Bryophytes Harbor Cultivable Actinobacteria With Plant Growth Promoting Potential

This study was designed to investigate the cultivable actinobacteria associated with bryophytes and their plant growth promoting ability. Thirteen actinobacteria were isolated and tested for their ability to promote growth of plant in vitro and in planta. All isolates were able to produce IAA and siderophores. Six isolates were identified as members of the genus Micromonospora. Five isolates belonged to the genus Streptomyces and one each of Microbispora and Mycobacterium. Micromonospora sp. CMU55-4 was inoculated to rare moss [Physcomitrium sphaericum (C. Ludw.) Fürnr.] and could increase the amount of carotenoid, fresh weight, and dry weight of this moss. In addition, this strain promoted capsule production, and rescued P. sphaericum’s gametophytes during acclimatization to land. Strain CMU55-4 was identified as Micromonospora chalcea based on whole genome sequence analysis. Its plant growth promoting potential was further characterized through genome mining. The draft genome size was 6.6 Mb (73% GC). The genome contained 5,933 coding sequences. Functional annotation predicted encoded genes essential for siderophore production, phosphate solubilization that enable bacteria to survive under nutrient limited environment. Glycine-betaine accumulation and trehalose biosynthesis also aid plants under drought stress. M. chalcea CMU55-4 also exhibited genes for various carbohydrate metabolic pathways indicating those for efficient utilization of carbohydrates inside plant cells. Additionally, predictive genes for heat shock proteins, cold shock proteins, and oxidative stress such as glutathione biosynthesis were identified. In conclusion, our results demonstrate that bryophytes harbor plant growth promoting actinobacteria. A representative isolate, M. chalcea CMU55-4 promotes the growth of P. sphaericum moss and contains protein coding sequences related to plant growth promoting activities in its genome.


INTRODUCTION
Plant microbiomes are involved in the well-being of their host health by inhibiting the growth of plant pathogens and help plant to tolerate stress conditions (Gabriele, 2016). Plant growthpromoting rhizobacteria (PGPR), which live around root area of plants, can be beneficial to plants by mediating plant growth through diverse biochemical mechanisms such as biosynthesis of indole-3-acetic acid (IAA), 1-aminocyclopropane-1-carboxylic acid (ACC) deaminase, siderophore production, and phosphate solubilization. PGPR which are abundant in soil and rhizosphere can penetrate into root endosphere through cellular disjunction during lateral root emergence (Bulgarelli et al., 2013) to colonize inside plants. Actinobacteria, a phylum of Grampositive bacteria that contained high amount of G+C in their DNA (>55 mol% of genomic DNA) are interesting due to their prolific metabolic capabilities (Yadav et al., 2018). Plant growth promoting actinobacteria (PGPA) are abundant in soil and can enhance nutrients availability, administer plant metabolism and reduce environmental stress in plants (Hamedi and Mohammadipanah, 2014). Bryophytes are group of plants that grow under specific climate conditions and their microbiomes are still not well studied. They are the oldest known land plants that lack true leaves, stem or true vascular system and considered as a remarkable reservoir of novel active compounds, natural products and antibiotics (Chandra, 2014). Actinobacteria associated with bryophytes could have some specific evolutionary adaptations to their host plants and may have potential of bioactive compounds production that need to be investigated. One of the easiest ways for bryophytes propagation is using tissue culture technique. However, the obtained in vitro plantlets are usually unable to compete with other native soil microbes when transfer to soil. In addition, micropropagated plants are struggled to cope with environmental conditions in the field which results in the morphological and anatomical changes of plantlets (Chandra and Bandopadhyay, 2010). The objectives of this study are 1. To investigate the cultivable actinobacteria associated with some bryophytes; 2. To study the plant growth promoting ability of actinobacteria from bryophytes.
One strain, Micromonospora chalcea CMU55-4 shows promising potential to promote the growth of tested bryophyte during acclimatization period in laboratory before transfer to the forest. Its taxonomic characterization and whole genome analysis are also reported.

