The draft genome of Nitzschia closterium f. minutissima and transcriptome analysis reveals novel insights into diatom biosilicification

Abstract

Background: Nitzschia closterium f. minutissima is a commonly available diatom that plays important roles in marine aquaculture. It was originally classified as Nitzschia (Bacillariaceae, Bacillariophyta) but is currently regarded as a heterotypic synonym of Phaeodactylum tricornutum. The aim of this study was to obtain the draft genome of the marine microalga N. closterium f. minutissima to understand its phylogenetic placement and evolutionary specialization. Given that the ornate hierarchical silicified cell walls (frustules) of diatoms have immense applications in nanotechnology for biomedical fields, biosensors and optoelectric devices, transcriptomic data were generated by using reference genome-based read mapping to identify significantly differentially expressed genes and elucidate the molecular processes involved in diatom biosilicification.

Results: In this study, we generated 13.81 Gb of pass reads from the PromethION sequencer. The draft genome of N. closterium f. minutissima has a total length of 29.28 Mb, and contains 28 contigs with an N50 value of 1.23 Mb. The GC content was 48.55%, and approximately 18.36% of the genome assembly contained repeat sequences. Gene annotation revealed 9,132 protein-coding genes. The results of comparative genomic analysis showed that N. closterium f. minutissima was clustered as a sister lineage of Phaeodactylum tricornutum and the divergence time between them was estimated to be approximately 17.2 million years ago (Mya). CAFF analysis demonstrated that 220 gene families that significantly changed were unique to N. closterium f. minutissima and that 154 were specific to P. tricornutum, moreover, only 26 gene families overlapped between these two species. A total of 818 DEGs in response to silicon were identified in N. closterium f. minutissima through RNA sequencing, these genes are involved in various molecular processes such as transcription regulator activity. Several genes encoding proteins, including silicon transporters, heat shock factors, methyltransferases, ankyrin repeat domains, cGMP-mediated signaling pathways-related proteins, cytoskeleton-associated proteins, polyamines, glycoproteins and saturated fatty acids may contribute to the formation of frustules in N. closterium f. minutissima.

Conclusions: Here, we described a draft genome of N. closterium f. minutissima and compared it with those of eight other diatoms, which provided new insight into its evolutionary features. Transcriptome analysis to identify DEGs in response to silicon will help to elucidate the underlying molecular mechanism of diatom biosilicification in N. closterium f. minutissima.

Keywords: Nitzschia closterium f. minutissima; Biosilicification; Draft genome; Frustule; Transcriptome analysis.

Fig. 1

Phylogenetic tree of Phaeodactylum tricornutum_PRJNA943072 and other 8 diatom species based on the 926 single-copy orthogroups. The estimated divergence time (million years ago, Mya) is shown as the blue numbers in the brackets and plotted at each node. The blue bars represent the 95% confidence interval of divergence time. Expansion and contraction of gene family are denoted as numbers with plus and minus signs (+ and -), respectively

Fig. 2

An analysis of significant expansion and contraction of gene families (P < 0.05). (A) Venn and upset plot diagrams of significant expansion and contraction of gene families among nine diatom species. The min overlap set size was 10. (B) GO enrichment analysis for biological process of significant expanded gene families specific to P. tricornutum_PRJNA943072 and P. tricornutum (Top 22). (C) GO enrichment analysis of significant contraction gene families specific to P. tricornutum_PRJNA943072 and P. tricornutum (Top 20)

Fig. 3

Microscopic and transcriptome analysis of synchronized cells of N. closterium f. minutissima after 0, 6 and 12 h of cultivation in 1/2f medium under normal silicate condition. Cells were stained by rhodamine 123 and microscopic images (A) were captured at 0, 6 and 12 h timepoints, bar = 10 μm; (a), (c) and (e), images obtained under bright field; (b), (d) and (f), images obtained under dark field. The new cell wall was detected by green fluorescence. Venn diagram (B) and heatmap (C) of significantly differentially expressed genes extracted from RNA-seq data. The data from 0_h was used as the control. Corrected P-value of 0.05 and absolute foldchange of 2 were set as the threshold for significantly differential expression and the experiment was repeated three times

Fig. 4

GO enrichment (A) and heatmap of the corresponding genes that significantly enriched in biological process and molecular function (B-F). GO category: mf, molecular function; cc, cellular component; bp, biological process. Top 10 of GO term in every category were shown. Si_0h_1, Si_0h_2 and Si_0h_3 represented three repeated samples that were harvested from 24 h silicon-starvation synchronized cultures; Si_6h_1, Si_6h_2 and Si_6h_3 represented three repeated samples that were maintained for 6 h in silicon replenishment media after 24 h silicon-starvation synchronization. Si_12h_1, Si_12h_2 and Si_12h_3 represented three repeated samples that were maintained for 12 h after synchronization

Fig. 5

KEGG pathway enrichment analysis of significantly differentially expressed genes (Top 20)

Fig. 6

Heatmap of genes encoding proteins involved in cytoskeleton-associated genes (A), epigenetic modification (B), protein interaction (C), sugar metabolism and transport (D), and fatty acid metabolism and desaturase (E)