Comparative Embryology and Transcriptomics of Asellus infernus, an Isopod Crustacean From Sulfidic Groundwater

Abstract

Sulfidic caves are harsh and extreme environments characterized by limited oxygen, low pH, and the presence of hydrogen sulfide. Amazingly, animals can live in sulfidic caves, one such animal being Asellus infernus, a representative of the Asellus aquaticus species complex, originating from Movile Cave and from old wells that represent windows of access to a sulfidic groundwater ecosystem located in southeast Romania. Little previous work has been done on lab-reared populations of A. infernus as they have been historically difficult to raise in the lab. Here, we develop resources for A. infernus, examining questions of timing of morphological differences in cave versus surface individuals, whether the environment (lab-bred vs. wild-caught) influenced size characteristics, and the genes and pathways showing differential expression between cave and surface samples. We found that A. infernus did not develop pigmentation embryonically, and juveniles had increased body length and longer antenna II as compared to surface individuals. Furthermore, we found that some of these measures differed between wild-caught and lab-reared juveniles for a given population, indicating that environmental differences can also influence these size characteristics. In addition, differential expression between cave and surface samples and allele-specific expression studies within F1 hybrids identified multiple genes, including those involved in sulfide metabolism and phototransduction. Strikingly, molecular convergence of genes involved in sulfide detoxification was observed between A. infernus and previous work on a fish that lives in both cave and sulfidic environments, Poecilia mexicana. In sum, we were able to develop embryonic and genomic tools for A. infernus, a model for understanding cave adaptation and adaptation to sulfidic environments.

Keywords: allele‐specific expression; cave; isopod; sulfidic water.

Figure 1

Comparison of adult morphology between cave, F1 hybrid, and surface individuals. (A, B, C) Cave individual. (D, E, F) F1 hybrid individual. (G, H, I) Surface individual. (A, D, G) Whole body. (B, E, H) Head. (C, F, I) Profile. In A, the scale bar is 2 mm and is relevant for A, D, and G. In B, the scale bar is 200 µm and relevant for B, C, E, F, H, and I. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2

Comparative embryology between cave, F1 hybrid, and surface embryos. (A–D) Cave embryos. (E–H) F1 hybrid embryos. (I–L) Surface embryos. (A, E, I) Embryos at approximately 70% of the way through embryonic development (mid‐stage embryos). (B, F, J) Embryos at approximately 75% of the way through embryonic development. (C, G, K) Embryos at approximately 80% of the way through embryonic development. (D, H, L) Embryos at approximately 90% of the way through embryonic development. White arrow: Pigmented eye can be seen in H and L. Scale bar is 200 µM for all panels. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3

Comparison of body and antenna length for the comparisons: cave lab versus surface lab, wild versus lab, and F1 versus cave and surface. “Lab” indicates juveniles from females that were raised from juveniles in the lab and bred in the lab. “Wild” indicates juveniles that were collected from wild‐caught individuals with embryos. F1 are F1 hybrids between surface females and cave males. (A) Body length of the juvenile. (B) Size of left antenna II. Black bars and asterisks are for cave lab versus surface lab comparison. Magenta bars and asterisks are for wild versus lab comparisons. Blue bars and asterisks are for F1 versus cave and lab comparisons. Table S4 holds p values of statistical tests of all possible comparisons. ANOVA and Tukey's HSD test were performed for antenna length, body length, and head width. ***p < 0.0005, **0.0005 < p < 0.005, *0.005 < p < 0.05. p > 0.05 was labeled NS. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4

First and second principal component scores of body measurements, which explain roughly 88% and 6.4% of total variation, respectively. Archetypal morphologies for extremes in each direction are depicted with isopod schematic diagrams, with the average for each subsample denoted by a cross. Higher first principal component scores generally correspond to larger body measurements overall and are generally held by isopods with longer body lengths, antennae, and wider heads; lower first principal component scores are associated with smaller measurements overall. Higher second principal component scores are held by isopods with longer antennae but shorter body lengths, and vice versa for isopods with lower second principal component scores. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 5

Genes showing allele‐specific expression. Three F1 hybrid samples were examined both in reference to the surface transcriptome and the cave transcriptome. A. Allele counting was performed using FreeBayes variant counting for five SNPs along the location of each gene, both the cave and surface versions of the transcript (shown as rows labeled as SNP1‐5 in parts B and C). C represents the cave allele, and S represents the surface allele. Here, a schematic of a hypothetical gene with cave‐biased expression is shown as an example. B. Genes that show cave‐biased allele‐specific expression are shown in red. C. Genes that show surface‐biased allele‐specific expression are in green. We performed binomial tests to show whether the ratio of surface to cave alleles deviates significantly from the expected 1:1 ratio. Multiple comparison correction was performed using the Benjamini–Yekutieli procedure. White cells denote non‐significance with p > 0.05, light green/red cells denote significance with 0.005 < p < 0.05, plain green/red cells denote significance with 0.0005 < p < 0.005, and dark green/red cells denote significance with p < 0.0005. Genes shown in B and C are those for which at least two SNPs showed a p < 0.05 for all hybrid samples to both the surface and cave transcript. The code before the gene name is the Tribolium castaneum gene ID. An asterisk indicates a gene that did not have a high‐quality, non‐chimeric primary alignment for either the surface or cave transcript. [Color figure can be viewed at wileyonlinelibrary.com]