Tocklai vegetative (TV) clones TV1, TV9, TV22, TV25, and Teenali 17 were explored for the habitation of probable endophytic actinobacteria. These tea clones were selected based on their economic value. These TV clones were vegetatively propagated by Tocklai Tea Research Institute (TTRI), Jorhat, Assam, India, and released to the industry for commercial cultivation, especially in Northeast India. These clones were identified manually with the help of experts in the tea industry. Healthy plant parts (leaf and roots) from three-year-old tea bush were collected from Kopati Tea estate, Assam, India (26.5875°N, 92.2507°E) and MEG Tea estate, Meghalaya, India (25.741729° N, 91.889094° E) by random sampling method ( Table 1 ). The samples were collected aseptically in zip lock bags and brought to the laboratory in an ice box, and further processed for isolation within one month.

Putative endophytic actinobacteria was isolated by the method described in our previous study (Hazarika et al., 2021). Briefly, surface-sterilized plant parts were macerated using mortar and pestle. Tissue extracts were then serially diluted in 0.9% sodium chloride and plated in selective media, namely Actinomycetes Isolation Agar, Streptomyces agar, ‘King’s B agar, International Streptomyces Project (ISP) series agar ISP7, ‘Pikovskaya’s agar, Nutrient agar with the addition of 75 µg L^-1 of Amphotericin B and 2.5 µg L^-1 of Rifampicin were used. Further, culture plates were incubated at 28 ± 2°C for 21 days.

All the pure EnA isolates were grown on GLM agar plates containing 3 g L^-1 each of malt extract and yeast extract, 5 g L^-1 of Peptone, 10 ml L^-1 of Glycerol, and 20 g L^-1 of Agar, pH 7.4 at 28 ± 2°C for 10 days for initial phenotypic characterization and to obtain clonal isolates. The distinguished colonies were characterized based on typical actinobacterial phenotypic morphologies such as aerial and substrate mycelia, filamentous growth, dusty, rough, dry, and frequently pigmented colonies. These morphological characteristics of the EnA isolates were accounted based on the International Streptomyces Project (Shirling and Gottlieb, 1966).

16S rRNA gene sequences were analyzed to determine the taxonomic relationships among the EnA isolates. Genomic DNA was extracted using Nucleopore Fungus/bacteria kit (Genetix, India). The concentration and purity of the eluted DNA were quantified by a Nanodrop spectrophotometer (Thermo Fisher Scientific, United States). 16S rRNA gene was amplified using universal primers 27F and 1492R (Hazarika et al., 2021). PCR cycling conditions were: 5 min at 94°C, 30 s at 94°C, 30 s at 52°C, 1 min at 72°C, and 7 min at 72°C. The 16S rRNA gene sequence of each isolate was compared with the homologous sequence using the BLAST (Basic Local Alignment Search Tool) program of NCBI (National Center for Biotechnology Information) to identify the closest known taxa. The 16S rRNA gene sequences obtained were submitted to GenBank after chimera check Decipher 2.20 (Wright et al., 2012), and accession numbers were obtained. The evolutionary relationship between the isolates was studied by the maximum likelihood method using RAxML (Stamatakis, 2014). Finally, with 1000 steps of bootstrap analysis, a phylogenetic tree was constructed and visualized using iTol (Letunic and Bork, 2019).

The species richness (S), species evenness (J), the Shannon diversity index (H’), and the Simpson diversity index (D or SDI) of the isolated endophytic actinobacteria were calculated and compared to literature (Shannon, 1948; Simpson, 1949; Pielou, 1966)

Diversity analysis (Shannon index) was calculated as:

The dominance index (Simpson diversity index) was calculated as:

Species evenness was calculated as:

where, H' _max =lnS The Pi = ni/N, the proportion of individuals found in the ith species; ni is the number of the species i in the sample; S is the number of species (species richness); Ln is the natural log; N is the total number of individuals of all species; n is the number of individuals of a single species

