Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

De novo activated transcription of inserted foreign coding sequences is inheritable in the plant genome

  • Takayuki Hata ,

    Roles Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Validation, Writing – original draft, Writing – review & editing

    ‡ Takayuki Hata and Naoto Takada are equal contributors and joint first authors.

    Affiliations Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan, Faculty of Agriculture, Setsunan University, Hirakata-shi, Osaka, Japan

    ⨯
  • Naoto Takada ,

    Roles Formal analysis, Investigation, Methodology, Writing – original draft

    ‡ Takayuki Hata and Naoto Takada are equal contributors and joint first authors.

    Affiliation Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Chihiro Hayakawa ,

    Contributed equally to this work with: Chihiro Hayakawa, Mei Kazama

    Roles Formal analysis, Investigation

    Affiliation Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Mei Kazama ,

    Contributed equally to this work with: Chihiro Hayakawa, Mei Kazama

    Roles Formal analysis, Investigation

    Affiliation Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Tomohiro Uchikoba,

    Roles Formal analysis, Investigation

    Affiliation Faculty of Life and Environmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Makoto Tachikawa,

    Roles Formal analysis, Investigation, Validation

    Affiliation Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Mitsuhiro Matsuo,

    Roles Data curation, Investigation, Validation, Writing – review & editing

    Affiliation Faculty of Agriculture, Setsunan University, Hirakata-shi, Osaka, Japan

    ⨯
  • Soichirou Satoh ,

    Roles Data curation, Formal analysis, Funding acquisition, Supervision, Writing – review & editing

    junichi.obokata@setsunan.ac.jp (JO); s-satoh@kpu.ac.jp (SS)

    Affiliations Graduate School of Life and Environfmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan, Faculty of Life and Environmental Sciences, Kyoto Prefectural University, Kyoto-shi, Kyoto, Japan

    ⨯
  • Junichi Obokata

    Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft, Writing – review & editing

    junichi.obokata@setsunan.ac.jp (JO); s-satoh@kpu.ac.jp (SS)

    Affiliation Faculty of Agriculture, Setsunan University, Hirakata-shi, Osaka, Japan

    ⨯

Abstract

The manner in which inserted foreign coding sequences become transcriptionally activated and fixed in the plant genome is poorly understood. To examine such processes of gene evolution, we performed an artificial evolutionary experiment in Arabidopsis thaliana. As a model of gene-birth events, we introduced a promoterless coding sequence of the firefly luciferase (LUC) gene and established 386 T2-generation transgenic lines. Among them, we determined the individual LUC insertion loci in 76 lines and found that one-third of them were transcribed de novo even in the intergenic or inherently unexpressed regions. In the transcribed lines, transcription-related chromatin marks were detected across the newly activated transcribed regions. These results agreed with our previous findings in A. thaliana cultured cells under a similar experimental scheme. A comparison of the results of the T2-plant and cultured cell experiments revealed that the de novo-activated transcription concomitant with local chromatin remodelling was inheritable. During one-generation inheritance, it seems likely that the transcription activities of the LUC inserts trapped by the endogenous genes/transcripts became stronger, while those of de novo transcription in the intergenic/untranscribed regions became weaker. These findings may offer a clue for the elucidation of the mechanism by which inserted foreign coding sequences become transcriptionally activated and fixed in the plant genome.

Citation: Hata T, Takada N, Hayakawa C, Kazama M, Uchikoba T, Tachikawa M, et al. (2021) De novo activated transcription of inserted foreign coding sequences is inheritable in the plant genome. PLoS ONE 16(6): e0252674. https://doi.org/10.1371/journal.pone.0252674

Editor: Hodaka Fujii, Hirosaki University Graduate School of Medicine, JAPAN

Received: December 23, 2020; Accepted: May 19, 2021; Published: June 10, 2021

Copyright: © 2021 Hata et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: This work was supported by the Japan Society for the Promotion of Science (JSPS) KAKENHI Grant Number: 17J04887 to Takayuki Hata, 26660008 to Soichirou Satoh, and 17K19358, 23125512, 23657040, 23117006, and 25650131 to Junichi Obokata, and by the Strategic Research Funds at Kyoto Prefectural University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Genomes exhibit a steady state of the dynamic activity between the gain and loss of genes. Comparative functional genomics among closely related species revealed how genomes acquired such genetic novelty during their evolution, i.e., duplication–diversification, transposition, gene transfer or de novo origination [1–4]. Importantly, to become functional, new coding sequences must acquire promoters during their evolution. A gene promoter is a region in which transcription is initiated and is a central component of gene regulation [5,6]. In eukaryotes, promoter-specific sequences and chromatin marks are well characterized [5,6]. The study of the origin of genic promoters has been led by the examination of evolutionarily young genes. Notably, the de novo gene, whose coding sequence arises from an ancestral non-coding sequence, is a valuable model to study how newly originated coding sequences became transcribed [3,4,7,8]. De novo genes were often found from ancestral non-coding RNA genes [9,10], near the bidirectional promoters [9,11–14], and within enhancers [12,15–17], open-chromatin regions [16,17] and pervasively/divergent transcribed regions [18–25]. However, because the young genes found in present-day genomes became fixed thousands to millions of years ago, the molecular portraits of the newly originated promoters shortly after their birth are largely overlooked.

Differing from the comparative genomics approach, artificial evolutionary experiments can dissect the molecular mechanisms of how newly emerged coding sequences acquire their promoters in the timescales of molecular biology and genetics. Previously, to mimic a gene-birth event, we introduced exogenous promoterless coding sequences into A. thaliana T87 cultured cells and analysed their transcriptional fates at the genome-wide level [26] (unpublished data; available at https://doi.org/10.1101/2020.11.28.401992). Interestingly, we found that promoterless coding sequences became transcriptionally activated via two distinct mechanisms: (1) the so-called promoter trapping, in which integrated sequences capture the endogenous promoter activities of pre-existing genes/transcripts; or (2) de novo transcriptional activation, which occurs ubiquitously across the entire genome, and does so stochastically in about 30% of the integration events independently of the chromosomal locus [26]. The transcription start sites (TSSs) analysis revealed that the de novo transcription occurred from the 5’ close proximal region of the inserted foreign coding sequences [27]. We speculated that the insertion of exogenous coding sequences might activate local chromatin remodelling to shape the promoter-like chromatin configuration, resulting in such de novo transcriptional activation in the plant genome.

Could this de novo transcriptional activation be a causative mechanism by which newly originated coding sequences acquire their transcriptional competency in the plant genome evolution? Testing this possibility requires the assessment of the genetic behaviour of the de novo transcription over generations. The cultured cell-based experiment is not suitable for such scope because the cultured cells continue only the vegetative growth. In this respect, artificial evolutionary experiments with whole plants could provide clues to the above question. The plant body develops with the continuous formation of various tissues and organs from stem cells. Heterogeneity of the transcriptome and epigenome among these different tissues and developmental stages are well characterized in A. thaliana plants [28–31]. During plant reproductive development, dynamic chromatin remodelling including the localizations of DNA methylation and specific histone species occurs [32–36]. It is unpredictable from the cultured cell-based experiment how such chromatin remodelling could have an influence on the de novo transcription in the plant genome over generations.

