Identification of the meiotic toolkit in diatoms and exploration of meiosis-specific SPO11 and RAD51 homologs in the sexual species Pseudo-nitzschia multistriata and Seminavis robusta

Abstract

Background: Sexual reproduction is an obligate phase in the life cycle of most eukaryotes. Meiosis varies among organisms, which is reflected by the variability of the gene set associated to the process. Diatoms are unicellular organisms that belong to the stramenopile clade and have unique life cycles that can include a sexual phase.

Results: The exploration of five diatom genomes and one diatom transcriptome led to the identification of 42 genes potentially involved in meiosis. While these include the majority of known meiosis-related genes, several meiosis-specific genes, including DMC1, could not be identified. Furthermore, phylogenetic analyses supported gene identification and revealed ancestral loss and recent expansion in the RAD51 family in diatoms. The two sexual species Pseudo-nitzschia multistriata and Seminavis robusta were used to explore the expression of meiosis-related genes: RAD21, SPO11-2, RAD51-A, RAD51-B and RAD51-C were upregulated during meiosis, whereas other paralogs in these families showed no differential expression patterns, suggesting that they may play a role during vegetative divisions. An almost identical toolkit is shared among Pseudo-nitzschia multiseries and Fragilariopsis cylindrus, as well as two species for which sex has not been observed, Phaeodactylum tricornutum and Thalassiosira pseudonana, suggesting that these two may retain a facultative sexual phase.

Conclusions: Our results reveal the conserved meiotic toolkit in six diatom species and indicate that Stramenopiles share major modifications of canonical meiosis processes ancestral to eukaryotes, with important divergences in each Kingdom.

Fig. 1

Phylogenetic tree of Rad21 proteins inferred from maximum likelihood analysis. Numbers on the branches indicate bootstrap support from 1000 replicates. Among-site substitution rate heterogeneity was corrected using two gamma-distributed substitution rate categories and WAG with frequencies (WAG + F) substitution model for amino acid substitutions. Diatom sequences are indicated in bold

Fig. 2

Phylogenetic tree of Spo11 proteins inferred from maximum likelihood analysis. Archaeal topoisomerase VIA protein sequences (underlined) were used as an out-group. Numbers on the branches indicate bootstrap support from 1000 replicates. Among-site substitution rate heterogeneity was corrected using two gamma-distributed substitution rate categories and LG substitution model for amino acid substitutions. Diatom sequences are indicated in bold

Fig. 3

Phylogenetic tree of Rad51 proteins inferred from maximum likelihood analysis. Archaeal Rad-A protein sequences (underlined) were used as an out-group. Numbers on the branches indicate bootstrap support from 1000 replicates. Among-site substitution rate heterogeneity was corrected using two gamma-distributed substitution rate categories and LG substitution model for amino acid substitutions. Diatom sequences are indicated in bold

Fig. 4

Expression profiles of the meiotic toolkit transcripts found in the transcriptome of Seminavis robusta. Expression values (normalized cpm) were determined for dark-synchronized monoclonal cultures (MT+ and MT-) (7 to 10 h after illumination) and for dark-synchronized mixed MT+ and MT- cultures sampled at the time for meiosis (9-10 h after illumination) and post-meiosis (19–22 h after illumination). Blue colour indicates down-regulation and yellow colour upregulation of expression

Fig. 5

Differential expression analysis of RAD21, SPO11 and RAD51 homologs during meiosis in Pseudo-nitzschia multistriata. Two time points, 13 h (a) and 15 h (b) after strains of opposite mating type were mixed together, were selected for expression analyses. Dark gray bars represent experiment 1 (Expt1, B937 (MT+) with B936 (MT-)) and light gray bars represent experiment 2 (Expt2, B938 (MT+) with B939 (MT-)). Relative fold changes, with respect to vegetatively growing cultures, in log2 scale, are reported on the Y-axis. A gene was considered differentially expressed if its relative expression change is 1.5 fold or greater (horizontal black line)