Because moss leaves have only one layer of cells, it is difficult to get the sterilized moss. To obtain sterile materials, NaClO was used to detoxify the gametophyte materials collected from the wild. we explored the effect of NaClO treatment time on the sterilization efficiency of B. argenteum. The results showed that 5 minutes of treatment with 5% NaClO had the best sterilization effect. Treated with 5% NaClO for 5 min, the gametophyte was grown on sterile culture media, B.argenteum became lighter in color. After several days of cultivation, almost all the materials became whitened and died. Still, after seven days, new green filamentous structures could grow from the top of some stems and leaves under a dissecting microscope (Fig. 1A). We divided the sterilization status of each stem tip into three categories: sterilization leading to the death of the entire plant, survival with bacterial growth, and survival without bacterial growth. The rates of sterilization and survival of B.argenteum were obtained (Fig. 1B).
Some gametophyte shoots were taken about 4 mm into liquid BCD medium with or without additional N or C sources. In liquid culture, the protonema cells grow from gametophyte stem and leaves and forms fuzzy balls. As shown in (Fig. 1C), on the 7th day of liquid culture, the gametophytes with additional carbon sources developed new gametophyte. In contrast, the samples without carbon sources mainly remain in the original filamentous phase. Continuing to culture until the 14th day, the gametophytes with added carbon sources grow longer, and the spherical bodies become more extensive. In contrast, the samples without carbon sources remain filamentous. After 21 days of culture, the spherical bodies of the two groups with added carbon sources continue to grow, and the number of gametophytes increases further. However, the gametophytes with added carbon and nitrogen sources are covered with more original protonema. The spherical bodies of the two groups without added carbon sources continue to grow larger, and more original protonema emerge. The sample with added nitrogen source but no carbon source has a much larger diameter, the longest original protonema, and almost no stem and leaf tissues, indicating that glucose promotes the development of protonema into gametophytes. Similar to overapplication of nitrogen (N) fertilizer delaying flowering in angiosperm (Yuan et al., 2016), ammonium tartrate as N sources inhibits moss development and can maintained moss at the protonema stage.
This typical drought-tolerant moss resurrection plant can achieve dehydration of more than 95% and rapidly recover within a few seconds when rehydrated. The filaments of P. patens only can survive up to 92% water loss but cannot recover from complete desiccation (Frank et al., 2005; Khandelwal et al., 2010). It was reported that P.patens was desiccated inside a laminar flow hood for 24 h, but the rate of water loss is instability because of the different levels of humidity in different places (Khandelwal et al., 2010). Still, this method cannot guarantee the experimental material’s dehydration level each time. We first attempted to use silica gel to dehydrate the material and observed the level of dehydration and stability. In the experiment, protonema material was selected. First, an excess amount of 5 cm in height of silica gel was added to a crystallization dish. Then the 7-day-old homogeneous slurry subculture material was transferred onto the silica gel in the crystallization dish and air-dried on a clean bench. The water loss rate of the material was measured at regular intervals (the weight of the filter paper was obtained after weighing the material, the filter paper was scraped clean after drying the material, and the weight of the filter paper was obtained after drying it under the same conditions). The results showed that the dehydration rate of the moss reached 95.42% (Supplementary Table S2). In order to test whether the moss material could revive after rehydration after the condition of extremely dehydration, we placed the dehydrated B.argenteum material into BCD medium after rehydration and grew it for one week, and found that the plants could grow normally. This indicates B.argenteum has a strong drought resistance and the ability to rapidly recover to the normal state after rehydration.
With the development of molecular genetics, more and more molecular markers are used for species identification and classification. Some molecular markers can be used for the identification and classification of mosses. We used three molecular markers (trnL-F, trnG, atpB-rbcL) on the chloroplast genome and a molecular marker (nad5) on the mitochondrial genome, and two fragments of the 26S gene in nuclear to identify B.argenteum. The lengths of the sequenced molecular markers were 520 bp, 638 bp, 684 bp, 1976 bp and 1132 bp, 1095 bp, respectively. Sequencing results for other species can be obtained from GenBank. The concatenated sequences of these five molecular markers were used to construct a phylogenetic tree using the Maxmum likelyhood method using IQ-TREE program, and B.argenteum was found to be most closely related to reported B.argenteum Hedw (Pedersen et al., 2007), indicating that B. argenteum also belongs to the Bryaceae branch (Fig. 2).
Material Selection and Genome Size Determination
Although the transcriptome data of B. argenteum was reported (Gao et al., 2015; Gao et al., 2017), but the genome size of this plant is still uncertain. the genome size of B.argenteum CNU51 is 313 Mb using k-mer (Supplementary Figure S2). The genome size of CNU51 is smaller than 700 Mb reported (Wang, 2016), which suggests different B.argenteum populations might have the different genome ploidy type B.argenteum CNU51 has a relatively small genome, making it a good model material for subsequent experiments. While, the moss model P. patens with about 500 Mb genome size has a disadvantage for that it has experienced multiple events of polyploidization during evolution that has resulted in a number of families of duplicated genes and functional redundancy (Rensing et al., 2008).