In this study, we aimed to establish a model system to elucidate the mechanism by which inserted foreign coding sequences acquire their promoters and become fixed as functional genes in the plant genome. We carried out a large-scale promoter-trap screening in the T2 generation of A. thaliana plants under an experimental scheme similar to that used in our previous study of cultured cells [26]. By comparing the results obtained in plants with those of cultured cells, we concluded that de novo transcriptional activation together with chromatin remodelling is an inheritable phenomenon in the plant genome. After one generation, the transcriptional activities of introduced coding sequences trapped by endogenous genes/transcripts became much stronger, while those of the intergenic/untranscribed regions became much weaker. These findings may contribute to the elucidation of how newly emerged coding sequences become transcriptionally activated and fixed in the plant genome at their early evolutionary stages.

Results

Establishment of transgenic lines for large-scale promoter-trap screening in A. thaliana

To investigate the mechanism of promoter birth and their genetic behaviours beyond one generation, we performed a promoter-trap screening using A. thaliana plants under conditions that were essentially the same as those used in a previous study of cultured cells [26]. Based on Agrobacterium-mediated transformation [37], we introduced the promoterless coding sequence of a firefly luciferase (LUC) gene into A. thaliana (Fig 1). Each LUC gene was tagged by distinct short random sequences called ‘barcodes’ (Fig 1), which were used as identification codes for individual transgenic lines in the subsequent in silico analysis. In this study, to analyse the transgenic lines without the selection bias caused by LUC gene function, we screened the T1 seeds against the kanamycin (Km) resistance of the co-transformed expression cassette, rather than the strength of the LUC luminescence (Fig 1). Finally, we established a T2 generation of 386 individual transgenic lines (termed T2-plants hereafter).

thumbnail
Download:
Fig 1. Experimental design of the promoter-trap experiment in A. thaliana plants.

Schematic illustration of the TRIP experiment performed in the T2 generation of A. thaliana transgenic lines. T-DNA including a barcode, a promoterless LUC gene and an expression cassette with a Km-resistance gene was introduced into A. thaliana via Agrobacterium-mediated transformation. T2 seeds were harvested from Km-resistant T1 lines. Three seeds per T2 transgenic line were grown under the non-selective condition and subjected to subsequent locus and transcription-level analysis based on the TRIP method. NptII, neomycin phosphotransferase II; nosp, nopaline synthase promoter; nost, nopaline synthase terminator.

https://doi.org/10.1371/journal.pone.0252674.g001

Genetic behaviours of de novo-activated transcription in A. thaliana

To identify the insertion loci and corresponding transcription levels of the individual LUC genes, we performed a massively parallel reporter assay based on the TRIP method [38].

First, three seeds per individual T2-plant were grown using the non-selective condition and seedlings were harvested as a mixed sample (Fig 1). Because the T2 generation is not homozygous, theoretically, one-fourth of T2 seeds were expected to be wild type (WT). However, as we grew three seedlings per line, no less than 98% of T2-plants (380/386) were expected to be recovered. In the TRIP method, individual transgenic lines are identified via in silico analysis based on the tagged barcode sequence of the reporter construct, as a molecular identifier. Specifically, we extracted DNA and RNA from the mixed samples and prepared the next-generation sequencing (NGS) library. For the determination of the insertion locus of each promoterless LUC gene, we performed inverse PCR followed by NGS, to read out the LUC–genome junction and barcode sequence. The relative transcription level of each LUC gene was determined utilizing NGS by counting the molecular abundance of each barcode in the RNA sample. Finally, each LUC gene-insertion locus and transcription level was assigned according to its barcode sequences. Note that T-DNAs are often inserted tandemly or with a large deletion on the reporter gene [39]. We carefully omitted such lines from further analysis because we could not determine their insertion loci uniquely. Based on this scheme, we determined individual insertion loci and corresponding transcription levels in 76 T2-plants (Figs 1 and 2A, and S1 Table). To confirm the results of the in silico analysis, we verified individual barcode sequences and insertion loci in randomly chosen T2-plants using Sanger sequencing and locus-specific PCR (S1 Fig). As shown in Fig 2A, promoterless LUC genes were inserted throughout the A. thaliana genome with low frequency in pericentromeric regions, which agreed with the reported preference of Agrobacterium T-DNA integration [40]. One-third of the 76 LUC genes (n = 27) were transcribed (Fig 2B). To examine further how these promoterless LUC genes became transcribed, we classified them according to their insertion types: an endogenous genic region with the sense (Genic Sense) and antisense (Genic AS) orientation, and the remaining intergenic regions (Intergenic). Based on this classification, the Genic Sense, Genic AS, and Intergenic types accounted for 26.3%, 21.1%, and 52.6% of the transcribed LUC genes, respectively (Fig 2C). Because the genic and intergenic regions of the A. thaliana genome have almost the same length [41], these results suggest that our established T2-plants exhibited no insertion-locus preference.

thumbnail
Download:
Fig 2. An artificial evolutionary experiment revealed the genetic behaviours of the activated transcription of coding sequences inserted in A. thaliana plants.

(A) The insertion loci and transcription levels of determined T2-plants (n = 76) were mapped on the A. thaliana chromosomes. The coloured bars indicate individual insertion sites and corresponding transcription levels based on their percentiles (High: 100–67, Mid: 66–34, and Low: 33–1). (B) Classification of T2-plants according to their transcription. (C) Number of T2-plants according to their insertion types: Genic sense, AS (antisense) or intergenic. The definition of each type is provided in Materials and Methods. (D) Fraction of transcribed LUC genes among T2-plants (n = 76) and T87 cultured cells (n = 4,443) (Satoh et al., 2020) against each insertion type, as in (C). (E) Fraction of LUC genes in T2-plants (upper panel) and T87 cells (lower panel) (Satoh et al., 2020) against their transcription levels, as normalized using the total number of each insertion type as 100%. ND, untranscribed LUC genes. (F) The abundance of transcribed LUC genes in each insertion type was classified according to their transcription levels; lower (101–104) and higher (105–107), as in (E). Each frequency was normalized to the number of transcribed LUC genes in each insertion type, which was set as 100%.

https://doi.org/10.1371/journal.pone.0252674.g002

Previously, based on the cultured cell experiment, we found that exogenously inserted promoterless genes became transcriptionally activated in two distinct types: promoter trapping and de novo transcriptional activation [26,27]. To examine whether similar transcriptional activation mechanisms occurred in our T2-plants, the abundance of the transcribed fraction was compared between the corresponding insertion types of T2-plants and cultured cells (Fig 2D and 2E). As shown in Fig 2D, ~30% of the promoterless LUC genes were transcribed similarly in T2-plants and cultured cells (Fig 2D). Their relative transcription levels ranged from 101 to 107 orders, with a peak at 104 (Fig 2E). Regarding the three insertion types, the abundances of transcribed LUC genes were almost the same in T2-plants and cultured cells, except for the Genic Sense type, in which the transcribed fraction was much greater in the T2-plants (Fig 2D). In both T2-plants and cultured cells, the Genic Sense type showed the highest transcriptional activity among the three insertion types, with 105 as a peak (Fig 2E).