Endophytic actinobacterial isolates were screened in vitro for the growth inhibition of phytopathogenic fungi, namely Exobasidium vexans (ITCC 938), Poria hypobrunnea (ITCC 4141), Phellinus lamaensis (ITCC 292), Fusarium oxysporum (MTCC 284), Rhizoctonia solani (MTCC 4633), Nigrospora sphaerica (KJ767520) which were collected from Indian Type Culture Collection (ITCC), IARI, New Delhi and Microbial Type Culture Collection (MTCC), IMTECH, Chandigarh and maintained in laboratory condition for experimental use. The pure culture of different pathogenic fungi was maintained in potato dextrose agar plates at 4°C, as glycerol stock at -20°C for experimental use, and as glycerol stock at -80°C for long storage. After a period of 3-4 months, the cultures are sub-cultured to maintain their metabolic activities. To determine the inhibition ability of the EnA isolates, an experiment was done in GLM agar media using the conventional spot inoculation method and incubated for 10 days at 28 ± 2°C. The five-day-old culture of fungal test pathogens was spread on plates with EnA isolates. Plates with only the target pathogen were used as control. Each experiment was conducted in three replicates. Plates were incubated for 7 days at 28 ± 2°C. The antifungal effect was considered by correlating the diameter of fungal mycelial on test plates with control plates and estimating the fungal growth inhibition (GI%). GI % was computed using the formula C-T/C×100, where C and T are the mycelial diameter on plates with control and EnA isolates, respectively (Sgroy et al., 2009).

Extracellular chitinase production was determined by using chitin agar plate consisting of (per liter) 0.5 g Yeast extract, 1 g (NH₄)₂SO₄, 0.3 g MgSO₄.6H₂O, 1.36 g KH₂PO₄, 2 g agar amended with 1.5 g colloidal chitin. The screening was done by spot inoculation and incubation of agar plates at 28 ± 2°C for 10 days. Chitinase production was confirmed by observing the clear or halo zones developed in the vicinity of the colonies (Roberts and Selitrennikoff, 1988).

The tomato (Solanum lycopersicum) seeds were surface sterilized with 2% sodium hypochlorite (NaOCl) for 10 min, and to remove traces of the disinfectant, the seeds were rinsed with sterile water for 5 min. The surface-sterilized seeds were then allowed to dry in a sterile environment. Surface sterilization was followed by seed bacterization by the method previously described (Niknam and Dhawan, 2003). In brief, fresh cultures of KA12 actinobacteria were prepared in GLM broth. Before seed bacterization, seeds were coated with 0.1% carboxymethylcellulose (CMC), followed by the addition of actinobacterial suspension (10⁸ cells mL^-1) as seed drench. For each experiment, a ratio of 10 ml per 30 seeds was applied and kept for 30 min, followed by air drying the seeds for 3 h in a sterile condition. The seed bacterization was validated by serial dilution and spread plating method.

To determine the seed biopriming effect of KA12, tomato seeds were treated with 100 μl of freshly cultured F. oxysporum spore suspension and incubated for 24 h. Seeds soaked under the same conditions in 10 ml of sterile distilled water with and without fungal inoculation served as a positive and negative control, respectively. Additionally, seeds incubated with GLM broth and 0.1% CMC served as the negative control. The seeds were then aseptically placed onto 3% w/v water agar plates moistened with sterile filter paper. Plates were incubated under a 16 h light/8 h dark cycle for up to 14 days at 28 ± 2°C. The degree of disease symptoms and inhibition of F. oxysporum on tomato seeds were determined by visual inspection after 14 days of inoculation. In addition, the seed germination rate per plate was determined, and the root architecture was studied in a stereomicroscope (Stemi 508, Carl Zeiss Microscope.). Each seed was assigned a disease severity score from 0 to 5, with 0 indicating no symptoms, 1 indicating less than 10% infection, 2 indicating 10% infection, 3 indicating 25% infection, 4 indicating 50% infection, and 5 indicating more than 75% infection. Further, the rate of seed germination was also scored, with 0 indicating no germination, 1 indicating germinated seed with seed infection, and 2 indicating germinated seed with no infection. The overall score in each treatment was calculated by multiplying the score awarded to each treatment by the number of seeds (n). Each experiment included three replications and 15 seeds per treatment (n).