Protoplast Isolation of Bryum argenteum from Protonema and Protoplast Regeneration
The 15-day-old protonema tissues in liquid were used to isolate protoplasts. Protonema were pre-incubated for 1 h in 0.5 M mannitol with slow rotation. As the enzymatic digestion time increased, the cell walls were gradually broken down, and the protoplasts were gradually separated, increasing the number of protoplasts. At 100 minutes, there were almost no protoplasts in the field of view (Fig. 3A a-l). Counting the number of protoplasts isolated at different treatment times showed that the longer the breakdown time, the more protoplasts were isolated (Fig. 3B). The protoplasts produced after breakdown showed intense activity after FDA staining (Fig. 3C).
We chose BCDG solid medium containing 8% mannitol for protoplast regeneration and used a single cell to illustrate the different stages of protoplast regeneration. The process of protoplast regeneration in Bryum argenteum is similar to that of other mosses observed under a microscope. First, the protoplast regenerates a cell wall (Fig. 4A a-b), and injured cells will die within 1–2 days. From days 2 to 5, cells with newly formed cell walls divide into 2–3 cells (Fig. 4A c-f). By day 10, they further develop into new protonema (Fig. 4A g-i). Young gametophytes are visible to the naked eye at 20 days (Fig. 4A j).
After protoplast differentiation and cell wall formation, the protoplasts begin to differentiate. about 50% of the cells divide into two, with one cell ceasing division and only the other cell dividing to produce the next daughter cell. After several days, it grows into a single filamentous cell (Fig. 4B a). 23% of the cells produce two daughter cells with equal division ability, each of which subsequently divides independently. Ultimately, one protoplast cell grows into two filaments (Fig. 4B b). A small number of cells divide twice, producing three or four daughter cells, but only one or two of these daughter cells continue to divide into single filaments, while the other cells cease division (Fig. 4B c-d). From the statistical results, it can be seen that the splitting mode of splitting into two is the most common (Fig. 4C). The phenomenon of initial division during protoplast regeneration is very similar to spore germination.
Transient Transfection of Protoplasts via PEG-Mediated and Screening of B.argenteum Stable Transformation Strains
To establish a transformation system for B.argenteum, we firstly explored suitable transformation conditions using a transient expression system. We directly transformed the protoplasts of B.argenteum with the pCambia1302:GFP and evaluated the transformation and transient expression efficiency. The PEG-mediated transformation method was based on the method of P.patens. After transformation, the regenerated cells were cultured on a liquid BCDG medium containing 8% mannitol for 16h.To assess the transient transformation efficiency, we observed GFP fluorescence. GFP fluorescence was dispersed throughout the cytoplasm after 16 hours of transient transformation with pCambia1302:GFP (Fig. 5A) At the same time,We used the method mentioned above to transform the pCambia1302:GFP into the protoplasts of the B.argenteum. After transformation, the regenerated cells were cultured on a solid BCDG medium containing 8% mannitol for ten days. Then, the screened cells were transferred to a hygromycin-containing medium at a 25 ug/mL concentration for the first seven-day selection round. After that, the recovered cells were transferred to a drug-free medium for growth recovery. Ten days later, they were transferred to a BCDG medium containing 50 ug/mL hygromycin for the second selection round. After five days of the second selection, almost all B.argenteum cells turned white and died. We obtained seven resistant plants. PCR analysis confirmed that seven plants contained the GFP fragment. Meanwhile, The expression of GFP was observed in the stable transformed positive plant 35S:GFP#2 (Fig. 5B).To examine chromosomal integration of transgenic constructs in transgenic lines, we assessed GFP transgene expression by semi-quantitative RT-PCR、real-time quantitative PCR (Fig. 5C&S4). Next, the ectopic expression of Citrine protein was examined by immuno-blot assay. Anti-green fluorescent protein (GFP) antibody was used to detect Citrine because of its high similarity to GFP (Fig. 5D). These results indicate that the transfection method described here can be used to generate CNU51 transfectants.
Establishment of CRISPR/Cas9 Knockout System in B.argenteum
Based on the transcriptome data of B.argenteum (Gao et al., 2015; Gao et al., 2017) and our unpublished genome data of B.argenteum, we retrieved two homologous sequences of ABI3 and named them BaABI3A and BaABI3B. We constructed a phylogenetic tree using the maximum likelihood method (ML) with the cloned amino acid sequences of ABI3 in moss. The phylogenetic analysis showed that mosses formed one branch, and seed plants, including Arabidopsis thaliana, Populus trichocarpa, and rice, formed another branch, with the three homologous genes of the AFL subfamily in rice and FUS3 forming a separate branch. BaABI3A and PpABI3C clustered in the moss branch, while BaABI3B clustered with PpABI3A and PpABI3B (Supplementary Figure S3). Comparative analysis of the phylogenetic tree and sequence proved that there are two homologous genes of ABI3 in the B.argenteum, which belong to the ABI3 family. To investigate the functions of the two BaABI3 genes in B.argenteum, we explored their expression patterns under various stress treatments. We selected CNU51 progeny cultures growing for nine days as the source material. We subjected them to four non-biological stress conditions, namely 10 µM ABA, 300 mM mannitol, drought/rehydration, and 250 mM NaCl. Under ABA stress treatment, the BaABI3A gene responded first, and its expression reached its maximum level in 2 hours, while the BaABI3B gene reached its maximum expression level in 8 hours. Under drought/rehydration treatment, both BaABI3 genes were highly expressed under drought/rehydration treatment, with BaABI3A mainly expressed during the drought stress period and BaABI3B primarily expressed during the rehydration period after the drought. The response and expression patterns of the two BaABI3 genes to salt stress were different, with BaABI3A being expressed in the later stages and BaABI3B being expressed in the earlier stages. Both BaABI3 genes responded to osmotic stress caused by mannitol more slowly (Fig. 6A).