To highlight the differences between the cultured cells and T2-plants, we divided the transcribed LUC lines into two fractions: that with a lower transcription level (101–104) and that with a higher transcription level (105–107). As shown in Fig 2F, the relative abundances of the higher and lower fractions in T2-plants exhibited a greater bipolarization than they did in cultured cells; LUC transcription became much stronger in the Genic Sense and AS types, while it became much weaker in the Intergenic type (Fig 2F). As the Genic Sense and AS types were transcribed presumably by trapping the transcriptional activities of endogenous genes [27], these features suggested that gene-trapping events are more prone to occur in plants than in cultured cells. Conversely, a type of transcriptional repression might have occurred on the Intergenic type in the T2 generation.

Taken together, these results suggest that the transcriptional behaviours of the promoterless LUC genes are remarkably similar between the T2-plants and the vegetatively growing cultured cells (Fig 2D and 2E). Therefore, it is likely that de novo transcriptional activation events are not specific to the vegetatively growing cultured cells; rather, they seem to be an inheritable phenomenon through a plant’s generation.

Comparison of LUC transcription with inherent transcriptional status

Are there any other similarities/differences between T2-plants and cultured cells? To address this question, we next focused on the correlation of the transcriptional status between the LUC genes and the corresponding WT loci. For this, we prepared a dataset of the transcribed regions of the WT genome using the publicly available RNA-seq data of A. thaliana, which were obtained using growth conditions similar to those used here. The WT dataset represents mostly (97.8%) the annotated genic regions, which cover 70.4% (19,308/27,416) of the annotated protein-coding genes. We classified the LUC insertion loci into four types by the combination of the transcriptional status of the LUC genes and the corresponding WT loci: (i) a LUC gene was transcribed in the WT transcribed region; (ii) a LUC gene was untranscribed in the WT transcribed region; (iii) a LUC gene was transcribed in the WT untranscribed region; and (iv) a LUC gene was untranscribed in the WT untranscribed region. The relative abundance of each type in T2-plants and T87 cells [26] was as follows: (i) 14.5% and 7.8%, (ii) 7.9% and 8.2%, (iii) 21.1% and 22.3% and (iv) 56.6% and 61.7%, respectively (Fig 3A). Based on these data, we evaluated the transcriptional activation rates in the WT untranscribed regions more precisely. We then redrew Fig 3A using the sum of types (iii) and (iv) as 100% (Fig 3B). In this data presentation, the transcriptional activation frequency was surprisingly similar between T2-plants and cultured cells (27.1 vs. 26.5 in Fig 3B), which supports the contention that de novo transcriptional activation in the untranscribed region occurs similarly in plants and cultured cells. Next, we performed the same analysis for the WT transcribed regions (types (i) and (ii)), and showed that the transcriptional activation frequency was higher in the T2-plants than in the cultured cells (66.0 vs. 48.7 in Fig 3C). This feature of the transcribed regions (Fig 3C) was reminiscent of the feature of the annotated genes (Fig 2D), because most of the WT transcribed regions (97.8%) represent annotated protein-coding genes. They both showed that the LUC inserts in the genic regions were activated more frequently in the T2-plants than in the cultured cells. The possible explanations for this feature from the viewpoint of plant life cycles and Km-based selection during the T2-plants establishment are referred to in the discussion section.

thumbnail
Download:
Fig 3. Transcription status of LUC insertion loci in T2-plants and WT plants.

(A) LUC insertion loci were classified by the combination of the transcription status of the LUC gene and the corresponding WT locus as follows; (i) LUC gene and the corresponding WT locus were both transcribed; (ii) LUC gene was untranscribed, but WT locus was transcribed; (iii) LUC gene was transcribed, but WT locus was untranscribed; and (iv) LUC gene and WT locus were both untranscribed. Upper panel, T2-plants; lower panel, T87 cells (Satoh et al., 2020). (B) Relative fractions of types (iii) and (iv) normalized to the sum of their abundances in (A), as 100%. (C) Relative fractions of types (i) and (ii) normalized to the sum of their abundances in (A), as 100%. (D) The transcription levels of T2-plants and corresponding inherent transcripts in WT plants were compared. R, Spearman’s correlation test.

https://doi.org/10.1371/journal.pone.0252674.g003

Generally, in promoter-trapping experiments, the expressed reporter genes are expected to reflect the activities of trapped endogenous promoters [42]. However, we previously found that the transcription levels of promoterless LUC genes did not reflect those of their inherent endogenous transcripts in the experiment that used cultured cells [26]. To confirm whether this feature was specific to the vegetatively growing cultured cells, we compared the transcription levels between T2-plants and their corresponding regions in the WT genome. We found that there was no correlation between them (Fig 3D). Thus, the observation that the trapping type of newly activated transcription events did not retain their inherent transcriptional status, at least in our experimental conditions, appeared to be a general feature of the plants and cultured cells. As insertions of the fragments were likely to disrupt the transcriptional activities of given loci, this result suggests two possibilities: (1) the original transcriptional activities had not yet been recovered in the vegetative propagation or within one generation; or (2) the transcriptional activities were overwritten by the de novo-activated transcription.

Chromatin remodelling occurred across the newly activated transcribed regions

Eukaryotic transcription is regulated by the control of the localization of transcription-related chromatin marks [5,6]. Therefore, next, we focused on the chromatin configuration around LUC inserts to examine whether the transcribed T2-plants were regulated by such chromatin marks. First, we screened T2-plants according to the following criteria: the existence of transcription evidence in the TRIP experiment, and an unlikeliness to be affected by the pre-existing promoters or transcription units. Based on these criteria, we finally selected two lines: T2:161 and T2:205 (Fig 4A and 4B). The T2:161 line was classified as a Genic AS type in which the LUC insert was found in the opposite strand of an endogenous gene (AT3G23750) (Fig 4A). In the T2:205 line, the LUC insert was located in an intergenic region, in which an endogenous gene (AT5G01110) was detected downstream of the LUC insert on the opposite strand (Fig 4B). Transcription of inserted promoterless LUC genes was verified in both lines by reverse-transcription quantitative-PCR (Figs 4C and S2), whereas no RNA-sequencing reads were mapped on the same strand of each LUC insert in the corresponding WT genome, indicating that they were inherently untranscribed regions. For these two lines, we scanned the localization of chromatin marks around the LUC insertion loci and compared them with those obtained from the WT genome. In this study, we analysed three transcription-related chromatin marks: methylated cytosine (mC), lysine 36 tri-methylation of histone H3 (H3K36me3), and the histone variant H2A.Z. In the WT genome, enrichments of mC and H3K36me3 were observed within the gene bodies of AT3G23750 and AT5G01110, respectively (Fig 4D and 4E, upper and middle panels), which agreed with the general properties of these chromatin marks [43,44]. However, in the T2-plants, these two chromatin marks were not found within the LUC gene bodies (Fig 4D and 4E, upper and middle panels). Although weak signals were observed 200 bp upstream from the LUC insert in the T2:161 line (Fig 4D, upper and middle panels), they reflected the chromatin marks of the WT allele in the T2-plant, because these plants were not homozygous. Conversely, the localization patterns of the H2A.Z variant were clearly different from those of the other two chromatin marks (Fig 4D and 4E, lower panels). Both lines showed significant enrichments of H2A.Z throughout the LUC gene bodies, while there were almost no H2A.Z signals in the corresponding regions in the WT genome (Fig 4D and 4E, lower panels). Although H2A.Z is a marker histone for the promoter region, it also appears in the gene bodies of genes with low expression [45–47]. In addition, mC and H3K36me3 were reportedly deposited within a gene body in a transcription-coupled manner [48], which would be undetectable in the low-expressed genes [49]. Thus, these distribution patterns of chromatin marks in the T2-plants were plausible because the transcriptional strength of these two lines was low compared with that of the constitutive genes (Figs 4C and S2).

thumbnail
Download:
Fig 4. Localization analysis of chromatin marks in selected T2-plants.