To evaluate plant growth in tea ‘clones’ an experiment was carried out in sleeves under the effect of actinobacterial inoculum. Three healthy tea clones of 3-4 months old, namely TV22, TV26, and TV9, were collected from Bateli Tea estate, Assam. The sleeves contained sandy loam soil with the succeeding characteristics: 50 - 70% sand greater than 0.2 mm, 0 - 50% silt 0.002 - 0.2 mm, and 15–20% clay less than 0.002 mm of good tilth with an ideal pH of 4.5 (Hazarika et al., 2021). For inoculum preparation, two potent actinobacterial strains KA12 (Streptomyces sp., GenBank Accession number MN483270.1) and MA34 (Streptomyces sp. GenBank Accession number OK483112.1), were cultured in GLM broth at 28 ± 2°C for 10 days as optimized. The obtained biomass was harvested by centrifugation at 7000 rev min^-1 for 10 min, washed thrice with 1X phosphate-buffered saline (PBS) buffer, and adjusted to 10⁸ CFU mL^-1. In nursery conditions (average temperature of 30°C, average soil moisture of 5-6 RH, light intensity of 250 μmol m^-2 s^-1), the tea clones were treated for 15 days for 6 months by the soil drenching method. Three treatments were conducted on the tea saplings, including T1: KA12 inoculation, T2: MA34 inoculation, T3: consortia of KA12 and MA34, and control, without actinobacterial inoculum. Each experiment included three replicates. Vegetative parameters of plants, including the fresh and dry weight of shoot and root, shoot and root length, and the number of leaves, were recorded and compared to control.

To determine the root colonization ability of the potent endophytic isolates to C. sinensis and S. lycopersicum roots, rhizospheric soil was carefully detached and washed with sterile distilled water to remove any adhered soil particles. To release the actinobacteria into the buffer, roots of 1 cm long uniform segments were cut and dipped into 5 ml of sterile 1X PBS buffer, pH 7.4, and vortexed for 5-6 times to release the actinobacteria into the buffer. To evaluate the population of actinobacteria, solutions of actinobacterial suspensions were spread on GLM agar plates and incubated for 48-72 h at 28 ± 2°C, followed by counting of CFU. The primary indication of their competence to endophytically colonize the roots was determined by the presence of viable actinobacteria (Passari et al., 2016b; Singh and Jha, 2016). In vitro biofilm formation by actinobacterial isolates was also studied by Confocal Laser Scanning Microscope (CLSM) (Hazarika et al., 2021).

The chlorophyll content in tea leaves was determined as described by (Hiscox and Israelstam, 1979; Hazarika et al., 2021). 100 mg of mature tea leaves were obtained from the treated and untreated tea clones for extraction of chlorophyll. The amount of chlorophyll was determined at 645 nm and 663 nm and calculated by using the following formula:

where A= absorbance at specific wavelengths; V= final volume of chlorophyll extract; W= fresh weight of tissue extracted.

Following the in vivo plant growth promotion experiment, the shoot elemental content of treated and untreated plants was evaluated. Sample digestion was carried out as outlined by AOAC method 935.13. For instance, 2.5 g dry matter of leaf were added to 30 ml of nitric acid (HNO₃) in a digestion flask and refluxed for 30 min and continued till brown fumes disappeared. It was allowed to cool to room temperature and 10 ml of 70% perchloric acid (HClO₄) was added and refluxed to make the solution clear. 50 ml of distilled water was then added and heated until effervescence was minimal. The final volume was adjusted to 100 ml using distilled water and filtered for further elemental analysis (AOAC, 2000; Almehdi et al., 2019). The phosphorus (P) content in the filtrates was determined by alkalimetric ammonium molybdophosphate method as described by AOAC official method 964.06 (AOAC, 2000). The other elemental components including Fe, Cu, Zn, Na, Ni, Mg, Mn, and Ca was assessed by Atomic Absorption Spectroscopy (AAS) (Model: AA7000, Shimadzu) as described by AOAC official method 975.03. CHNSO Analyzer (Model: FLASHSmart, Thermo Scientific) was used to determine the nitrogen (N) content in the shoots by automated method as described by AOAC official method 972.43 (AOAC, 2000).