Selective Isolation
Each bryophyte sample (1 g fresh weight) was surface sterilized (70% ethanol 1 min, 3% sodium hypochlorite 10 min, 3 times sterile distilled water). Sterile samples were crushed in sterile mortars, diluted with sterile distilled water up to 10 −3 and plated on starch casein agar (Küster and Williams, 1964), humic acid vitamin B agar (Hayakawa and Nonomura, 1987), R2A agar and water proline agar (1% proline). All media were supplemented with 25 µg/ml nystatin and 10 µg/ml cycloheximide. Plates were incubated for 30 days at 25 • C. All colonies appeared on agar were picked and purified on ISP2 agar (Shirling and Gottlieb, 1966). The effectiveness of the surface sterilization method was evaluated by spread the final rinse water of each bryophyte sample on ISP2 agar.

Plant Growth Promoting Activity
IAA production was determined following the methods of Glickman and Dessaux (1995). In brief, actinobacteria were grown in ISP2 broth with and without 2 mg/ml L-tryptophan for 7 days in the dark. After that, the supernatant was mixed with Salkowski's reagent (1 ml of 0.5 M FeCl 3 in 50 ml 35% HClO 4 ), stored at room temperature for 30 min and measured the optical density at 530 nm. IAA production is also confirmed by HPLC (Fluorescent detector) with a Restek Ultra C18, 5 µm (150 × 4.6 mm) using 0.1 M acetic acid as mobile phase A and 0.1 M acetic acid in methanol as mobile phase B (1 ml/min flow rate). Standard methods were used to determine siderophore production qualitatively and quantitatively. Siderophore production was first screened on CAS agar (Schwyn and Neilands, 1987) using King's B agar as a basal medium (King et al., 1954). The appearance of a yellow to orange zone around the actinobacterial colony after 7 days incubation in the dark was an indication of siderophore production. Quantitative determination of siderophore was carried out in liquid medium. Actinobacteria were grown in King's B broth for 7 days. The culture supernatant was mixed with ferric perchlorate solution for hydroxamate siderophore (Atkin et al., 1970), and 0.5 M HCl, nitrite-molybdate reagent and 1 M NaOH for catecholate siderophore (Arnow, 1937). Phosphate solubilization was determined from clear zone formation on Pikovskaya (PVK) agar supplemented with 0.5% (w/v) tricalcium phosphate after 7 days incubation (Nautiyal, 1999). Drought tolerant ability of actinobacteria was also investigated using sorbitol as an osmotic adjustment in water agar medium (Hallsworth et al., 1998).

Taxonomic Characterization
DNA extraction was performed following the method of Kudo et al. (1998). Genomic DNA was used as the template for polymerase chain reaction (PCR) following the method of Lee et al. (2014). 27F (5 AGAGTTTGATCMTGGCTCAG 3 )   (Kumar et al., 2016). Tree topology was evaluated using the bootstrap resampling method at 1000 bootstrap. Strain CMU55-4 was also identified based on whole genome sequence analysis. The ANI value was calculated and compared in JSpeciesWS (Ritcher et al., 2016), web server tool, using ANI-Blast (ANIb) and ANI-MUMmer (ANIm) algorithms Growth Promotion on Physcomitrium sphaericum (C. Ludw.) Fürnr.
Physcomitrium sphaericum (C. Ludw.) Fürnr. was used as a moss model to determine beneficial effects of selected actinobacteria on bryophytes during transplantation. P. sphaericum is a rare moss species in Thailand with the risk from extinction. The re-introduction of this moss back to nature is badly needed. Spores of P. sphaericum were picked from healthy plants from the nature under stereo microscope using sterile forceps and placed into Hoagland agar (Hoagland and Arnon, 1950) and incubated for 3 months at 24 • C, 35,000-40,000 lux intensity to let spore germinate and form protonema. Then, the moss protonema was transferred to Hoagland agar and incubated in the same condition for 3 months. The moss plantlets were transferred from agar media to soil. Strain CMU55-4 was chosen to be introduced to moss due to its high IAA production in vitro and fast growth within 3 days which is an important trait of PGPA to compete with other soil microbes. Strain CMU55-4 was grown in ISP2 broth for 7 days, 25 • C, 150 rpm. Culture broth was centrifuged to collect cells. Cells were washed and resuspended in sterile distilled water and adjusted the concentration to OD 600 = 1, which is equivalent to 10 6 cells/ml. A suspension of strain CMU55-4 (1 ml) was dropped into the autoclaved soil around the plants.
Sterile distilled water was used as a control solution. The plants were maintained in the incubator at 24 • C, 35,000-40,000 lux for 1 month. Chlorophyll and carotenoid contents were determined from fresh samples using standard method (Chappelle et al., 1992). Dry weight was obtained from plants that were dried in 60 • C oven for 7 days.