(A and B) Locus details of the (A) T2:161 and (B) T2:205 lines. The genomic loci of LUC inserts are represented as the individual position of the RB–genome junction. (C) Transcription levels of the T2:161 and T2:205 lines relative to the endogenous UBQ10 gene (AT4G05320). (D and E) Localization patterns of three chromatin marks (mC: upper panel; H3K36me3: middle panel; and H2A.Z: lower panel) around individual LUC insertion loci of the (D) T2:161 and (E) T2:205 lines. Individual localization signals were normalized to the enrichment of the control locus of each chromatin mark (see Materials and Methods) as 100%. The red bars indicate the analysed positions, which were normalized to the genomic position of the start codon of LUC inserts as zero. Error bar, ±SD of two biological replicates.

https://doi.org/10.1371/journal.pone.0252674.g004

In the T2:161 line, H2A.Z was newly localized 200 bp upstream from the LUC insert (Fig 4D, lower panel), which suggests that chromatin remodelling occurred even outside the LUC insert. We hypothesized that H2A.Z is localized throughout the transcribed region of the LUC insert. To confirm this hypothesis, next we analysed the TSS of LUC inserts. However, it was challenging to determine the TSSs of T2-plants using general methods [50,51] because of the low transcription levels of these plants. The template-switching method has the advantage of yielding full-length cDNAs from low-input RNA [52]. In this study, we applied inverse PCR to this template-switching method to specifically amplify the full-length cDNAs of LUC genes. Based on this method, we analysed TSS distribution in T2-plants. Unfortunately, the transcription level of the T2:205 line was too low to obtain any TSS signals. Conversely, in the T2:161 line, a TSS was found ~1.1 kb upstream of the LUC insertion locus (Figs 5A and S3). Sanger sequencing revealed that this transcript was spliced (Figs 5A and S3). We reanalysed the distribution profiles of H3K36me3 and H2A.Z around the determined TSS (Fig 5B). There was no significant enrichment of H3K36me3 around the LUC-TSS, as the enrichment levels were almost the same among the transgenic plants and the WT genome (Fig 5B, upper panel). In contrast, we observed that H2A.Z was newly localized starting from the LUC-TSS, whereas H2A.Z was not observed in the corresponding locus in the WT genome (Fig 5B, lower panel).

thumbnail
Download:
Fig 5. Localization of chromatin marks around the transcription start site of the T2:161 line.

(A) Transcription start site of the T2:161 line, as determined using a template-switching-based method. (B) Localization analysis of H2A.Z and H3K36me3 in T2:161 and WT plants, as in Fig 4D.

https://doi.org/10.1371/journal.pone.0252674.g005

Overall, the chromatin and TSS analyses revealed that exogenously inserted promoterless genes acquired a brand-new chromatin configuration, and that such chromatin remodelling occurred throughout the newly activated transcription unit. In addition, this chromatin remodelling might have been involved in the transcriptional behaviour of the trapping type of LUC transcription (Fig 3D): de novo-activated transcription events concomitant with the chromatin remodelling might overwrite their inherent transcriptional status.

Discussion

In this study, based on the large-scale promoter-trap screening of A. thaliana plants, we demonstrated the genetic behaviour of the newly activated transcription of exogenous genes. A comparison with the results of a previous study using cultured cells [26] showed that de novo transcriptional activation is an inheritable phenomenon of the plant genome (Figs 1–3). We also demonstrated that chromatin remodelling occurred across the transcribed regions of the inserted coding sequences in the selected two transgenic lines (Figs 4 and 5), which probably regulated the newly activated transcription of these loci by overwriting the inherent chromatic and transcriptional status.

In the T2:161 line, the TSS was located on the 3′ end of an endogenous gene (AT3G23750), where no detectable transcripts existed in the WT genome (Fig 5A). It is plausible to propose that this was caused by activating (rather than trapping) a cryptic antisense transcript of the given locus [27]. Conversely, we speculated that the T2:205 line may be transcribed from a de novo-activated TSS located in the proximal intergenic region, although we could not identify this TSS in this study. This speculation was based on a previous finding from the cultured cell experiment: de novo TSS occurs about 100 bp upstream of the inserted coding sequences in the intergenic region [27]. The localization pattern of H2A.Z in the T2:205 line agreed with this prediction, as the H2A.Z signal clearly dropped to almost zero at 200 bp upstream of the LUC insert (Fig 4E).

Generally, in promoter-trap screening, transgenic lines are screened based on the expression of the inserted promoterless reporter genes [42]. In contrast, we did not carry out the screening of T2-plants according to the expression of LUC genes; rather, we selected them according to the activity of the co-transformed Km-resistance gene (Fig 1). This selection method enabled the isolation of lines without the selection bias that was caused by the transcription levels of the LUC genes. However, we found differences between the results of plants and cultured cells, despite the similar experimental conditions used in the two experiments. For instance, compared with the cultured cells, plants were more prone to be transcriptionally activated by the trapping of endogenous gene/transcripts (Figs 2D and 3C), and the transcriptional strength of such activated transcription tended to be bipolarized to lower and higher transcription levels according to the insertion type (Fig 2F). How can these features of T2-plants be explained? Although transgenic cultured cells were regarded as the T1 generation, we used the T2 generation of transgenic plants in this study. Plants require a greater number of genes than do cultured cells during this one-cycle generation, because plants experience germination, development, differentiation and sexual reproduction, while the cultured cells are only in the state of vegetative propagation in a constant culture condition. Gene-insertion events might cause lethal effects on a certain population of transgenic plants by disrupting various genes that are essential for their growth over the life cycle [53]. Therefore, although we assumed that the T2-plant lines were established under a non-selective condition for LUC activity, the population might be distorted through a generation. Km-based selection might also affect the T2-plant population; T-DNA insertion sometimes fails to confer Km resistance and causes embryonic lethality [54,55]. In addition, under the selective condition, T-DNAs tended to be inserted in open-chromatin and hypomethylated regions [56]. Thus, Km-based selection might enrich transgenic lines in which inserts were located in the transcriptionally permissive regions where the Km-resistance genes could function sufficiently. We observed a weak insertion preference of LUC genes in the accessible chromatin regions (S1 Table) by utilizing a Plant Chromatin State Database [57] (http://systemsbiology.cau.edu.cn/chromstates). However, we could not evaluate any clear correlation between the transcriptional activation of promoterless LUC genes and the chromatin states of the corresponding WT loci, probably because the detected LUC population was not sufficiently large for such an analysis. Overall, the transcriptional fates of promoterless LUC inserts were likely to be affected by the experienced life stages and selective conditions during the establishment of transgenic plants. Hence, to grasp the extent to which inserted promoterless coding genes actually become transcribed in plants, alternative experimental strategies are needed; for example, selection-free transformation or the use of a binary vector system to introduce reporter and selection marker genes independently [58].

