Mitochondrial Genome of Non-Photosynthetic Mycoheterotrophic Plant Hypopitys 1 Monotropa,Its Structure, Gene Expression And
Total Page：16
File Type：pdf, Size：1020Kb
bioRxiv preprint doi: https://doi.org/10.1101/681718; this version posted December 3, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1 Mitochondrial genome of non-photosynthetic mycoheterotrophic plant Hypopitys 2 monotropa,its structure, gene expression and RNA editing 3 4 Viktoria Y. Shtratnikova1, Mikhail I. Schelkunov2,3, Aleksey A. Penin2, Maria D. Logacheva1,2,3 5 6 1A. N. Belozersky Institute of Physico-Chemical Biology, Lomonosov Moscow State University, 7 Moscow, Russian Federation 8 2Laboratory of Plant Genomics, Institute for Information Transmission Problems of the Russian 9 Academy of Sciences, Moscow, Russian Federation 10 3Skolkovo Institute of Science and Technology, Moscow, Russian Federation 11 12 Corresponding Author: 13 Maria D. Logacheva1,2,3 14 1Lomonosov Moscow State University, Leninskie Gory, GSP-1, Moscow 119991, Russia 15 2BolshoyKaretny lane, 19/1, Moscow 127051, Russia 16 3Skolkovo Institute of Science and Technology, Nobel St. 3, Moscow 143026, Russia 17 18 Email address: [email protected] 19 20 Abstract 21 Heterotrophic plants – the plants that lost the ability to photosynthesis – are characterized by a 22 number of changes at all levels of organization. Heterotrophic plants divide into two large 23 categories – parasitic and mycoheterotrophic. The question of to what extent these changes are 24 similar in these two categories is still open. Plastid genomes of non-photosynthetic plants are 25 well characterized and they demonstrate similar patterns of reduction in both groups. In contrast, 26 little is known about mitochondrial genomes of mycoheterotrophic plants. We report the 27 structure of the mitochondrial genome of Hypopitys monotropa, a mycoheterotrophic member of 28 Ericaceae, and the expression of mitochondrial genes. In contrast to its highly reduced plastid 29 genome, the mitochondrial genome of H. monotropa is larger than that of its photosynthetic 30 relative Vaccinium macrocarpon, its complete size is ~810 Kbp. We found an unusually long 31 repeat-rich structure of the genome that suggests the existence of linear fragments. Despite this 32 unique feature, the gene content of the H. monotropa mitogenome is typical of flowering plants. 33 No acceleration of substitution rates is observed in mitochondrial genes, in contrast to previous 34 observations on parasitic non-photosynthetic plants. Transcriptome sequencing revealed trans- 35 splicing of several genes and RNA editing in 33 genes of 38. Notably, we did not find any traces 36 of horizontal gene transfer from fungi, in contrast to plant parasites which extensively integrate 37 genetic material from their hosts. 38 Introduction 39 Hypopitys monotropa (Ericacae, Monotropoideae) is a non-photosynthetic plant gaining carbon 40 from fungi that are ectomycorrhizal with tree roots (Bjorkman, 1960). In contrast to most other bioRxiv preprint doi: https://doi.org/10.1101/681718; this version posted December 3, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 41 mycoheterotrophic (MHT) plants, which are very rare and/or very narrowly distributed, 42 Monotropoideae, including H. monotropa, are quite widespread, being associated with old- 43 growth conifer forests. Thus, H. monotropa is used as a model system in studies of plant- 44 mycorrhizal associations and developmental biology of MHT plants (e.g. Olson, 1993, 1990). 45 Recent advances in DNA sequencing allow expanding the studies of mycoheterotrophs into 46 genomics. By now, the most attention was focused on plastid genomes of MHT plants. They are 47 highly reduced in size and gene content, includingcomplete absence or pseudogenization of 48 genes of photosynthesis electron transport chain (for review see Graham et al., 2017). Thus, the 49 MHT lifestyle strongly affects plastids, but what about the mitochondrial genome? 50 In contrast to animals, where mitochondrial genomes are usually conserved in size and gene 51 content across large taxonomic groups, in plants they are highly variable and may be very 52 dissimilar even in closely related species. The size of the angiosperm mitogenome is ranging 53 from 66 Kb in the hemiparasitic mistletoe Viscum scurruloideum (Skippington et al., 2015), 222 54 Kb in the autotrophic Brassica napus (Handa, 2003) to more than 11 Mb in Silene noctiflora . 55 Despite such huge variation in size, the large part of mitochondrial genes – the ones that encode 56 the components of the oxidative phosphorylation chain complexes and proteins involved in the 57 biogenesis of these complexes – are stable in content and have very low sequence divergence. 58 More variation exists in the group of genes involved in translation – i.e. ribosomal proteins and 59 transfer RNAs (Adams et al., 2002; Gualberto et al., 2014), presumably due to the transfer to the 60 nuclear genome that occurred in several plant lineages or to the non-essentiality of these genes. 