 4eBP mRNA, which may reflect an inverse association with IIS signaling, or alternatively with 4eBP induction by REPTOR that occurs when TOR is repressed [35]. Expression of dilp8 was weakly correlated with egg production (Table 2). Dilp8 in adult ovaries may regulate reproduction, although caloric content itself may coordinately reduce both traits.

Longevity in Drosophila is readily manipulated by ablating insulin-producing cells and through mutations of the signaling pathway [10,4244]. Dietary restriction extends Drosophila lifespan [23], and Lee et al. [24] implicated protein-to-carbohydrate ratio as the specific, operative nutrient factor. Given these many observations, whether and how insulin-like peptides mediate the impact of diet upon longevity remains surprisingly unclear. Here we find that dilp1 and dilp2 are highly expressed at 1:16 protein-to-carbohydrate ratio, which presumably will inactive the pro-longevity FOXO transcription factor [12,45]. This ridge of expression correlates to elevated longevity in Lee et al. [24]. On the other hand, Skorupa et al. [36] demonstrated low caloric diets with a protein-to-carbohydrate close to 1:1 as most favorable for longevity. Such diets would occur in the lower left corner of our plots, where all dilp mRNA is reduced and 4eBP expression is greatest. More recently, the longevity response to P:C ratio was measured using a holidic diet [46]: survival was optimized at protein-to-carbohydrate ratios of approximately 1:2 or 1:4. These ratios correspond to diets we find to have low overall dilp expression, in which case reduced net expression of all dilps may be required for longevity assurance.

An emerging theme among geometric framework analyses of lifespan emphasizes the importance of protein-to-carbohydrate ratio rather than caloric content as the operative nutrient feature [24,36,37,46]. Our analysis suggests that the interaction between protein-to-carbohydrate ratio and caloric content affects insulin/IGF signaling, and the corresponding life history traits. The geometric framework analysis permits finer scale dissection of how dilps are regulated and correlated with respect to diet and each other, and how these patterns may be disrupted when a single dilp is mutated. Ultimately, studies combining mutant and Geometric Framework approaches [20] are needed to provide a thorough analysis of individual dilp functions in life history traits and metabolism.

Supporting Information

S1 Fig. Feeding on normal food, wildtype flies express ~100-fold more dilp2 than dilp1 mRNA.

https://doi.org/10.1371/journal.pone.0155628.s001

(DOCX)

S2 Fig. Upd2 expression along the geometric topography is hidden by extremely high expression at one outlier diet (Grubbs/ESD test p<0.05).

https://doi.org/10.1371/journal.pone.0155628.s002

(DOCX)

S1 Table. Relative gene expression values are the average of three technical replicates and are standardized to RP49 expression.

https://doi.org/10.1371/journal.pone.0155628.s003

(PDF)

S2 Table. Multivariate multiple regression MANOVA table.

https://doi.org/10.1371/journal.pone.0155628.s004

(DOCX)

Acknowledgments

We are grateful to Rochele Yamamoto and Aleksandra Norton for fly stocks and technical assistance; to Wenjing Zhang for dilp8 and Upd2 primers; to Hua Bai for helpful discussion and comments; and to Stephen Simpson for providing us with comparison data.

Author Contributions

Conceived and designed the experiments: SP MT. Performed the experiments: SP. Analyzed the data: SP. Contributed reagents/materials/analysis tools: SP. Wrote the paper: SP MT.