Genome Mining for Plant Growth Promoting Potential
Whole genome sequencing of Micromonospora sp. CMU55-4 was carried out by an Illumina Miseq platform (Illumina, Inc., San Diego, US-CA) using 2 × 250 bp paired-end reads. Raw reads quality was checked using FASTQC software (Andrews, 2010). Adaptors and poor-quality reads were removed using Trim Galore (Krueger, 2015), and the filtered reads were used as an input for Unicycler (Wick et al., 2017), genome assembly program. Annotation of assembled genome was done using Prokka Version 1.13 (Seemann, 2014), and NCBI Prokaryotic Genome Annotation Pipeline (PGAP). The draft genome sequence of strain CMU55-4 was mined using RAST annotation server (Aziz et al., 2008) and analyzed through SEED viewer (Overbeek et al., 2014) for genes responsible for plant growth promoting properties. FIGURE 3 | Effects of Micromonospora strain CMU55-4 toward the growth of P. sphaericum. a, b indicated significant difference in statistical analysis tested by SPSS independent T-test (p < 0.05), n = 3.

GenBank Accession Number
The accession number of the draft genome sequence is JAAOLH000000000.

Statistical Analysis
Statistical analysis was performed by SPSS statistics 17.0 program.

Plant Growth Promoting Activity
All isolates produced indole-3 acetic acid and siderophores at varying amount ( Table 1). Isolates CMU51-1 and CMU55-4 produced the highest amount of IAA without L-tryptophan supplement (2.73 µg/ml) and in 2 mg/ml of L-tryptophan (11.35 µg/ml), respectively. For siderophores production, isolates CMU51-1 produced the highest amount of both hydroxamate type (992.50 ± 50.76 µMole/l) and catecholate type siderophores (484.47 ± 27.91 µMole/l) ( Table 1). For drought tolerant ability, no growth was observed below a w 0.957 for all isolates. None of the obtained isolates could solubilize tricalcium phosphate on PVK agar.

Taxonomic Characterization
Comparison of 16S rRNA gene sequence similarity of the obtained actinobacteria in EzBiocloud database, assigned them as members of the following genera: Micromonospora, Streptomyces, Mycolicibacterium and Microbispora ( Table 2). The majority of isolates were Micromonospora (46%) followed by Streptomyces (38%). The 16S rRNA gene similarity values were ranged from 96.31 to 100%. Phylogenetic analysis confirmed the assignment of these actinobacteria at genus level based on BLAST results (Figure 1)  Growth Promotion on Physcomitrium sphaericum (C. Ludw.) Fürnr.
Physcomitrium sphaericum (C. Ludw.) Fürnr. plantlets that were inoculated with Micromonospora strain CMU55-4 showed better growth appearance compared to control plants. The control plants have more chlorosis leaves and showed growth retardation (Figure 2A). The plants that had been treated with strain CMU55-4 started to develop new leaves and showed the production of capsules ( Figure 2B). Carotenoid, fresh weight and dry weight of P. sphaericum were significantly higher than the control. However, total chlorophyll content was not different in control and strain CMU55-4 inoculated plantlets (Figure 3).