61 Besides this, many plant mitochondrial genomes carry open reading frames (ORF) that 62 potentially encode functional proteins (Qiu et al., 2014); such ORFs are highly lineage- 63 specific.Non-photosynthetic plants divide into two large groups—those that are parasitic on other 64 plants and mycoheterotropic. By now only few complete mitochondrial genomes of non- 65 photosynthetic plants are characterized, and most of them are parasitic. A comparative analysis 66 of mitogenomes of several parasitic, hemiparasitic and autotrophic Orobanchaceae (Fan et al., 67 2016) showed that the gene content does not depend on trophic specialization in family range. 68 Mitogenomes of two non-related lineages of parasitic plants: Rafflesiaceae and Cynomoriaceae 69 also do not show reduction in gene content; besides that, they are the example of massive HGT 70 from other plants (including, but not only, their hosts). This is not, however, a trait unique for 71 parasitic plants—e.g. Amborella trichopoda, an autotrophic plant from basal angiosperms, has in 72 its mitochondrial genome a large fraction acquired from green algae, mosses, and other 73 angiosperms (Rice et al., 2013). In contrast, in the hemiparasitic plant Viscum scurruloideum the 74 mitogenome is drastically reduced in size and gene content, it lacks all nine nad genes, and matR 75 (Skippington et al., 2015). Mitogenomes of other Viscum species are not reduced in length but 76 have reduced gene content, similar to V. scurruloideum ((Petersen et al., 2015); but see 77 (Skippington et al., 2017)). The sampling is obviously insufficient even for parasitic plants; 78 regarding mycoheterotrophs the data are almost completely lacking. There are only two 79 mitochondrial genomes of a MHT plants characterized by date - that of orchid Gastrodia elata 80 (Yuan et al., 2018) and Epirixanthes elongata (Polygalaceae) (Petersen et al., 2019). The former bioRxiv preprint doi: https://doi.org/10.1101/681718; this version posted December 3, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 81 one is characterized as a part of genome sequencing project and is not assembled completely 82 being represented by 19 contigs. G. elata mitogenome is large (~ 1.3 Mb) while in E. elongata it 83 is much smaller (~0,4 Mb). However, both above-mentioned studies lack comparative analysis 84 with autotrophic relatives that would allow to highlight the changes associated with 85 mycoheterotrophy. 86 Taking this into account, we set the following objectives: 1) to characterize the structure and 87 gene content of the mitochondrial genome of H. monotropa 2) to estimate if horizontal gene 88 transfer (HGT) from fungi took place 3) to study the mitochondrial gene expression and RNA 89 editing. 90 The photosynthetic plant with characterized mitochondrial genome phylogenetically closest to H. 91 monotropa is cranberry, Vaccinium macrocarpon (Fajardo et al., 2014). In this study we use V. 92 macrocarpon as a basis for comparative analysis aimed at finding the patterns (if there are any) 93 of mitochondrial genome changes associated with mycoheterotrophy. 94 Materials & Methods 95 2.1. Sample collection and sequencing 96 Sample collection, DNA and RNA libraries preparation for most datasets (DNA shotgun and 97 mate-pair libraries and inflorescence transcriptome library) and sequencing were described in in 98 the papers of Logacheva et al. (2016) and Schelkunov et al. (2018). In addition to the data 99 generated in previous studies we used four new datasets which represent transcriptomes of 100 anthers and ovules of H. monotropa. The samples were collected in the same location as 101 previous ones but in 2018. RNA was extracted using RNEasy kit (Qiagen) with the addition of 102 Plant RNA isolation aid reagent (Thermo fisher) to the lysis buffer. Removal of ribosomal RNA 103 was performed using Ribo-Zero plant leaf kit (Illumina) and further sample preparation using 104 NEBNext RNA library preparation kit (New England Biolabs). The libraries were sequenced on 105 HiSeq 4000 (Illumina) in 150-nt paired-end mode. The reads are deposited in NCBI Sequence 106 Read Archive under the BioProject PRJNA573526. 107 2.2. Assembly 108 Read trimming and assembly were made as described in the paper of Logacheva et al. (2016). 109 The contig coverages were determined by mapping all reads in CLC Genomics Workbench v. 110 7.5.1 (https://www.qiagenbioinformatics.com/products/clc-genomics-workbench/), requiring at 111 least 80% of a read's length to align with at least 98% sequence similarity. To find contigs 112 corresponding to the mitochondrial genome, we performed BLASTN and TBLASTX alignment 113 of Vaccinium macrocarpon genome (GenBank accession NC_023338) against all contigs in the 114 assembly. BLASTN and TBLASTX were those of BLAST 2.3.0+ suite (Camacho et al., 2009), 115 the alignment was performed with the maximum allowed e-value of 10-5. Low-complexity 116 sequence filters were switched off in order not to miss genes with extremely high or low GC- 117 content.