References

  1. 1. Brogiolo W, Stocker H, Ikeya T, Rintelen F, Fernandez R, Hafen E. An evolutionarily conserved function of the Drosophila insulin receptor and insulin-like peptides in growth control. Curr Biol. 2001;11(4):213–21. pmid:11250149.
  2. 2. Teleman AA. Molecular mechanisms of metabolic regulation by insulin in Drosophila. Biochem J. 2010;425(1):13–26. pmid:20001959.
  3. 3. Toivonen JM, Partridge L. Endocrine regulation of aging and reproduction in Drosophila. Mol Cell Endocrinol. 2009;299(1):39–50. pmid:18682271.
  4. 4. Veenstra JA, Agricola HJ, Sellami A. Regulatory peptides in fruit fly midgut. Cell Tissue Res. 2008;334(3):499–516. pmid:18972134.
  5. 5. Soderberg JA, Birse RT, Nassel DR. Insulin production and signaling in renal tubules of Drosophila is under control of tachykinin-related peptide and regulates stress resistance. PLoS One. 2011;6(5):e19866. pmid:21572965.
  6. 6. Slaidina M, Delanoue R, Gronke S, Partridge L, Leopold P. A Drosophila insulin-like peptide promotes growth during nonfeeding states. Dev Cell. 2009;17(6):874–84. pmid:20059956.
  7. 7. Bai H, Kang P, Tatar M. Drosophila insulin-like peptide-6 (dilp6) expression from fat body extends lifespan and represses secretion of Drosophila insulin-like peptide-2 from the brain. Aging Cell. 2012;11(6):978–85. pmid:22935001.
  8. 8. Colombani J, Andersen DS, Leopold P. Secreted peptide Dilp8 coordinates Drosophila tissue growth with developmental timing. Science. 2012;336(6081):582–85. pmid:22556251.
  9. 9. Hariharan IK. How growth abnormalities delay "puberty" in Drosophila. Sci Signal. 2012;5(229):pe27. pmid:22715466.
  10. 10. Gronke S, Clarke DF, Broughton S, Andrews TD, Partridge L. Molecular evolution and functional characterization of Drosophila insulin-like peptides. PLoS Genet. 2010;6(2):e1000857. pmid:20195512.
  11. 11. Rulifson EJ, Kim SK, Nusse R. Ablation of insulin-producing neurons in flies: growth and diabetic phenotypes. Science. 2002;296(5570):1118–20. pmid:12004130.
  12. 12. Hwangbo DS, Gershman B, Tu MP, Palmer M, Tatar M. Drosophila dFOXO controls lifespan and regulates insulin signalling in brain and fat body. Nature. 2004;429(6991):562–6. pmid:15175753.
  13. 13. Broughton S, Alic N, Slack C, Bass T, Ikeya T, Vinti G, et al. Reduction of DILP2 in Drosophila triages a metabolic phenotype from lifespan revealing redundancy and compensation among DILPs. PLoS One. 2008;3(11):e3721. pmid:19005568.
  14. 14. Okamoto N, Yamanaka N, Yagi Y, Nishida Y, Kataoka H, O'Connor MB, et al. A fat body-derived IGF-like peptide regulates postfeeding growth in Drosophila. Dev Cell. 2009;17(6):885–91. pmid:20059957.
  15. 15. Ikeya T, Galic M, Belawat P, Nairz K, Hafen E. Nutrient-dependent expression of insulin-like peptides from neuroendocrine cells in the CNS contributes to growth regulation in Drosophila. Curr Biol. 2002;12(15):1293–1300. pmid:12176357.
  16. 16. Min K, Yamamoto R, Buch S, Pankratz M, Tatar M. Drosophila lifespan control by dietary restriction independent of insulin-like signaling. Aging Cell. 2008;7(2):199–206. pmid:18221413.
  17. 17. Broughton SJ, Slack C, Alic N, Metaxakis A, Bass TM, Driege Y, et al. DILP-producing median neurosecretory cells in the Drosophila brain mediate the response of lifespan to nutrition. Aging Cell. 2010;9(3):336–46. pmid:20156206.
  18. 18. Okamoto N, Nishimura T. Signaling from Glia and Cholinergic Neurons Controls Nutrient-Dependent Production of an Insulin-like Peptide for Drosophila Body Growth. Dev Cell. 2015;35(3):295–310. pmid:26555050.
  19. 19. Kannan K, Fridell YW. Functional implications of Drosophila insulin-like peptides in metabolism, aging, and dietary restriction. Front Physiol. 2013;4:288. pmid:24137131.
  20. 20. Piper MD, Partridge L, Raubenheimer D, Simpson SJ. Dietary restriction and aging: a unifying perspective. Cell Metab. 2011;14(2):154–60. pmid:21803286.
  21. 21. Simpson SJ, Raubenheimer D. Caloric restriction and aging revisited: the need for a geometric analysis of the nutritional bases of aging. J Gerontol A Biol Sci Med Sci. 2007;62(7):707–13. pmid:17634316.
  22. 22. Simpson SJ, Raubenheimer D. Macronutrient balance and lifespan. Aging. 2009;1(10):875–80. pmid:20157561.
  23. 23. Tatar M, Post S, Yu K. Nutrient control of Drosophila longevity. Trends Endocrinol Metab. 2014;25(10):509–17. pmid:24685228.
  24. 24. Lee KP, Simpson SJ, Clissold FJ, Brooks R, Ballard JW, Taylor PW, et al. Lifespan and reproduction in Drosophila: New insights from nutritional geometry. Proc Natl Acad Sci U S A. 2008;105(7):2498–503. pmid:18268352.