Genome Mining for Plant Growth Promoting Potential
From whole genome sequence data, the estimated genome size of Micromonospora sp. CMU55-4 was 6.6 Mb. The GC content was 73.0%. The summary of sequence assembly and genome annotation is displayed in Table 3. Phylogenomic tree clearly classified isolate CMU55-4 in the same species and subspecies level with M. chalcea DSM 43026 T (Figure 4). Prokka predicted 5,946 coding regions with 6,077 estimated genes ( Table 4) contained 67 tRNAs, 3 rRNAs and 2 repeat regions. Gene annotation in RAST server showed that genes were grouped into 25 subsystems (Supplementary Material 1). The majority of genes play roles in amino acids, carbohydrate, and protein metabolisms. From various subsystems, essential roles of protein coding sequences according to their plant growth promoting traits was grouped in Table 5. Twenty-four genes were associated with siderophore production, namely siderophore assembly kit, siderophore desferrioxamine E, and siderophore aerobactin. Indole-3-glycerol phosphate synthase and tryptophan synthase plays role in IAA synthesis. Gene encoding exopolyphosphatase with possible role in phosphate solubilization was also found. In addition, protein coding sequences involved in nitrogen metabolism such as assimilatory nitrate reductase, nitrate/nitrite transporter, and ammonium transporter were also detected. Strain CMU55-4 was well equipped with protein coding sequences related to oxidative stress response such as SoxR, NsrR, FIGURE 4 | Phylogenomic tree of strain CMU55-4 and their closely related type strains of the genus Micromonospora. Tree inferred with FastME 2.1.6.1 (Lefort et al., 2015) from Genome BLAST Distance Phylogeny (GBDP) distances calculated from genome sequences. The branch lengths are scaled in terms of GBDP distance formula d 5 . The numbers above branches are GBDP pseudo-bootstrap support values from 100 replications, with an average branch support of 92.6 %. The tree was rooted at the midpoint (Farris, 1972). Leaf labels were annotated by affiliation to species and subspecies clusters, genomic G+C content, δ values, overall genome sequence length, number of proteins, and the kind of strain, respectively.
organic hydroperoxide reductase, glutathione peroxidases and trehalose synthesis genes. Glycine betaine transporter (OpuD) for osmotic adjustment under osmotic stress was also present. CspA protein family, DnaK and DnaJ chaperones responsible for cold or heat stress were identified. Strain CMU55-4 also exhibited genes involved in the utilization of saccharides found inside plant cells such as xylose, arabinose, mannose and D-galacturonate (  (Figure 5). The strain grew at 20 to 45 • C (optimally at 30 • C), pH 5.0 to 12.0 (optimally pH at 8.0) and tolerated up to 4% (w/v) NaCl. Strain CMU55-4 utilized various type of carbohydrates such as L-arabinose, dulcitol, D-mannose, and D-mannitol and those found in plant cells including D-glucose, sucrose, cellulose, amygdalin, and starch ( Table 6). Cell-wall peptidoglycan of strain CMU55-4 contained meso-diaminopimelic acid. Galactose, arabinose and xylose were detected as diagnostic sugars in the whole-cell hydrolysates. The predominant phospholipids were diphosphatidylglycerol (DPG), phosphatidylethanolamine (PE), and phosphatidylinositol (PI). Major menaquinones were
Members of this genus have been isolated from various habitats including plants (Trujillo et al., 2010Carro et al., 2012Carro et al., , 2013Genilloud, 2015). Several new species of Micromonospora are endophytes of various plants as exemplified by the description of M. costi (Thawai, 2015), M. globbae (Kuncharoen et al., 2018), M. phytophila (Carro et al., 2018b), M. luteiviridis (Carro et al., 2018b), M. oryzae (Kittiwongwattana et al., 2015), and M. terminaliae (Kaewkla et al., 2017). Streptomyces isolates also dominant represent around 38.5%. Our results provide additional evidence that these two actinobacterial genera are also dominant in bryophytes. Mycolicibacterium stellerae was recently    (Nouioui et al., 2019b). However, this is the first time that member of this genus was found associated with a bryophyte. The remaining isolate, CMU55-5 was closely related to Microbispora rosea subsp. rosea. The low 16S rRNA gene similarity value (96.3%) between CMU55-5 and its closest neighbor, suggests that this strain may represent a new species. However, polyphasic taxonomic characterization is needed for its formal description which will be the subject of future investigation.
No actinobacteria was obtained from liverwort. This may be an effect of "oil bodies, " the highly distinctive organelle uniquely found in liverworts (He et al., 2013). This organelle is responsible for toxic compound accumulation that usually use for protection of liverworts from herbivore (Stahl, 1888), pathogens, low temperature and excessive light (Hieronymus, 1892). We opined that during the isolation of actinobacteria, the liverwort sample was crushed which may cause the release of toxic compounds from oil bodies and inhibit the growth of actinobacteria.