To date, studies of the evolutionary processes by which genetic novelty emerges were mainly led by comparative genomics [19,24,59–61]. However, because such genomics approaches are established based solely on the evolutionary winners, the resultant scenario lacks perspective from the great majority of evolutionary losers. The resolution depends on the divergence time, ranging from millions to billions of years. Conversely, our artificial evolutionary approach sheds light even on evolutionary losers within a much shorter timescale. Specifically, as the LUC genes used in this study are not profitable for plants, most of them would presumably become silenced or pseudogenized, while a few of them might occasionally be retained. How many generations and populations are needed to reach such endings? Our approach based on the use of plants will reveal the types of genomic/epigenomic variations that become winners/losers, thus enabling the tracing of the fates of newly activated transcripts in the population over the generations. In contrast, the cultured cells will be a useful model to investigate the molecular mechanisms underlying promoter birth, thus providing a homogeneous and simple experimental system.

It is intriguing to utilize a stress-tolerance/inducible gene as a promoterless reporter gene in our artificial evolutionary experiment. This would be a useful model to investigate how newly emerged genes adapt and evolve against exposed stress or selective environments. It is also interesting to try such experiments among different developmental phases and tissues. For example, the promoterless genes might be more prone to be transcribed in the pollen, where new genes often arise because of the transcriptionally permissive status caused by the accessible chromatin configuration [62]. Such an approach allows the investigation of gene evolution in multicellular organisms, thus providing insights into how newly emerged genes become integrated into pre-existing spatio-temporal genetic networks.

In conclusion, our artificial evolutionary experiment provided insight into the initial genetic behaviour of newly activated transcription in the plant genome. We showed that the de novo-activated transcription accompanying the local chromatin remodelling was inheritable. To evaluate the contribution of this phenomenon to the plant genome evolution, examination of the genetic behaviour of the de novo transcribed genes over an increasing number of generations with/without selective pressures will provide further clues regarding this phenomenon.

Materials and methods

Plant materials and transformation

A. thaliana (ecotype; Col-0) plants were grown at 23°C with continuous illumination (20–50 μmol m–2 s–1). Ti-plasmid libraries containing short sequences (5’-aggcctcgacgttatcagcttacag-3’), a 12 bp random sequence (‘barcode’), a promoterless LUC-coding sequence, a nos-terminator and an expression cassette of a kanamycin (Km)-resistance gene between the left (LB) and right (RB) borders of the T-DNA were constructed using a modified pGreenII vector [63,64]. Agrobacterium tumefaciens (GV3101) cells were transformed with the Ti-plasmid libraries. Agrobacterium-mediated transformation of A. thaliana was performed according to the floral-dip method [37]. Transformed seeds were selected on Murashige and Skoog (MS) medium [1× strength of MS plant salt mixture (Nihon Pharmaceutical), 1% sucrose, 0.05% MES, 0.8% agar, pH 5.7] supplemented with 25 μg ml–1 of Km. The screened 386 individual Km-resistant T1 seedlings were grown at 23°C with continuous illumination (20–50 μmol m–2 s–1). The seeds of individual T2-generation plants were harvested. For the promoter-trap experiment, three seeds of individual T2-plants were stratified at 4°C in the dark for 2 days, then grown on MS medium [half-strength MS medium including vitamins (Duchefa Biochemie), 1% sucrose, 0.8% agar, pH 5.7] at 23°C with continuous illumination (40–60 μmol m–2 s–1) for 10 days. All seedlings were harvested and ground in liquid nitrogen to a fine powder, for thorough mixing. DNA and RNA were extracted using a DNeasy Plant Mini Kit (QIAGEN) and RNeasy Plant Mini Kit (QIAGEN), respectively, and subjected to the preparation of the NGS libraries.

Determination of the LUC insertion sites

NGS libraries for determining LUC insertion loci were prepared according to the TRIP method [38,65] with modifications as follows. Genomic DNA (2.0 μg) was digested completely with DpnII, MseI or ApoI, and then purified using the QIAquick PCR purification Kit (QIAGEN). Each digested DNA (600 ng) was independently circularized with T4 DNA ligase. An aliquot of each circularized DNA was subjected to inverse PCR using primer sets that hybridize within the LUC ORF. Subsequently, NGS libraries were prepared by two rounds of PCR; the first round was performed to add Illumina adapters, and the second was carried out using Nextera XT index primers (Illumina). Sequencing was performed using a 301 bp paired-ended protocol on an Illumina MiSeq platform. All primers used in this study are listed in S2 Table.

For the determination of each LUC insertion site, NGS reads were first processed, before mapping to the genome according to a method previously described [27], with the following modifications. NGS reads were aligned to the T-DNA vector sequence (5’- tcaaggcctcgacgttatcagcttacagNNNNNNNNNNNNATGGAAGACGCCAAAAACATAAAGAAAGGCCCGGCGCCATTCTATCCTCTAGAG-3’; lowercase, border sequence; N, barcode; underlined, LUC fraction) using Blastn (version: 2.4.0+) [66], to obtain individual flanking sequences from the LUC insert and barcode. The obtained flanking sequences were mapped on the TAIR10 version of the A. thaliana genome using bowtie [67] allowing three mismatches. Precise locus–barcode pairs were determined according to the following criteria: (1) at least two read counts; (2) the read count of the most frequent locus–barcode pair accounted for ≥60% of them, including their PCR/sequencing artefacts; and (3) exclusion from subsequent analysis of two or more distinct LUC inserts with the same barcode sequences. The insertion loci of T2-plants were classified according to the TAIR10 version of the genomic annotation of A. thaliana under the following classification: genomic regions where annotated protein-coding genes were defined as ‘Genic’ regions, whereas the remainder of the genome was classified as ‘Intergenic’. The insertion strand of LUC genes was considered.

Determination of the relative transcription level of LUC genes

NGS libraries for determining LUC transcription level were prepared according to the TRIP method [38,65] with modifications as follows. RNA (5.0 μg) was subjected to reverse transcription using an oligo dT15 primer and SuperScript III Reverse Transcriptase (Thermo Fisher Scientific). An NGS library (termed RNA library, hereafter) was prepared by amplification of the barcode region of LUC-cDNA using primer sets with an Illumina adapter extension, followed by the tailed-PCR using Nextera XT index primers. From an aliquot of DNA used in the LUC insertion site determination, barcode regions were amplified and an NGS library (termed DNA library, hereafter) was prepared according to the method described above. Sequencing was performed on the Illumina MiSeq under a 76 bp paired-ended protocol.

To obtain an indicator of the molecular abundances of each LUC-mRNA per transgenic cells, barcode sequences were extracted from the sequencing reads and counted. Barcodes with a read number ≤5 in the DNA library were omitted from further analysis. In the RNA library, barcodes with a read number ≤5 were set as zero. For DNA and RNA libraries, the read number of each barcode was normalized to the total sequencing reads of the corresponding library. The relative transcription level of each LUC gene was calculated as follows: the RNA read number of each barcode was divided by the corresponding DNA read number and multiplied by 10,000. Subsequently, individual LUC loci and transcription levels were associated based on their barcode sequences.

