Profiling sex-biased gene expression during parthenogenetic reproduction in Daphnia pulex

Abstract

Background: Sexual reproduction is a core biological function that is conserved throughout eukaryotic evolution, yet breeding systems are extremely variable. Genome-wide comparative studies can be effectively used to identify genes and regulatory patterns that are constrained to preserve core functions from those that may help to account for the diversity of animal reproductive strategies. We use a custom microarray to investigate gene expression in males and two reproductive stages of females in the crustacean Daphnia pulex. Most Daphnia species reproduce by cyclical parthenogenesis, alternating between sexual and clonal reproduction. Both sex determination and the switch in their mode of reproduction is environmentally induced, making Daphnia an interesting comparative system for the study of sex-biased and reproductive genes.

Results: Patterns of gene expression in females and males reveal that 50% of assayed transcripts show some degree of sex-bias. Female-biased transcription is enriched for translation, metabolic and regulatory genes associated with development. Male-biased expression is enriched for cuticle and protease function. Comparison with well studied arthropods such as Drosophila melanogaster and Anopheles gambiae suggests that female-biased patterns tend to be conserved, whereas male-biased genes are evolving faster in D. pulex. These findings are based on the proportion of female-biased, male-biased, and unbiased genes that share sequence similarity with proteins in other animal genomes.

Conclusion: Some transcriptional differences between males and females appear to be conserved across Arthropoda, including the rapid evolution of male-biased genes which is observed in insects and now in a crustacean. Yet, novel patterns of male-biased gene expression are also uncovered. This study is an important first step towards a detailed understanding of the genetic basis and evolution of parthenogenesis, environmental sex determination, and adaptation to aquatic environments.

Figure 1

Ratio vs. intensity and log odds vs. fold change plots. (A, D) Males vs. juvenile females; (B, E) males vs. mature females; (C, F) self vs. self hybridizations. Ratio-intensity plots (A-C) depict values for a representative experimental replicate, while log odds-fold change plots (D-F) are based on four replicate hybridizations.

Figure 2

Comparison of microarray (MA) and quantitative RT-PCR (PCR) results. For MA results, normalized ratio and intensity values were used to back-calculate red and green fluorescence intensity for each experimental comparison and are reported as mean ± one standard error. For PCR results, standard curves were used to generate estimates of starting amounts for each reaction on a relative scale. Biological and technical replicates were combined for each group to calculate mean ± one standard error.

Figure 3

Log-odds vs. fold changes males and females. Using a linear model, log-odds and log fold changes were calculated for pregnant females vs. juvenile females (A) and male vs. male comparisons (B) and plotted based on four experimental replicates.

Figure 4

Scatterplot of highly differentially expressed genes. All significantly differentially expressed genes at two-fold change or above between males and mature females (X-axis) and between males and juvenile females (Y-axis) are depicted with male vs. mature as red squares; male vs. juvenile as blue diamonds; and mature vs. juvenile females as black triangles.