The use of water proline agar yielded the highest number of actinobacterial isolates in this study. Water proline agar is a low nutritional medium that simulate oligotrophic status in natural environments. Proline is also serving as a compatible solute that commonly found in plant cells. The composition of the minimal media is suggested to allow an easier adaptation for endophytes (Alain and Querllou, 2009) and allow slow growing endophytic strains a chance to develop (Eevers et al., 2015). Complex media composed of rich carbon and nitrogen sources are suggested to be unsuitable for the growth of slow-growing endophytic bacteria as they do not resemble the environment inside plant tissues (Eevers et al., 2015). Low nutritional media have been successfully used to isolate actinobacteria from various environmental samples (Janssen et al., 2002;Wang D.-S. et al., 2014;Ruttanasutja and Pathom-aree, 2015).
Strain CMU55-4 was classified into the genus Micromonospora of the family Micromonosporaceae based on the sequence of 16S rRNA gene and its unique phenotypic characteristics. It forms a characteristic orange colony at the early stage, which turns to dark olive brown color with age. The strain produced extensively branched substrate hyphae, lack of aerial mycelia which is a unique morphological characteristic of the genus Micromonospora (Ørskov, 1923). The strain formed a single non-motile warty-globular spore on the vegetative mycelium similar to the closest type strains, M. chalcea DSM 43026 T (Ørskov, 1923). The formation of single spores is also the main morphological characteristic of the genus Micromonospora . A meso-diaminopimelic acid was found in the cell wall peptidoglycan of strain CMU55-4 with arabinose, xylose and galactose as diagnostic sugars in the whole-cell hydrolysates corresponding to the cell wall type II and sugar type D (Lechevalier et al., 1971). Strain CMU55-4 exhibited phospholipid type II, comprising of phosphatidylethanolamine (PE), diphosphatidylglycerol (DPG), and phosphatidylinositol (PI) as the major phospholipids (Lechevalier et al., 1977). The strain contained a large amount of MK-9(H 4 ) and MK-9(H 6 ) which generally found in the genus Micromonospora (Nouioui et al., 2018). Strain CMU55-4 shared 98.47% of ANIb values and 91.8% digital DNA-DNA hybridization to M. chalcea DSM 43026 T . The value of <95-96% ANI or 70% dDDH is proposed as a cutoff value for delineating of new species (Chun et al., 2018). Therefore, strain CMU55-4 was identified as M. chalcea CMU55-4.
Micromonospora have long been recognized as antibiotic producers (Hirsch and Valdes, 2010). Recently, endophytic Micromonospora have been regarded as plant growth promoting bacteria especially in legumes (Trujillo et al., 2014;Benito et al., 2017). Physcomitrium sphaericum moss is known to be distributed only in the temperate regions; North America, Europe, Russia, China, and Japan. Recently, it has been found in the tropical country and reported as new record to Thailand (Printarakul et al., 2014). The inoculation of M. chalcea CMU55-4 on P. sphaericum moss resulted in an increase of carotenoid content, fresh and dry weight significantly higher than the control (Figure 3). This promotion effect appears to be related to the production of phytohormone IAA as M. chalcea CMU55-4 produced the highest IAA among all tested actinobacteria. Interestingly, P. sphaericum inoculated with M. chalcea CMU55-4 developed new leaves and showed the production of capsules ( Figure 2B). Capsule is an important part of bryophyte reproduction as it represent sporophyte stage and benefits in asexual reproduction through spores dispersion (Sundberg et al., 2014). It is likely that the moss-bacterium interaction between M. chalcea CMU55-4 and P. sphaericum may play a role in promote leaves and capsule development though the exact mechanism still required further investigation. Effect of moss associated bacteria on Pylaisiella selwynii moss development was reported that the obtained Gram-negative bacteria could promote protonemal growth and gametophore initiation of P. selwynii (Spiess et al., 1981). Recently, auxin has been shown to involve in the development of moss (Physcomitrella patens) including gametophore and sporophyte development (Thelander et al., 2018). Almost all works regarding plant growth promoting actinobacteria are carried out with vascular plants. For examples, the inoculation of Micromonospora strain SB3 promotes plant biomass, root length and increases the acclimatization rate of Lolium multiflorum plantlets after 4 weeks of in vitro culture (Della Mónica et al., 2018). Recently, an endophytic M. chalcea UAE1 from halophytic crop, Salicornia bigelovii has been reported to promote the growth of S. bigelovii mainly by ACC-deaminase production (El-Tarabily et al., 2019). To the best of our knowledge, this is the first report on potential of Micromonospora species on the growth and development of non-vascular plants such as mosses.