All the experiments were performed in replicates of three and represented as mean ± standard deviation (SD). To study the significance level in the interrelationships among the treatments and vegetative parameters, the data obtained were statistically analyzed by two-way ANOVA (analysis of variance) in SPSS package 28.0. Variance in elemental analysis was determined by one-way ANOVA and post hoc test by Tukey’s highest significance difference. The significance of the effect was determined at p ≤ 0.05. In addition, fold change analysis in a log2 scale was established to study the up or down-regulation of parameters in the presence of treatments compared to the control. Further, multivariate principal component analysis (PCA) was administered on the datasets using (MATLAB R (2017a)). PCA analysis established a relationship among the samples to assess the influence of treatment on plant growth promotion.

A total of 88 endophytic actinobacterial isolates of distinct colony morphologies were isolated from five different commercially important tea clones collected from the tea estates of Assam and Meghalaya ( Figure S1 and Table S1 ). Largest part of the endophytic actinobacteria isolates belonged to the roots (n = 63, 71.59%) followed by leaves (n = 25, 28.40%).

Based on 16S rRNA gene sequencing results and analysis using the nucleotide BLAST tool of NCBI ( Table S2 ), 88 actinobacterial isolates were divided into 9 genera. The respective genera abundance was different among the samples of tea clones TV1, TV9, TV22, TV25, and Teenali 17. Isolates belonging to genera Streptomyces was most abundant in all the tea clones irrespective of the plant parts ( Figure 1 ). Taxonomy delineation revealed that all the strains belong to Phylum Actinobacteria, Class Actinobacteria, 5 orders, namely Streptomycetales, Micrococcales, Mycobacteriales, Streptosporangiales, and Propionibacteriales, 7 families namely Streptomycetaceae, Microbacteriaceae, Intrasporangiaceae, Nocardiaceae, Gordoniaceae, Nocardiopsaceae, and Kribbellacea, and 9 genera namely Streptomyces, Microbacterium, Curtobacterium, Janibacter, Rhodococcus, Nocardia, Gordonia, Nocardiopsis, and Kribbella. The evolutionary relationship between the isolates is shown in the phylogenetic tree constructed based on the maximum likelihood method ( Figure 2 ).

Our study observed an average population size (µ) of 9.78. The species richness index (S) was found to be 9, and due to the reduced number of culturable isolated species, the diversity indexes were calculated for the overall samples. The Shannon diversity (H’), Simpson dominance (D), and evenness index (J) of the endophytic actinobacteria isolated from the tea clones were calculated. H’ was 0.934, which is based on randomness and represents a less diverse population. D was 0.623, which expressed overall dominance that considers the proportion of species. The uniformity of individual species in a community is determined by J, which ranges between 0 to 1(0 - no evenness and 1- complete evenness). The J index was 0.425, signifying a non-even distribution of genera among the isolated species.

In vitro assessment of 88 endophytic actinobacterial isolates for the production of plant growth-promoting traits using qualitative and quantitative assays revealed that 59 (67.05%) isolates produced phytohormone IAA in the range 3.09 - 143 µg/ml, 24 (27.27%) isolates solubilized calcium phosphate available in the media in the range 2.48 - 62.19 µg/ml, 71 (80.68%) isolates produced ammonia in the range 0.39 - 4.92 µmol/ml, 66 (75%) isolates were found to be siderophore producers in the range 14.09- 86.33% SU. Nitrogen fixation was shown by 51 (57.95%) isolates, by their ability to grow in nitrogen-free media. 42 (47.73%) isolates produced ACC deaminase, an enzyme involved in abiotic stress alleviation, Maximum IAA of 143 µg/ml was produced by isolate KA12 (Streptomyces sp.), isolate MA16 (Streptomyces sp.) solubilized highest amount of phosphate of 62.19 µg/ml, maximum ammonia of 4.92 µmol/ml was produced by isolate M15 (Microbacterium sp.), isolate MA37 (Streptomyces sp.) produced maximum siderophore of 86.33% SU among the 88 isolates. Most isolates belonging to the genus Streptomyces were observed to fix nitrogen and produce ACC deaminase ( Table 2 ). The capability of endophytic actinobacteria to ameliorate plant growth was also represented by a phylogenetic approach. Most of the positive strains were related to genus Streptomyces. The results obtained are illustrated in Figure 3 . Richness in the PGP traits was consistently observed in the isolates KA12 (Streptomyces sp.), KA83 (Streptomyces sp.), MA3 (Nocardiopsis sp.), MA34 (Streptomyces sp.), and MA45 (Streptomyces sp.).