Validation of LUC insertion loci and barcode sequences

Randomly chosen T2-plants were stratified at 4°C in the dark for 2 days, then grown on MS medium supplemented with 25 μg ml–1 of Km at 23°C with continuous illumination (20–30 μmol m–2 s–1) for 10 days. Km-resistant seedlings were harvested and subjected to DNA extraction. Four types of PCR were performed to amplify the barcode region, the RB–genome junction, the LB–genome junction and the T-DNA insert, respectively. The PCR products obtained were then analysed by agarose gel electrophoresis and Sanger sequencing, for validation of the insertion locus and barcode sequence, respectively.

Comparison with WT transcriptome data

RNA-seq data of WT A. thaliana (col-0) plants were retrieved from the NCBI Short-Read Archive under accessions SRR6388204, SRR6388205 and SRR770510. The sequencing reads were subjected to adapter trimming and quality trimming, followed by mapping to the A. thaliana genome (TAIR10) using STAR (v2.5.3) [68] with the following parameters: STAR–alignIntronMax 6000 –outSAMstrandField intronMotif–two passMode Basic. Transcribed regions and their transcription levels (in fragments per kilobase of exon per million reads mapped (FPKM)) were analysed using StringTie (v2.1.4) [69]. Subsequently, the transcription level of each T2-plant was compared with the FPKM of the inherent transcribed region in the WT genome. In the case of inherent transcripts with multiple isoforms, each FPKM was summed up.

Chromatin immunoprecipitation (ChIP) and MBD immunoprecipitation (MBDIP) analysis

The T2:161 and T2:205 lines were stratified at 4°C in the dark for 3 days, then grown on MS medium [half-strength MS medium including vitamins (Duchefa Biochemie), 1% sucrose, 0.8% agar, pH 5.7] supplemented with 15 μg ml–1 of Km at 23°C with continuous illumination (20–30 μmol m–2 s–1) for 8 days. Km-resistant seedlings were harvested and subjected to ChIP and MBDIP analysis. For control experiments, transgenic A. thaliana harbouring an expression cassette of the Km-resistance gene without the LUC reporter gene (termed WT in Figs 4D and 4E, and 5B) were prepared and grown under the same condition as that used for T2-plants. ChIP and MBDIP were performed according to a method previously described [27], with the following modifications. For the ChIP assay, ~10 ng of solubilized chromatin (median, 200 bp) and antibodies (2.4 μg of an anti-H2A.Z antibody [70] and 2.0 μg of an anti-H3K36me3 antibody (Abcam: ab9050), respectively) were used for each experiment. For the MBDIP assay, the methylated DNA fraction (mC) was collected from 1.0 μg of sheared DNA (median, 200 bp) using an EpiXplore Methylated DNA Enrichment Kit (Clontech) according to the manufacturer’s instructions. Successful enrichment of ChIPed DNA and mC was validated by qPCR in the control sites (S2 Table) according to Deal et al. [71] for H2A.Z, to Yang et al. [72] for H3K36me3 and to Erdmann et al. [73] for mC. In both T2-plants and WT, relative enrichments of H2A.Z, H3K36me3 and mC around the LUC insertion loci were calculated based on the enrichment of the control sites, which was set as 100%.

Expression and TSS analysis

The T2:161 and T2:205 lines were grown and harvested under the same condition as that used for the ChIP experiments. Total RNA was isolated using an RNeasy Plant Mini Kit followed by DNase I treatment. For expression analysis, cDNA was synthesized from 5.0 μg of the total RNA using an oligo dT20 primer and SuperScript III Reverse Transcriptase. The transcription level of the LUC gene was normalized to that of the ubiquitin gene (UBQ10: AT4G05320).

LUC-TSS was analysed according to a published method [52,74], with the following modifications. Specifically, polyadenylated RNA was extracted using a Dynabeads mRNA Purification Kit (Invitrogen) according to the manufacturer’s protocol. Polyadenylated RNA (1.0 μg) was used for reverse-transcription and template-switching reactions. During these reactions, SgfI sites were added at both ends of the full-length cDNA by the primer used for reverse transcription and the template-switching oligo. The full-length cDNAs obtained were then digested completely by SgfI. Subsequently, digested cDNAs were circularized and subjected to inverse PCR to specifically amplify LUC-containing cDNAs. The resulting nested PCR products were analysed by Sanger sequencing.

Acknowledgments

We thank Moyuru Shirasu for his help in maintaining the transgenic A. thaliana plants.