  25. 25. Clancy DJ, Gems D, Hafen E, Leevers SJ, Partridge L. Dietary Restriction in Long-Lived Dwarf Flies. Science. 2002;296(5566):319. pmid:11951037.
  26. 26. Houthoofd K. Life extension via dietary restriction is independent of the Ins/IGF-1 signalling pathway in Caenorhabditis elegans. Experimental Gerontology. 2003;38(9):947–54. pmid:12954481.
  27. 27. Zhang SD, Gant TW. Effect of pooling samples on the efficiency of comparative studies using microarrays. Bioinformatics. 2005;21(24):4378–83. pmid:16234321.
  28. 28. Diz AP, Truebano M, Skibinski DO. The consequences of sample pooling in proteomics: an empirical study. Electrophoresis. 2009;30(17):2967–75. pmid:19676090.
  29. 29. Ponton F, Chapuis MP, Pernice M, Sword GA, Simpson SJ. Evaluation of potential reference genes for reverse transcription-qPCR studies of physiological responses in Drosophila melanogaster. J Insect Physiol. 2011;57(6):840–50. pmid:21435341.
  30. 30. Jensen K, McClure C, Priest NK, Hunt J. Sex-specific effects of protein and carbohydrate intake on reproduction but not lifespan in Drosophila melanogaster. Aging Cell. 2015;14(4):605–15. pmid:25808180.
  31. 31. Rajan A, Perrimon N. Drosophila cytokine unpaired 2 regulates physiological homeostasis by remotely controlling insulin secretion. Cell. 2012;151(1):123–37. pmid:23021220.
  32. 32. Osman D, Buchon N, Chakrabarti S, Huang YT, Su WC, Poidevin M, et al. Autocrine and paracrine unpaired signaling regulate intestinal stem cell maintenance and division. J Cell Sci. 2012;125(Pt 24):5944–9. pmid:23038775.
  33. 33. Puig O, Marr MT, Ruhf ML, Tjian R. Control of cell number by Drosophila FOXO: downstream and feedback regulation of the insulin receptor pathway. Genes Dev. 2003;17(16):2006–20. pmid:12893776.
  34. 34. Bai H, Kang P, Hernandez AM, Tatar M. Activin signaling targeted by insulin/dFOXO regulates aging and muscle proteostasis in Drosophila. PLoS Genet. 2013;9(11):e1003941. pmid:24244197.
  35. 35. Tiebe M, Lutz M, De La Garza A, Buechling T, Boutros M, Teleman AA. REPTOR and REPTOR-BP Regulate Organismal Metabolism and Transcription Downstream of TORC1. Dev Cell. 2015;33(3):272–84. pmid:25920570.
  36. 36. Skorupa DA, Dervisefendic A, Zwiener J and Pletcher SD. Dietary composition specifies consumption, obesity, and lifespan in Drosophila melanogaster. Aging Cell. 2008;7(4):478–90. pmid:18485125.
  37. 37. Solon-Biet SM, McMahon AC, Ballard JW, Ruohonen K, Wu LE, Cogger VC et al. The ratio of macronutrients, not caloric intake, dictates cardiometabolic health, aging, and longevity in ad libitum-fed mice. Cell Metab. 2014;19(3):418–30. pmid:24606899.
  38. 38. Good TP, Tatar M. Age-specific mortality and reproduction respond to adult dietary restriction in Drosophila melanogaster. J Insect Physiology. 2001;47(12):1467–73. pmid:12770153.
  39. 39. Sato-Miyata Y, Muramatsu K, Funakoshi M, Tsuda M, Aigaki T. Overexpression of dilp2 causes nutrient-dependent semi-lethality in Drosophila. Front Physiol. 2014;5:147. pmid:24795642.
  40. 40. Miguel-Aliaga I, Thor S, Gould AP. Postmitotic specification of Drosophila insulinergic neurons from pioneer neurons. PLoS Biol. 2008;6(3):e58. pmid:18336071.
  41. 41. Colombani J, Andersen DS, Boulan L, Boone E, Romero N, Virolle V, et al. Drosophila Lgr3 Couples Organ Growth with Maturation and Ensures Developmental Stability. Curr Biol. 2015;25(20):2723–9. pmid:26441350.
  42. 42. Broughton SJ, Piper MD, Ikeya T, Bass TM, Jacobson J, Driege Y, et al. Longer lifespan, altered metabolism, and stress resistance in Drosophila from ablation of cells making insulin-like ligands. Proc Natl Acad Sci U S A. 2005;102(8):3105–10. pmid:15708981.
  43. 43. Yamamoto R, Tatar M. Insulin receptor substrate chico acts with the transcription factor FOXO to extend Drosophila lifespan. Aging Cell. 2011;10(4):729–732. pmid:21518241.
  44. 44. Tatar M, Kopelman A, Epstein D, Tu MP, Yin CM, Garofalo RS. A mutant Drosophila insulin receptor homolog that extends life-span and impairs neuroendocrine function. Science. 2001;292(5514):107–10. pmid:11292875.
  45. 45. Villa-Cuesta E, Sage BT, Tatar M. A role for Drosophila dFoxO and dFoxO 5’UTR internal ribosome entry sites during fasting. PLoS One. 2010;5(7):e11521. pmid:20634900.
  46. 46. Lee KP. Dietary protein:carbohydrate balance is a critical modulator of lifespan and reproduction in Drosophila melanogaster: a test using a chemically defined diet. J Insect Physiol. 2015;75:12–9. pmid:25728576.