Currently, no reports of the CRISPR/Cas9 gene editing system in B.argenteum. Here, we used the CRISPR/Cas9 gene editing method from P.patens (Collonnier et al., 2016) to explore its application in B.argenteum. Using CNU51 moss as the material, we targeted two sites for BaABI3A, sgRNA 1 and sgRNA 2. Four mutant plants, Baabi3a-5, Baabi3a-14, Baabi3a-20, and Baabi3-a21 were obtained through screening with the G418 antibiotic and sequencing identification. For BaABI3B, six target sites were designed, sgRNA 3, sgRNA 4, sgRNA 5, sgRNA 6, sgRNA 7, and sgRNA 8, and two mutant plants, Baabi3b-16 and Baabi3b-36 were obtained through screening with the G418 antibiotic and sequencing identification (Fig. 6B). We attempted different combinations of sgRNAs and found that only the plasmids containing sgRNA 1 and sgRNA 4 underwent gene editing. The statistical results showed that the editing efficiency of the BaABI3A gene in B.argenteum was 4.5%, and the editing efficiency of the BaABI3B gene was 1.9%. This result provides us with confidence to further explore the functions of other genes in B. argenteum (Table 1).
Table 1
Targeting efficiency of genome editing on BaABI3A and BaABI3B gene using the CRISPR-Cas9 system in Bryum argenteum
Gene
|
sgRNA used for
transformation
|
G418 clones★
|
Number of
analysed clones†
|
GT efficiencies (%)‡
|
BaABI3A
|
sgRNA1
|
89
|
4
|
4.5
|
BaABI3B
|
sgRNA4
|
103
|
2
|
1.9
|
sgRNA1 and sgRNA4 target BaABI3A and BaABI3B genes respectively (see Supplementary Table S3).
G418 clones★ are the stable antibiotic-resistant clones that survived after subculture on G418 medium.
Number of analysed clones† where the donor DNA template have been edited with deletions or insertions.
GT efficiencies (%) ‡express the frequency of edited clones among the population of antibiotic-resistant transgenic clones.
There are no relevant literature reports on drought, salt stress, and osmotic stress treatment conditions in B.argenteum. Therefore, BaABI3 gene mutants were used to explore the tolerance of B. argenteum to various abiotic stresses. A gradient of 100 mM, 200 mM, and 400 mM NaCl was set to analyze the optimal concentration for NaCl stress treatment. We found that wild-type and mutant strains could grow well under low-concentration salt treatment for three days and then transferred to a standard BCDG medium, which indicates that B. argenteum material is tolerant to short-term low-concentration salt stress. After two days of growth on BCDG medium containing 400 mM NaCl, Baabi3a 、Baabi3b mutant and wild-type B. argenteum materials appeared whitish, indicating significant salt stress. When transferred to a normal culture medium, the wild-type could recover the growth. In contrast, Baabi3a-14 mutant strain could also recover growth (subsequent repeated experiments showed that Baabi3a-14 could only partially recover growth, and its status was not as good as that of the WT plants, while Baabi3a-5, Baabi3a-20, and Baabi3a-21 mutant strains could only partially or not recover growth, showing a salt-sensitive phenotype) (Supplementary Figure S5).
To determine whether wild-type and mutants have salt-sensitive phenotype of B.argenteum, we treated the original prostrate material of the wild-type and mutants with 500 mM NaCl. The experimental results showed that the material treated directly with 500 mM NaCl turned white after 2 days of growth on the medium. The mutant strains Baabi3a-5, Baabi3a-14, and Baabi3b-16 did not recover or only partially recovered during the recovery process. The results showed that the ABI3 genes have the function in tolerating salt stress (Fig. 6C).
B. argenteum is a typical desiccation-tolerant (DT) plant. In this study, drying treatment was also performed on five mutants (Baabi3a-5, Baabi3a-14, Baabi3a-20, Baabi3a-21, Baabi3b-16) and the wild-type. Rehydration after 24 hours of drought, it was observed that the wild-type materials were able to survive, while only a little portion of the mutants resumed growth compared to the control. These results indicate that the Baabi3a and Baabi3b mutants are sensitive to drought stress, and both gene BaABI3A and BaABI3B function in desiccation for B. argenteum protonema (Fig. 6D)