Approximate genome sizes of 40 Micromonospora type strains were reported to be ranged from 6.1 Mbp for Micromonospora marina DSM 45555 T to 7.9 Mbp for M. carbonacea DSM 43168 T with the average genome size for all of the Micromonospora strains was 7 ± 0.4 Mbp (Carro et al., 2018a). The estimated genome size of our M. chalcea CMU55-4 was within these previously reported ranges (6.6 Mb) with 73.0% GC content. The genome size of M. chalcea DSM 43026 T was 7.0 Mb and 72.8% G+C content (Carro et al., 2018a). The slightly bigger genome size of 7.086 Mb was reported from plant growth promoting rhizosphere Micromonospora sp. strain MW-13 with a similar G+C content of 73.3% (Jahanshah et al., 2019).
Omics techniques, including genome sequencing, comparative genomics, microarray, next generation sequencing, metagenomics, and metatranscriptomics can be used to explain plant-endophyte relationship (Kaul et al., 2016). Endophytic lifestyle of M. chalcea CMU55-4 was supported by data from genome mining and phenotypic studies as this strain was able to utilize various kind of saccharides and carbohydrates which found inside plant cells. Saccharides molecules were discharged from plant cell walls during elongation process and served as nutrients to promote the colonization ability of plantassociated bacteria (Etesami et al., 2015). M. chalcea CMU55-4 is also able to produce both catecholate and hydroxamate type siderophores in vitro. In addition, several predictive genes related to siderophore production were also detected in its genome for examples desferrioxamine E and aerobactin and genes encoding transporter proteins. Neilands and Leong (1986) suggested that siderophore systems may play a role in infections of plant tissue. Thus, the ability of M. chalcea CMU55-4 to produce siderophores may help this strain to enter and live inside plant cells. Metallophores producing Micromonospora strains were recently reported to promote the growth of Arabidopsis thaliana under heavy metal condition by playing roles in metal acquisition, iron metabolism and resistance to toxic compounds (Ortúzar et al., 2020). Genes mutually found in Ortúzar et al. (2020) and this study are ferric hydroxamate ABC transporter (FhuB, FhuC, FhuD), siderophore desferrioxamine E (DesA, DesB, DesC, DesD), and siderophore aerobactin (FhuB, FhuC, FhuD).
Previous studies on actinobacterial genome revealed genes related to plant growth promoting activities primarily production of IAA, siderophore, phosphate solubilization, and phytopathogen inhibition (Gupta et al., 2014;Liu et al., 2019;Nouioui et al., 2019a). Predictive genes obtained from genome mining in this study supported the experimental data on IAA and siderophore producing abilities of M. chalcea CMU55-4 in vitro. Tryp-dependent pathway could be a pathway for IAA synthesis in strain CMU55-4 as predictive genes for tryptophan synthase alpha chain (TSA1), tryptophan synthase beta chain (TSB), indole-3-glycerol phosphate synthase (IGS) were found. Indole-3-glycerol phosphate or indole is a branch point for Tryp-dependent pathway that directly leads to IAA synthesis (Spaepen and Vanderleyden, 2011;Di et al., 2016). IAA promotes plant growth by advocating cell division, cell elongation, stimulates primary growth, and plays role in stress resistant. IAA production trait is considered as a useful criterion for selection of endophytic and rhizospheric bacteria in rice growth promotion (Etesami et al., 2015). High IAA production without L-tryptophan was recorded in some isolates such as CMU51-1 and CMU51-5. This observation suggests the possibility that these isolates can produce IAA by Tryp-independent pathway. M. chalcea CMU55-4 failed to solubilize tricalcium phosphate on PVK agar though predictive gene for exopolyphosphatase was detected. Since only tricalcium phosphate was used for phosphate solubilizing property, we cannot rule out the possibility of M. chalcea CMU55-4 to solubilize other types of inorganic phosphates. M. chalcea CMU55-4 also genetically possessed abilities related to nitrogen metabolism in particular nitrate/nitrite ammonification and ammonia assimilation. These  abilities could enhance nitrogen uptake system of plant as nitrate (NO 3 − ) and ammonium (NH 4 + ) are the main resources of inorganic nitrogen (N 2 ) absorbed by the roots of higher plants (Wickert et al., 2007).
