Research ArticleEVOLUTIONARY BIOLOGY

The genetic mechanism of selfishness and altruism in parent-offspring coadaptation

See allHide authors and affiliations

Science Advances  03 Jan 2020:
Vol. 6, no. 1, eaaw0070
DOI: 10.1126/sciadv.aaw0070
  • Fig. 1 Experimental design and expression heat map.

    (A) The transcriptomic screen for candidate genes underlying parent-offspring coadaptation was based on experimental manipulation of the presence and absence of egg attendance and posthatching parental care: No care (NC), eggs were removed upon the completion of oviposition. No nymph sample was used because of insufficient hatching success of untended eggs (fig. S1 and table S1). Egg care (EC), mothers tended their eggs for 20 days and were then sampled shortly before eggs hatched (2). Nymphs were kept for 6 days without tending females before they were sampled. Full care (FC), mothers tended their eggs until hatching and cared for their nymphs until 6 days after hatching when both were sampled. Antennae, head, abdomen, and ovaries from mothers and the whole body of nymphs were sampled in each treatment. On the basis of this screen, the expression of candidate genes was manipulated using in vivo RNAi to assess their causal effects on behavior and components of evolutionary fitness. (B) Venn diagram illustrating that only two genes with confirmed insect origin were differentially coexpressed in mothers and offspring when they behaviorally interacted according to our selection criterion (see main text). (C to G) Heat maps of differentially expressed genes in different tissues from mothers and the nymphs (PFDR < 0.01). Rows are genes, columns are samples. Samples were clustered according to expression patterns. We color-coded each sample in the horizontal bar above each panel and the clustering trees according to its treatment. FC samples are in yellow; EC samples are in light green, and NC samples are in dark green.

  • Fig. 2 Differential expression of Th and other dopamine-related genes in maternal tissues and in offspring.

    (A) Illustration of a known dopamine synthesis and degradation pathway catalyzed by tyrosine hydroxylase (Th) as the rate-limiting enzyme as well as DOPA decarboxylase (AADC) and DNAT (15, 17). (B to F) Differential expression of Th across treatments. (G to K) Differential expression of AADC. (L to P) Differential expression of DNAT. The y axes are normalized read counts. Treatments are color-coded, with NC in dark green, EC in light green, and FC in yellow. The box plots are shown with medians, interquartile range (box), and 1.5× interquartile range (whiskers). PFDR = 0.053 for Th differential expression between FC and EC in ovaries, *PFDR = 0.056 in offspring and ***PFDR < 0.001.

  • Fig. 3 Differential expression of PebIII and acyl-CoA desaturase genes in maternal tissues and in offspring.

    (A to E) Differential expression of PebIII. (F to T) Differential expression of acyl-CoA desaturase genes potentially for CHC synthesis and chemical communication. To distinguish multiple transcripts of desaturase genes, the corresponding contig ID is labeled next to each gene name. y axes are normalized read counts. Treatments are color-coded with NC in dark green, EC in light green, and FC in yellow. The box plots are shown with medians, interquartile range (box), and 1.5× interquartile range (whiskers). *PFDR = 0.034, **PFDR < 0.01, and ***PFDR < 0.001.

  • Fig. 4 Effects of Th and PebIII knockdown on behavior and measures of fitness in earwig mothers and nymphs.

    (A and B) Results for Th knockdown. (C) Pictures of an earwig mother tending hatched nymphs (top) (photo credit to J. Meunier) and tending her eggs (bottom). (D to F) Results for PebIII knockdown. In (E), the relative size of the second clutch was calculated on the basis of the first and second clutch egg numbers. In (A) and (B), frequencies are shown, Note that zero observed provisioning in the M+/O− YFP treatment means that no instance of provisioning was observed during the observation sessions (see Materials and Methods). (D to F) Box plots with median, interquartile range (box), and 1.5× interquartile range (whiskers). Target genes were knocked down in three treatments: only in mothers (M−/O+), only in offspring (M+/O−), and in both (M−/O−). Three corresponding YFP treatments were used to control for the injection of exogenous double-stranded RNA. To test the specific effects of target gene knockdown over and above effects of injection of double-stranded RNA, we used a GLM for statistical analysis with the main effects gene, maternal treatment, and offspring treatment (table S2). Significant gene–by–maternal treatment interaction or gene–by–offspring treatment interactions indicate a social function of a target gene, which are labeled with asterisks; *P < 0.05 and **P < 0.01.