Among the 88 EnA isolates, 35 (39.77%) of the isolates were observed to exhibit antifungal activities against at least one of the tested pathogens. The highest inhibition activity was ascertained against Fusarium oxysporum, followed by Exobasidium vexans and Poria hypobrunnea ( Figure 4 and Figure S2 ). The isolate KA12 showed extensive antifungal activity against most of the tested pathogens. Additionally, the highest reduction in mycelial growth of 90.62% was exhibited by isolate MA40 against F. oxysporum, followed by KA62 against P. lamaensis (90%), KA13 against E. vexans (80%), KA13 against R. solani (75%), KA20 against N. sphaerica (67.86%), and KA10 against P. hypobrunnea (65%) ( Table 2 ).

Chitin, a polymer of 1,4-β-linked N-acetyl-D-glucosamine, is the building block of fungal cell walls. Chitinases are enzymes that degrade chitin and exhibit antagonism against fungus. Thus, actinobacteria with chitinolytic activity will be beneficial for the biocontrol of pathogens. In our study, in vitro screening of chitinolytic activity by the EnA isolates was determined using substrate chitin in the medium. 51 (57.95%) isolates showed chitinase enzyme production by forming clear halo zones around the colonies ( Table 3 ).

In order to select the most performing EnA isolates, the isolates were organized hierarchically (a non-arbitrary ranking approach). The strains were ranked based on in vitro screening for their potential to promote plant growth and biocontrol ability. The obtained ranking was compared to the bonitur scale, where all the PGP and biocontrol attributes were allocated points. The utmost attainable score here is 29 points: IAA production and phosphate solubilization were each given 3 points, nitrogen fixation, ammonia, siderophore, ACC deaminase, and chitinase production were given 1 point each. Depending on the percentage of inhibition, antifungal activity against six tested pathogens was assigned 3 points each leading to a sum of 18 points ( Table 4 and Table S3 ).

Biofilm formation by the potent EnA strains KA12 and MA34 was found to be strong during the stationary growth phase at 96 h. The significant biofilm formation, as examined by the crystal violet staining method, started after 24 h of incubation ( Figure 5A ). Along with this, CLSM studies confirmed the biofilm formation on glass slides. The biofilms formed in the glass slides by KA12 and MA34 appeared to be 9 µm and 7 µm thick, respectively, as demonstrated in the z-plane by CLSM microscopy ( Figures 5B–E ).

Plant growth-promoting actinobacteria can also act as a defensive symbiont by biopriming response. In this study, strain Streptomyces sp. (KA12) was found to have potential antagonistic activity against phytopathogens viz. F. oxysporum, P. hypobrunnea, E. vexans, R. solani, and N. sphaerica. Hence, KA12 was evaluated as seed biopriming agent. The seed germination rate (%) of bacterized tomato seeds showed enhanced germination of 80 ± 0.43 compared to the non-bacterized control seeds, which were 20 ± 0.28 – 26.6 ± 0.4. The seeds inoculated with test isolate and fungal pathogen showed a seed germination rate (%) of 73.33 ± 0.25 ( Figure 6A and Table S4 ). The severity of disease by inoculating F. oxysporum in bacterized tomato seeds was reduced to a scale of 0.6 compared to the non-bacterized seeds with disease severity of 4.66 ( Figure 6B ). The results obtained by conducting both pot experiments and water agar plate assay displayed resemblance confirming the biopriming effect of KA12 and effectiveness for fungal resistance ( Figures 6C, D ).

