The Terrestrial Carnivorous Plant Utricularia reniformis Sheds Light on Environmental and Life-Form Genome Plasticity

Abstract

Utricularia belongs to Lentibulariaceae, a widespread family of carnivorous plants that possess ultra-small and highly dynamic nuclear genomes. It has been shown that the Lentibulariaceae genomes have been shaped by transposable elements expansion and loss, and multiple rounds of whole-genome duplications (WGD), making the family a platform for evolutionary and comparative genomics studies. To explore the evolution of Utricularia, we estimated the chromosome number and genome size, as well as sequenced the terrestrial bladderwort Utricularia reniformis (2n = 40, 1C = 317.1-Mpb). Here, we report a high quality 304 Mb draft genome, with a scaffold NG50 of 466-Kb, a BUSCO completeness of 87.8%, and 42,582 predicted genes. Compared to the smaller and aquatic U. gibba genome (101 Mb) that has a 32% repetitive sequence, the U. reniformis genome is highly repetitive (56%). The structural differences between the two genomes are the result of distinct fractionation and rearrangements after WGD, and massive proliferation of LTR-retrotransposons. Moreover, GO enrichment analyses suggest an ongoing gene birth-death-innovation process occurring among the tandem duplicated genes, shaping the evolution of carnivory-associated functions. We also identified unique patterns of developmentally related genes that support the terrestrial life-form and body plan of U. reniformis. Collectively, our results provided additional insights into the evolution of the plastic and specialized Lentibulariaceae genomes.

Keywords: ABC transporters; evolution; genome fractionation; transcription factors; transposable elements; whole-genome duplication.

Figure 1

Utricularia reniformis metaphases with 2n = 40. (A) Metaphase stained with DAPI. (B) Karyogram using the same metaphase presented in (A). Dots in (A) represent the distended and unstained region. The bar is equivalent to 10 µm.

Figure 2

Genome assembly analysis: Read k-mer frequency versus Utricularia reniformis assembly copy number stacked histograms generated by the KAT comp tool. Read content in black is absent from the assembly, red occurs once, purple twice, and green occurs three times, indicating a scenario of both heterozygous and polyploidy genome. The heterozygous content is represented by the first peak at x = 40 and the homozygous content in the second peak at x = 80. The hidden content (the black peak) represents the heterozygous content that is lost when the assembly bubbles are collapsed. The genome assembly contains most (but not all) of the heterozygous content and introduces duplications on both heterozygous and homozygous content.

Figure 3

Utricularia reniformis vs. U. gibba genome comparisons. (A) Macrosynteny karyotype visualization of U. gibba chromosome 1 (CM007989.1), 2 (CM007990.1), 3 (CM007991.1), and 4 (CM007992.1), against the corresponding U. reniformis scaffolds showing matches (generated with the MCscan tool with minspan = 10 option). (B) Dual synteny bar plot, created by the ‘dual-bar plotter’ from the MCScanX. The upper panel represents U. gibba chromosomes and scaffolds; the lower panel represents the U. reniformis scaffolds. The syntenic blocks are shown in colors. Most of the white blocks present in each U. reniformis scaffolds corresponds to exclusive regions and transposable elements. The asterisks correspond to the U. reniformis scaffolds showing no matches to U. gibba chromosomes and scaffolds.

Figure 4

Microsynteny analysis of Utricularia reniformis against U. gibba and other eudicotyledons: (A,B) Polyploid subgenomes of U. reniformis vs. U. gibba microsynteny showing an alternative deletion of the duplicated genes. (C) Highly conserved regions of U. reniformis vs. U. gibba microsynteny. (D) Arabidopsis thaliana vs. U. reniformis, with U. gibba showing moderately conserved regions of microsynteny. (E) Vitis vinifera vs. U. reniformis, with U. gibba showing a considerable fractionation. (F) Lycopersicon esculentum vs. U. reniformis, with U. gibba showing a moderate fractionation.

Figure 5

Overall similarity heatmap and maximum likelihood and Bayesian inference phylogenetic tree of 336 core-eukaryotic genes shared among seven carnivorous plant genomes (Utricularia reniformis, U. gibba, Genlisea aurea, G. nigrocaulis, G. pygmaea, G. repens, and G. hispidula) and Arabidopsis thaliana as an out-group. Numbers above and below the lines indicate the maximum likelihood bootstrap values and the Bayesian posterior probabilities for each clade. The G. pygmaea and G. repens genomes used in this analysis are yet to be published. However, the concatenated core nucleotide gene fasta sequence used to construct this dataset is freely available at http://doi.org/10.5281/zenodo.3268745.

Figure 6

Transposable elements discovery in Utricularia reniformis and U. gibba genomes: Length occupied (bp) of each super-family and evolutionary lineage. A detailed distribution of each element identified is presented in Table S6. The y-axis on the left indicated the total length occupied in the megabases of each evolutionary lineage, whereas the y-axis on the right meant the % of the corresponding genome. Blue and Grey asterisk correspond to the absent evolutionary lineages in U. gibba and U. reniformis, respectively.

Figure 7

Functional enrichment comparative analysis of tandem duplicates among Utricularia reniformis and U. gibba: The enriched GO terms and KEGG enzymes with the corrected p-value < 0.05 are presented. The color of the circle represents the statistical significance of the enriched GO terms (A) and the KEGG Enzymes (B). The size of the circles represents the number of occurrences of a GO term (A) and the KEGG Enzyme (B).

Figure 8

(A) Distribution of ABC transporters and (B) developmental-related transcription factors among Utricularia reniformis and U. gibba. Exclusive genes are species-specific genes for which no orthologs could be found in any of the other species compared. (C) Venn diagram showing the distribution of shared orthologous gene families among U. reniformis, U. gibba, Arabidopsis thaliana, Vitis vinifera, and Lycopersicon esculentum (using an MCL inflation factor of 1.5). The values present inside the box correspond to the number of core genes of each species. The number present inside the parentheses represent the total number of species-specific genes from each species cluster.