It is also interesting to note the gene sequences responsible for other functions, such as the production of antioxidants and heat shock/ cold shock proteins were detected though these properties had not been investigated under laboratory conditions in this study. M. chalcea CMU55-4 contained predictive genes involved with oxidative stress response such as SoxR, NsrR, organic hydroperoxide resistance protein, trehalose synthesis genes, glutathione S-transferase omega, and carotenoid production, which are the molecules help plant to tolerate to oxidative stress. Reactive oxygen species (ROS) created by the plant cells, are generally neutralized by the production of enzymes such as superoxide dismutases (SOD), catalases (CatA), peroxidases (POD), alkyl hydroperoxide reductases (AhpC), genes conferring nitrosative stress caused by reactive nitrogen species, and glutathione-S-transferases (GSTs) in endophytes (Khare et al., 2018). GST was found in the genome of Streptomyces scabrisporus NF3, an endophyte of Amphipterygium adstringens (Ceapã et al., 2018). The ability to alleviate oxidative stress is one attractive property of plant growth promoting bacteria. Forty-two genomes of Micromonospora strains including the type strain of M. chalcea, were analyzed to determine the presence of PGP genes and other characteristics related with plant growth promotion (Carro et al., 2018a). Genomes of Micromonospora strains contain genes related to stress responses such as cspA, cspC for cold shock response, dnaK and grpE for heat shock responses, and sod for oxidative stress. Similar observation was also found in this study.
Bryophytes require high water for growth. They are more sensitive to water deficit condition than other plants. Accumulation of compatible solute such as glycine betaine therefore can help bryophytes to maintain water balance. Glycine betaine transporter OpuD was found in M. chalcea CMU55-4 genome. The opuD gene product was reported to be essential for glycine betaine uptake and osmoprotection in E. coli (Kappes et al., 1996). Glycine betaine accumulation had been reported to increase the drought tolerant in Streptomyces chatreusis . Predictive genes involve in trehalose metabolism was also present in M. chalcea CMU55-4 genome including gene encoded for trehalose synthase. Trehalose is another compatible solute that can protect plant proteins and cellular membranes from inactivation or denaturation caused by a variety of stress conditions, including desiccation, dehydration, heat, cold, and oxidation (Elbein et al., 2003).
Bryophytes usually grow in low temperature and high humidity area. Cold shock protein may be essential for the adaptation of actinobacteria to live with bryophytes under such environment. M. chalcea CMU55-4 genome is equipped with CspA family of cold shock protein and several heat shock protein genes. This protein plays an essential role in inhibition of DNA replication during cold-adaptation in Streptomyces sp. AA8321 (Kim et al., 2007). Similarly, heat shock proteins (HSP) can be benefit to bryophytes for survival under high temperature. HSPs remove denatured proteins to prevent formation of large protein aggregates and cell death while DnaK/DnaJ/GrpE or GroEL/GroES chaperones are part of sigma-32 heat shock regulon that regulates cell during heat stress and maintains protein homeostasis in living cells (Schumann, 2016).
In conclusion, this study provides the first evidence of cultivable actinobacteria associated with three high altitude moss species, Bryum apiculatum, Syntrichia gemmascens and Campylopus involutus. These actinobacteria show plant growth promoting ability in vitro. The inoculation of the selected strain, M. chalcea CMU55-4, can promote the growth of rare moss species, P. sphaericum. The response of this moss to M. chalcea CMU55-4 suggested that actinobacteria of the genus Micromonospora might occur naturally in association with the moss and might commonly affect moss development in nature. Genome mining data also support plant growth promotion potential of M. chalcea CMU55-4 as a good candidate for breeding program of P. sphaericum especially those under acclimatization and other moss species.

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be found in online the repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm. nih.gov/genbank/, JAAOLH000000000.

AUTHOR CONTRIBUTIONS
CI performed overall experiments and wrote the manuscript. WP, NC, and ST supervised CI. CI, WP, NK, and NC revised the manuscript. WP conceived the idea and designed research outline. All authors contributed to the article and approved the submitted version.

FUNDING
The authors would like to acknowledge funding support from the Development and Promotion of Science and Technology Talents Project (DPST), Ministry of Education, Thailand, and Canada-ASEAN Scholarships and Educational Exchanges for Development (SEED) program, supported by the Government of Canada. This work was also partially supported by the Chiang Mai University.