Figure 5

Analysis of Gene Ontology (GO) terms. Genes were categorized as significantly differentially expressed in a two-way comparison with false discovery rate correction and p < 0.05, and information about associated GO terms was extracted using Blast2GO [39]. Fisher's exact test was used to calculate the probability that the observed distribution of terms in the test set was significantly different (p-value < 0.05) from the distribution of all assembled sequences for the following comparisons: A) females vs. males; B) males vs. females; C) juvenile females vs. mature females; D) mature females vs. juvenile females; E) mature females vs. males; F) Daphnia and Drosophila conserved sex-biased genes. GO terms for each column are: 1, chromatin; 2, nucleosome; 3, nucleotide binding; 4, microtubule cytoskeleton organization and biogenesis; 5, lytic vacuole; 6, pattern binding; 7, nucleic acid binding; 8, DNA binding; 9, structural constituent of ribosome; 10, translation elongation factor activity; 11, catalytic activity; 12, GTPase activity; 13, endopeptidase activity; 14, serine-type endopeptidase activity; 15, chymotrypsin activity; 16, trypsin activity; 17, hydrolase activity, hydrolyzing O-glycosyl compounds; 18, chitinase activity; 19, structural molecule activity; 20, GTP binding; 21, cellular component; 22, extracellular region; 23, intracellular; 24, cell; 25, nucleus; 26, chromosome; 27, cytoplasm; 28, lysosome; 29, vacuole; 30, endoplasmic reticulum; 31, ribosome; 32, microtubule; 33, carbohydrate metabolism; 34, polysaccharide metabolism; 35, chitin metabolism; 36, nucleobase, nucleoside, nucleotide and nucleic acid metabolism; 37, DNA packaging; 38, establishment and/or maintenance of chromatin architecture; 39, chromatin assembly or disassembly; 40, nucleosome assembly; 41, transcription; 42, transcription, DNA-dependent; 43, regulation of transcription, DNA-dependent; 44, protein biosynthesis; 45, translational elongation; 46, protein complex assembly; 47, proteolysis and peptidolysis; 48, amino acid and derivative metabolism; 49, amino acid metabolism; 50, aromatic compound metabolism; 51, organelle organization and biogenesis; 52, chromosome organization and biogenesis (sensu Eukaryota); 53, cytoskeleton organization and biogenesis; 54, microtubule-based process; 55, physiological process; 56, chitin binding; 57, translation factor activity, nucleic acid binding; 58, biological_process; 59, metabolism; 60, peptidase activity; 61, serine-type peptidase activity; 62, catabolism; 63, macromolecule catabolism; 64, biosynthesis; 65, macromolecule biosynthesis; 66, cellular process; 67, microtubule cytoskeleton; 68, small ribosomal subunit; 69, cell organization and biogenesis; 70, hydrolase activity; 71, hydrolase activity, acting on glycosyl bonds; 72, isomerase activity; 73, purine nucleotide binding; 74, guanyl nucleotide binding; 75, regulation of nucleobase, nucleoside, nucleotide and nucleic acid metabolism; 76, regulation of metabolism; 77, protein metabolism; 78, protein catabolism; 79, carbohydrate binding; 80, polysaccharide binding; 81, ribonucleoprotein complex; 82, microtubule polymerization or depolymerization; 83, structural constituent of cuticle; 84, translation; 85, macromolecule metabolism; 86, organelle; 87, membrane-bound organelle; 88, non-membrane-bound organelle; 89, intracellular organelle; 90, intracellular membrane-bound organelle; 91, intracellular non-membrane-bound organelle; 92, protein complex; 93, biopolymer metabolism; 94, biopolymer catabolism; 95, cellular metabolism; 96, primary metabolism; 97, cellular catabolism; 98, cellular biosynthesis; 99, cellular protein catabolism; 100, cellular macromolecule metabolism; 101, cellular carbohydrate metabolism; 102, cellular polysaccharide metabolism; 103, cellular macromolecule catabolism; 104, cellular protein metabolism; 105, translation regulator activity; 106, tubulin; 107, regulation of transcription; 108, microtubule polymerization; 109, regulation of biological process; 110, regulation of physiological process; 111, cellular physiological process; 112, protein polymerization; 113, chromosome organization and biogenesis.

Figure 6

Scatterplot of male vs. female ratios for Daphnia and Drosophila. D. pulex female/male ratios (females positive) are plotted on the X-axis and D. melanogaster female/male ratios (females positive) are plotted on the Y-axis. Genes with an e-value less than e^-08 were considered homologs, and all D. pulex genes show significant differential expression between males and females. Numbers of genes with significant homology include 27 from e^-08 to e^-10; 21 from e^-11 to e^-20; 54 from e^-21 to e^-50; 33 from e^-51 to e^-80; and 15 at e^-81 and lower.

Figure 7

Sequence homology and differential expression. Histogram of sequence homology vs. count percentage for weakly biased (ratio less than 2-fold different), strongly biased (ratio greater than 2-fold different) and all biased genes in male vs. female comparison. Homology is defined as strong if bit score is 101 or greater, corresponding to an e-value of roughly 1 × 10^-20; moderate if bit score is greater than 50, corresponding to an e-value of roughly 1 × 10^-08; or weak if bit score is 50 or less.