A case of hermaphroditism in the gonochoristic sea urchin, Strongylocentrotus purpuratus, reveals key mechanisms of sex determination†

Abstract

Sea urchins are usually gonochoristic, with all of their five gonads either testes or ovaries. Here, we report an unusual case of hermaphroditism in the purple sea urchin, Strongylocentrotus purpuratus. The hermaphrodite is self-fertile, and one of the gonads is an ovotestis; it is largely an ovary with a small segment containing fully mature sperm. Molecular analysis demonstrated that each gonad producedviable gametes, and we identified for the first time a somatic sex-specific marker in this phylum: Doublesex and mab-3 related transcription factor 1 (DMRT1). This finding also enabled us to analyze the somatic tissues of the hermaphrodite, and we found that the oral tissues (including gut) were out of register with the aboral tissues (including tube feet) enabling a genetic lineage analysis. Results from this study support a genetic basis of sex determination in sea urchins, the viability of hermaphroditism, and distinguish gonad determination from somatic tissue organization in the adult.

Keywords: hermaphrodite; hermaphroditism; ovary; ovotestis; sea urchin; sex determination; testis.

Graphical Abstract

Figure 1

Physical description of hermaphrodite S. purpuratus (scale bar = 200 μM). (A) Darkfield image of gonad  1, a normal testis. (B) Close-up of lobes of gonad  1, a testis, note presence of sperm. (C) Seawater around gonad  1 containing viable, swimming sperm. (D) Darkfield image of gonad  2, appearance is that of a normal ovary. (E) Close-up of lobes of gonad  2, note presence of mature, unfertilized eggs. (F) Darkfield image of gonad  2 and associated unfertilized eggs. (G– I) Darkfield image of gonads  3,  4, and  5. Each of these were normal testes, indistinguishable from  1.

Figure 2

Summary of crosses (scale bar = 100 μM). (A–D) Normal development of Sp embryos using control sperm and control eggs from unrelated male and female urchins. Timepoints embryos were imaged and collected correspond to first cleavage (2hpf), blastula (14hpf), late gastrula (48hpf), and larval stage (4dpf). (E–H) Normal development of embryos from hermaphrodite self-fertilization, sperm from testis and ovary of the same animal were crossed, yielding normal development through day 4. Images are a summary of data; for a full table of crosses, see Supplementary Figure S8. (I–L) Normal development through day 4 from crossing hermaphrodite sperm with control unrelated female eggs. (M–P) Normal development through day 4 from crossing hermaphrodite eggs with unrelated control male sperm. All crosses yielded unremarkable embryos with normal developmental features.

Figure 3

Immunofluorescence of hermaphrodite gonads. (A) Normal control ovary with Hyalin− and Vasa− localization. Vasa highlights the oogonia and early oocytes with intense green cytoplasmic stain, and a cluster of mature Hyalin+ eggs is clearly visible in the center of the gonad lobe (scale bar = 100 μM, inset = 20 μM). (B) 60× image of Hyalin+ mature eggs in a control ovary. (C) Hermaphrodite gonad  2, ovary Hyalin− and Vasa− localization. Several large, Hyalin+ eggs are clearly visible. (D) Close-up of Hyalin+ mature eggs in ovary portion of hermaphrodite gonad  2. (E) Inset showing Hyalin− and Vasa− localization in the ovotestis portion of hermaphrodite gonad  2, note distinct lack of not only Hyalin+ eggs but oogonia and early oocytes are not visible as well. This region was noted to contain sperm. (F) Bindin− and tubulin− localization in a normal control testis. Bindin puncta are visible in green at the tip and collar of the sperm, while tubulin marks the sperm flagella. (G) Inset showing cluster of spermatozoa with highlighted sperm tails. (H) Bindin and tubulin localization in hermaphrodite gonad  1, a normal testis. Localization is indistinguishable from the control. (I) Inset showing tubulin flagella of normal sperm in hermaphrodite gonad  1. (J) Bindin and tubulin localization in ovotestis portion of hermaphrodite gonad  2, appearing indistinguishable from localization in control male.

Figure 4

Fluorescent imaging of hermaphrodite gonads (scale bar = 100 μM, inset = 20 μM). (A–B) Hermaphrodite gonad  1, normal sperm (scale bar = 20 μM). (C, D) Hermaphrodite gonad  2, normal eggs. (E, F) Hermaphrodite gonad  3. (G–H) Hermaphrodite gonad  4. (I–J) Hermaphrodite gonad  5. (K, L) Control ovary. (M, N) Control testis.

Figure 5

Fluorescent imaging of hermaphrodite ovotestis (scale bar = 100 μM, inset = 20 μM). (A, B) Partial ovotestis in a lobe of hermaphrodite gonad  2, where all morphology appears to be that of a normal testis; however, a single egg is present among spermatozoa (asterisk). (C) Partial ovotestis in hermaphrodite gonad  2, a single lobe where a normal looking nest of mature eggs is present; however, all somatic structures resemble that of a testis. (D) Close-up of the partial ovotestis showing spermatozoa with triangular heads (arrowheads) next to a mature egg within the same lobe of the ovotestis.

Figure 6

Hermaphrodite has genetic differences at the Dmrt1 Locus. (A) PCR genotyping control, primers to exon 1 of a doublesex family gene, Doublesex-Like (DsxL), amplify in both male and female control DNA (lanes 1 and 2, respectively), as well as in each of the five gonads of the hermaphrodite Sp. (B) PCR genotyping of Dmrt1 exon 2 yields a genetic difference in control male and female DNA, where the product amplifies in male DNA but not in female DNA. This same difference is observed across the five hermaphrodite gonads; note the absence of the PCR product in the lane corresponding to gonad  2, the ovary, identical to the result in the control female DNA lane.

Figure 7

Differential gene expression profiling. (A) Heatmap of gene expression. Each square represents relative expression of genes on the Y-axis normalized to ubiquitin control, with sperm-related and male-specific transcripts at the top and oocyte-specific transcripts at the bottom. Control male and control female testis and ovary expression patterns are highlighted with a box. (B) PCA plot of gene expression values from individual qPCR results, hermaphrodite gonad expression profiles are shown in pink, control male is purple, and control female is orange.

Figure 8

(A) Map of the Nodal promoter region sequenced for genotyping analysis; primers are shown as blue arrows, and SNP locations are highlighted with red asterisks. Numbers refer to distance (in bp) from the Transcription Start Site (0) which precedes the 5′UTR region, denoted after the arrow. In descending order, genotype at each SNP location for a control male DNA sample, a control female DNA sample, and each of the five hermaphrodite gonads. Alternative genotypes (differing from the majority genotype at each SNP location) are highlighted in different colors. (B) SNP analysis of intergenic region in five unrelated testis samples from a control male. Primers were generated to an intergenic locus at random, producing a 750-bp product. The SNPS within this 750-bp region were quantified and totaled. (C) Analyses of variance of SNPs from intergenic region. Each of the five control male testes SNPs were compared to those in the five hermaphrodite gonads, asterisks represent P-values. For reference, a full table of all pairwise comparisons is given in the supplement, with lowest P-values (four asterisks, P < 0.0001) highlighted in yellow.

Figure 9

We hypothesize that the hermaphrodite is a mosaic. Integrating expression profiles, the Dmrt1 genomic difference, and the SNP profiles, we hypothesize here that at some point during development, chimeric embryo composed of two different genomes developed into this animal. At some point during metamorphosis, a small region of the soma developed female and thus produced a single ovary. It is still unclear exactly what mechanisms underlie this phenotype.