References

  1. 1. Kaessmann H. Origins, evolution, and phenotypic impact of new genes. Genome Res. 2010;20(10):1313–1326. pmid:20651121
  2. 2. Cardoso-Moreira M, Long M. The origin and evolution of new genes. Methods Mol Biol. 2012;856:161–186. pmid:22399459
  3. 3. McLysaght A, Guerzoni D. New genes from non-coding sequence: the role of de novo protein-coding genes in eukaryotic evolutionary innovation. Philos Trans R Soc Lond B Biol Sci. 2015;370(1678):20140332. pmid:26323763
  4. 4. Van Oss SB, Carvunis AR. De novo gene birth. PLoS Genet. 2019;15(5):e1008160. pmid:31120894
  5. 5. Haberle V, Stark A. Eukaryotic core promoters and the functional basis of transcription initiation. Nat Rev Mol Cell Biol. 2018;19(10):621–637. pmid:29946135
  6. 6. Andersson R, Sandelin A. Determinants of enhancer and promoter activities of regulatory elements. Nat Rev Genet. 2020;21(2):71–87. pmid:31605096
  7. 7. McLysaght A, Hurst LD. Open questions in the study of de novo genes: what, how and why. Nat Rev Genet. 2016;17(9):567–578. pmid:27452112
  8. 8. Schlötterer C. Genes from scratch—the evolutionary fate of de novo genes. Trends Genet. 2015;31(4):215–219. pmid:25773713
  9. 9. Xie C, Zhang YE, Chen JY, Liu CJ, Zhou WZ, Li Y, et al. Hominoid-specific de novo protein-coding genes originating from long non-coding RNAs. PLoS Genet. 2012;8(9):e1002942. pmid:23028352
  10. 10. Ruiz-Orera J, Messeguer X, Subirana JA, Alba MM. Long non-coding RNAs as a source of new peptides. Elife. 2014;3:e03523. pmid:25233276
  11. 11. Kalitsis P, Saffery R. Inherent promoter bidirectionality facilitates maintenance of sequence integrity and transcription of parasitic DNA in mammalian genomes. BMC Genomics. 2009;10:498. pmid:19860919
  12. 12. Gotea V, Petrykowska HM, Elnitski L. Bidirectional promoters as important drivers for the emergence of species-specific transcripts. PLoS One. 2013;8(2):e57323. pmid:23460838
  13. 13. Neme R, Tautz D. Phylogenetic patterns of emergence of new genes support a model of frequent de novo evolution. BMC Genomics. 2013;14:117. pmid:23433480
  14. 14. Vakirlis N, Hebert AS, Opulente DA, Achaz G, Hittinger CT, Fischer G, et al. A Molecular Portrait of De Novo Genes in Yeasts. Mol Biol Evol. 2018;35(3):631–645. pmid:29220506
  15. 15. Long HK, Prescott SL, Wysocka J. Ever-Changing Landscapes: Transcriptional Enhancers in Development and Evolution. Cell. 2016;167(5):1170–1187. pmid:27863239
  16. 16. Werner MS, Sieriebriennikov B, Prabh N, Loschko T, Lanz C, Sommer RJ. Young genes have distinct gene structure, epigenetic profiles, and transcriptional regulation. Genome Res. 2018;28(11):1675–1687. pmid:30232198
  17. 17. Majic P, Payne JL. Enhancers Facilitate the Birth of De Novo Genes and Gene Integration into Regulatory Networks. Mol Biol Evol. 2020;37(4):1165–1178. pmid:31845961
  18. 18. Zhou Q, Zhang G, Zhang Y, Xu S, Zhao R, Zhan Z, et al. On the origin of new genes in Drosophila. Genome Res. 2008;18(9):1446–1455. pmid:18550802
  19. 19. Carvunis AR, Rolland T, Wapinski I, Calderwood MA, Yildirim MA, Simonis N, et al. Proto-genes and de novo gene birth. Nature. 2012;487(7407):370–374. pmid:22722833
  20. 20. Wu X, Sharp PA. Divergent transcription: a driving force for new gene origination? Cell. 2013;155(5):990–996. pmid:24267885
  21. 21. Ruiz-Orera J, Hernandez-Rodriguez J, Chiva C, Sabidó E, Kondova I, Bontrop R, et al. Origins of De Novo Genes in Human and Chimpanzee. PLoS Genet. 2015;11(12):e1005721. pmid:26720152
  22. 22. Neme R, Tautz D. Fast turnover of genome transcription across evolutionary time exposes entire non-coding DNA to de novo gene emergence. Elife. 2016;5:e09977. pmid:26836309
  23. 23. Durand É, Gagnon-Arsenault I, Hallin J, Hatin I, Dubé AK, Nielly-Thibault L, et al. Turnover of ribosome-associated transcripts from de novo ORFs produces gene-like characteristics available for de novo gene emergence in wild yeast populations. Genome Res. 2019;29(6):932–943. pmid:31152050
  24. 24. Zhang L, Ren Y, Yang T, Li G, Chen J, Gschwend AR, et al. Rapid evolution of protein diversity by de novo origination in Oryza. Nat Ecol Evol. 2019;3(4):679–690. pmid:30858588
  25. 25. Blevins WR, Ruiz-Orera J, Messeguer X, Blasco-Moreno B, Villanueva-Cañas JL, Espinar L, et al. Uncovering de novo gene birth in yeast using deep transcriptomics. Nat Commun. 2021;12(1):604. pmid:33504782
  26. 26. Satoh S, Hata T, Takada N, Tachikawa M, Mitsuhiro M, Kushnir S, et al. Plant genome response to incoming coding sequences: stochastic transcriptional activation independent of integration loci. BioRxiv [Preprint]. 2020 bioRxiv 401992 [posted 2020 Nov 28; revised 2021 Feb 4]. Available from: https://www.biorxiv.org/content/10.1101/2020.11.28.401992v2
  27. 27. Hata T, Satoh S, Takada N, Matsuo M, Obokata J. Kozak sequence acts as a negative regulator for de novo transcription initiation of newborn coding sequences in the plant genome. Mol Biol Evol. 2021. msab069, pmid:33705557
  28. 28. Slane D, Kong J, Berendzen KW, Kilian J, Henschen A, Kolb M, et al. Cell type-specific transcriptome analysis in the early Arabidopsis thaliana embryo. Development. 2014;141(24):4831–4840. pmid:25411212
  29. 29. Palovaara J, Saiga S, Wendrich JR, van ’t Wout Hofland N, van Schayck JP, Hater F, et al. Transcriptome dynamics revealed by a gene expression atlas of the early Arabidopsis embryo. Nat Plants. 2017;3(11):894–904. pmid:29116234
  30. 30. Shulse CN, Cole BJ, Ciobanu D, Lin J, Yoshinaga Y, Gouran M, et al. High-Throughput Single-Cell Transcriptome Profiling of Plant Cell Types. Cell Rep. 2019;27(7):2241–2247.e4. pmid:31091459
  31. 31. Shi D, Jouannet V, Agustí J, Kaul V, Levitsky V, Sanchez P, et al. Tissue-specific transcriptome profiling of the Arabidopsis inflorescence stem reveals local cellular signatures. Plant Cell. 2020. pmid:33582756
  32. 32. Ingouff M, Rademacher S, Holec S, Soljić L, Xin N, Readshaw A, et al. Zygotic resetting of the HISTONE 3 variant repertoire participates in epigenetic reprogramming in Arabidopsis. Curr Biol. 2010;20(23):2137–2143. pmid:21093266
  33. 33. Jullien PE, Susaki D, Yelagandula R, Higashiyama T, Berger F. DNA methylation dynamics during sexual reproduction in Arabidopsis thaliana. Curr Biol. 2012;22(19):1825–1830. pmid:22940470
  34. 34. Kawakatsu T, Nery JR, Castanon R, Ecker JR. Dynamic DNA methylation reconfiguration during seed development and germination. Genome Biol. 2017;18(1):171. pmid:28911331
  35. 35. Tao Z, Shen L, Gu X, Wang Y, Yu H, He Y. Embryonic epigenetic reprogramming by a pioneer transcription factor in plants. Nature. 2017;551(7678):124–128. pmid:29072296
  36. 36. Gehring M. Epigenetic dynamics during flowering plant reproduction: evidence for reprogramming? New Phytol. 2019;224(1):91–96. pmid:31002174
  37. 37. Clough SJ, Bent AF. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 1998;16(6):735–743. pmid:10069079
  38. 38. Akhtar W, de Jong J, Pindyurin AV, Pagie L, Meuleman W, de Ridder J, et al. Chromatin position effects assayed by thousands of reporters integrated in parallel. Cell. 2013;154(4):914–927. pmid:23953119