  • Fig. 5 Differential expression of genes related to insulin signaling, JH, and Vg.

    Bars above zero indicate enhanced expression in FC mothers compared to EC mothers in the RNA-seq experiment, while bars below zero indicate reduced expression. Dotted lines indicate thresholds for 2 and −2 log2 fold changes. ILP, insulin-like peptide; IRS, insulin receptor substrate; InR, insulin receptor; FOXO, forkhead domain transcription factor; JHIP, JH ineducable protein; VgR, Vg receptor. To distinguish multiple transcripts of the same gene, the contig ID is labeled next to the corresponding gene name. # ILP showed a trend of differential expression with PFDR = 0.065 in the mothers’ head and PFDR = 0.058 in the ovaries, but there was no expression difference found for IRS or InR. PFDR < 0.01 for all the other genes.

  • Fig. 6 A hypothesized molecular pathway for a link between PebIII expression and reproduction.

    PebIII, an OBP, may play a role in chemical communication between mothers and offspring. PebIII indirectly interact with Vg and JH via apolipophorin and lipophorin (25). Insulin signaling, JH, and Vg interplay with each other and regulate maternal reproduction (26, 27, 29).

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/6/1/eaaw0070/DC1

    Fig. S1. Hatching success of 5073 total eggs in RNA-seq experiment.

    Fig. S2. RT-qPCR validation for Th and PebIII knockdown.

    Fig. S3. Nonsignificant behavior and fitness results for Th and PebIII knockdown.

    Table S1. Hatching success of 5073 total eggs in RNA-seq experiment.

    Table S2. GLM results on behavior and fitness for Th and PebIII knockdown.

    Table S3. Nonsignificant GLM results of behavior and fitness assay for Th and PebIII knockdown.

    Table S4. Primers for double-stranded RNA synthesis.

    Table S5. Sample sizes in the behavioral and fitness assay of the RNAi experiment.

    Table S6. Mortality of mothers and offspring in RNAi experiment.

    Movie S1. Food provisioning in earwigs.

    Data file S1. List of genes responsive to parent-offspring interaction in earwig mothers’ antennae.

    Data file S2. List of genes responsive to parent-offspring interaction in earwig mothers’ head.

    Data file S3. List of genes responsive to parent-offspring interaction in earwig mothers’ abdomen.

    Data file S4. List of genes responsive to parent-offspring interaction in earwig mothers’ ovaries.

    Data file S5. List of genes responsive to parent-offspring interaction in earwig offspring.

  • Supplementary Materials

    The PDFset includes:

    • Fig. S1. Hatching success of 5073 total eggs in RNA-seq experiment.
    • Fig. S2. RT-qPCR validation for Th and PebIII knockdown.
    • Fig. S3. Nonsignificant behavior and fitness results for Th and PebIII knockdown.
    • Table S1. Hatching success of 5073 total eggs in RNA-seq experiment.
    • Table S2. GLM results on behavior and fitness for Th and PebIII knockdown.
    • Table S3. Nonsignificant GLM results of behavior and fitness assay for Th and PebIII knockdown.
    • Table S4. Primers for double-stranded RNA synthesis.
    • Table S5. Sample sizes in the behavioral and fitness assay of the RNAi experiment.
    • Table S6. Mortality of mothers and offspring in RNAi experiment.
    • Legend for movie S1
    • Legends for data files S1 to S5

    Download PDF

    Other Supplementary Material for this manuscript includes the following:

    • Movie S1 (.mp4 format). Food provisioning in earwigs.
    • Data file S1 (Microsoft Excel format). List of genes responsive to parent-offspring interaction in earwig mothers’ antennae.
    • Data file S2 (Microsoft Excel format). List of genes responsive to parent-offspring interaction in earwig mothers’ head.
    • Data file S3 (Microsoft Excel format). List of genes responsive to parent-offspring interaction in earwig mothers’ abdomen.
    • Data file S4 (Microsoft Excel format). List of genes responsive to parent-offspring interaction in earwig mothers’ ovaries.
    • Data file S5 (Microsoft Excel format). List of genes responsive to parent-offspring interaction in earwig offspring.

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article