  39. 39. De Buck S, Podevin N, Nolf J, Jacobs A, Depicker A. The T-DNA integration pattern in Arabidopsis transformants is highly determined by the transformed target cell. Plant J. 2009;60(1):134–145. pmid:19508426
  40. 40. Kim SI, Veena , Gelvin SB. Genome-wide analysis of Agrobacterium T-DNA integration sites in the Arabidopsis genome generated under non-selective conditions. Plant J. 2007;51(5):779–791. pmid:17605756
  41. 41. Berardini TZ, Reiser L, Li D, Mezheritsky Y, Muller R, Strait E, et al. The Arabidopsis information resource: Making and mining the "gold standard" annotated reference plant genome. Genesis. 2015;53(8):474–485. pmid:26201819
  42. 42. Springer PS. Gene traps: tools for plant development and genomics. Plant Cell. 2000;12(7):1007–1020. pmid:10899970
  43. 43. Jones PA. Functions of DNA methylation: islands, start sites, gene bodies and beyond. Nat Rev Genet. 2012;13(7):484–492. pmid:22641018
  44. 44. Wagner EJ, Carpenter PB. Understanding the language of Lys36 methylation at histone H3. Nat Rev Mol Cell Biol. 2012;13(2):115–126. pmid:22266761
  45. 45. Lashgari A, Millau JF, Jacques P, Gaudreau L. Global inhibition of transcription causes an increase in histone H2A.Z incorporation within gene bodies. Nucleic Acids Res. 2017;45(22):12715–12722. pmid:29036442
  46. 46. Gómez-Zambrano Á, Merini W, Calonje M. The repressive role of Arabidopsis H2A.Z in transcriptional regulation depends on AtBMI1 activity. Nat Commun. 2019;10(1):2828. pmid:31249301
  47. 47. Lei B, Frederic B. H2A Variants in Arabidopsis: Versatile Regulators of Genome Activity. Plant Communications. 2020;1:100015. pmid:33404536
  48. 48. Teissandier A, Bourc’his D. Gene body DNA methylation conspires with H3K36me3 to preclude aberrant transcription. EMBO J. 2017;36(11):1471–1473. pmid:28442531
  49. 49. Cermakova K, Smith E, Veverka V, Hodges H. Dynamics of transcription-dependent H3K36me3 marking by the SETD2:ISW1:SPT6 ternary complex. BioRxiv [Preprint]. bioRxiv [posted 2019 May 14]. Available from: https://www.biorxiv.org/content/10.1101/636084v1
  50. 50. Maruyama K, Sugano S. Oligo-capping: a simple method to replace the cap structure of eukaryotic mRNAs with oligoribonucleotides. Gene. 1994;138(1–2):171–174. pmid:8125298
  51. 51. Carninci P, Kvam C, Kitamura A, Ohsumi T, Okazaki Y, Itoh M, et al. High-efficiency full-length cDNA cloning by biotinylated CAP trapper. Genomics. 1996;37(3):327–336. pmid:8938445
  52. 52. Salimullah M, Sakai M, Mizuho S, Plessy C, Carninci P. NanoCAGE: a high-resolution technique to discover and interrogate cell transcriptomes. Cold Spring Harb Protoc. 2011;2011(1):pdb.prot5559. pmid:21205859
  53. 53. Meinke DW. Genome-wide identification of EMBRYO-DEFECTIVE (EMB) genes required for growth and development in Arabidopsis. New Phytol. 2020;226(2):306–325. pmid:31334862
  54. 54. Errampalli D, Patton D, Castle L, Mickelson L, Hansen K, Schnall J, et al. Embryonic Lethals and T-DNA Insertional Mutagenesis in Arabidopsis. Plant Cell. 1991;3(2):149–157. pmid:12324593
  55. 55. Francis KE, Spiker S. Identification of Arabidopsis thaliana transformants without selection reveals a high occurrence of silenced T-DNA integrations. Plant J. 2005;41(3):464–477. pmid:15659104
  56. 56. Shilo S, Tripathi P, Melamed-Bessudo C, Tzfadia O, Muth TR, Levy AA. T-DNA-genome junctions form early after infection and are influenced by the chromatin state of the host genome. PLoS Genet. 2017;13(7):e1006875. pmid:28742090
  57. 57. Liu Y, Tian T, Zhang K, You Q, Yan H, Zhao N, et al. PCSD: a plant chromatin state database. Nucleic Acids Res. 2018;46(D1):D1157–D1167. pmid:29040761
  58. 58. Komari T, Hiei Y, Saito Y, Murai N, Kumashiro T. Vectors carrying two separate T-DNAs for co-transformation of higher plants mediated by Agrobacterium tumefaciens and segregation of transformants free from selection markers. Plant J. 1996;10(1):165–174. pmid:8758986
  59. 59. Zhao L, Saelao P, Jones CD, Begun DJ. Origin and spread of de novo genes in Drosophila melanogaster populations. Science. 2014;343(6172):769–772. pmid:24457212
  60. 60. Li ZW, Chen X, Wu Q, Hagmann J, Han TS, Zou YP, et al. On the Origin of De Novo Genes in Arabidopsis thaliana Populations. Genome Biol Evol. 2016;8(7):2190–2202. pmid:27401176
  61. 61. Li C, Lenhard B, Luscombe NM. Integrated analysis sheds light on evolutionary trajectories of young transcription start sites in the human genome. Genome Res. 2018;28(5):676–688. pmid:29618487
  62. 62. Wu DD, Wang X, Li Y, Zeng L, Irwin DM, Zhang YP. "Out of pollen" hypothesis for origin of new genes in flowering plants: study from Arabidopsis thaliana. Genome Biol Evol. 2014;6(10):2822–2829. pmid:25237051
  63. 63. Hellens RP, Edwards EA, Leyland NR, Bean S, Mullineaux PM. pGreen: a versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol Biol. 2000;42(6):819–832. pmid:10890530
  64. 64. Hirashima M, Satoh S, Tanaka R, Tanaka A. Pigment shuffling in antenna systems achieved by expressing prokaryotic chlorophyllide a oxygenase in Arabidopsis. J Biol Chem. 2006;281(22):15385–15393. pmid:16574646
  65. 65. Akhtar W, Pindyurin AV, de Jong J, Pagie L, Ten Hoeve J, Berns A, et al. Using TRIP for genome-wide position effect analysis in cultured cells. Nat Protoc. 2014;9(6):1255–1281. pmid:24810036
  66. 66. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, et al. BLAST+: architecture and applications. BMC Bioinformatics. 2009;10:421. pmid:20003500
  67. 67. Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009;10(3):R25. pmid:19261174
  68. 68. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21. pmid:23104886
  69. 69. Pertea M, Pertea GM, Antonescu CM, Chang TC, Mendell JT, Salzberg SL. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol. 2015;33(3):290–295. pmid:25690850
  70. 70. Kudo H, Matsuo M, Satoh S, Hachisu R, Nakamura M, Yamamoto Y, Yoshiharu, et al. Cryptic promoter activation occurs by at least two different mechanisms in the Arabidopsis genome. BioRxiv 399337 [Preprint]. bioRxiv [posted 2020 Nov 28]. Available from: https://www.biorxiv.org/content/10.1101/2020.11.28.399337v1
  71. 71. Deal RB, Topp CN, McKinney EC, Meagher RB. Repression of flowering in Arabidopsis requires activation of FLOWERING LOCUS C expression by the histone variant H2A.Z. Plant Cell. 2007;19(1):74–83. pmid:17220196
  72. 72. Yang H, Howard M, Dean C. Antagonistic roles for H3K36me3 and H3K27me3 in the cold-induced epigenetic switch at Arabidopsis FLC. Curr Biol. 2014;24(15):1793–1797. pmid:25065750
  73. 73. Erdmann RM, Souza AL, Clish CB, Gehring M. 5-hydroxymethylcytosine is not present in appreciable quantities in Arabidopsis DNA. G3 (Bethesda). 2014;5(1):1–8. pmid:25380728
  74. 74. Plessy C, Bertin N, Takahashi H, Simone R, Salimullah M, Lassmann T, et al. Linking promoters to functional transcripts in small samples with nanoCAGE and CAGEscan. Nat Methods. 2010;7(7):528–534. pmid:20543846