On the visualization of root architecture using a stereomicroscope, variation of root morphology was observed among treated and control groups. The number of root hairs in the control group was less than that compared to the roots colonized with KA12. The number of roots in the treatment group with KA12 and the fungal pathogen was also more than in the untreated control ( Figure 6E ). Thus, this indicates that the isolate KA12 enhances the root hair formation, thereby improving the soil anchorage and positively influencing growth.

In all the experiments performed in nursery condition, on comparison of the treated plants to the control plants, a significant (p < 0.05) increase in all the vegetative parameters were observed ( Figure 7 and Figure S3 ). Further, the level of growth in vegetative parameters of treated plants correlated to control plants was determined by fold change analysis. The fold change level varied with the various EnA treatments in the tea clones used for the study. In all the tea clones TV9, TV22, and TV26, the actinobacterial inoculum of Streptomyces sp. KA12 (T1) showed a significant increase in growth parameters in terms of fold change compared to other inoculums. In the TV9 clone, T1 enhanced fresh and dry biomass of roots by 4.42-fold and 3.8-fold, and shoot biomass by 4.15 and 3.6-fold, respectively, root and shoot length was increased by 4.68-fold and 5.6-fold, respectively, and leaf number was increased by 4.56-fold. However, the degree of growth promotion in TV22 and TV26 clones varied with various treatments. In the TV22 clone, enhancement in shoot biomass (3.38-fold) and leaf number (4.17-fold) were observed in treatment with T1. The increase in root length (3.98-fold) was observed in treatment with T2. The enhancement in root biomass (2.79-fold) and shoot length (5.29-fold) were observed in treatment with T3. In the TV26 clone, an increase in shoot biomass (3.2-fold), root length (4.38-fold), and leaf number (3.7-fold) were observed in treatment with T1. The enhancement in root biomass (3.78-fold) was observed in treatment with T2, and the increase in shoot length (5.22-fold) was observed in treatment with T3. All these growth augmentations were observed compared to the untreated control plants. The graph of fold change analysis and their log values are provided in Table 5 and Figure S4 .

The various PGP parameters were reduced to two principal components, PC1 and PC2, by principal component analysis (PCA) analysis. A maximum variance of 70.96% was found in PC1, followed by 15.21% in PC2 ( Figure 8 ). As indicated in the PCA plot, a distinct variance was observed throughout the PC1 between treated and control groups as the treatments formed separate clusters in all the tea clones used. The degree of correlation of parameters fresh and dry weight root (FWR and DWR) was represented toward clones TV9 and TV26 in all the treatment groups. Likewise, in all the treatment groups, fresh and dry weight shoot parameters, root length, the number of leaves, and shoot length were correlated mostly toward clone TV22.

The extent of effective colonization of symbiotic actinobacteria determines the influence of plant-plant growth-promoting bacteria interactions on plant growth and yield. Thus, we assessed the efficiency of Streptomyces sp. KA12 to colonize tomato plant by determining the CFU, which was found to be 7.2 x 10³ CFU/g of the root after inoculating tomato plantlets for 15 days. However, sparse actinobacterial colonies of 1.1 x 10² CFU/g were recovered from control plants. On the assessment of the efficiency of Streptomyces sp. KA12 and Streptomyces sp. MA34 to colonize tea plants, CFU was found to be 4.7 x 10³ and 4.1 x 10^3, respectively.

In the treatment group, tea clones showed an increase in chlorophyll content. Comparing the chlorophyll a and b content of the treated tea clones to the control group, the variance was found to be significant at p < 0.05 ( Table 6 ; Figure 7 ). This denotes the impact of prospective actinobacteria on photosynthesis and plant development.

The results of elemental analysis are listed in Table 7 . The endophytic actinobacteria inoculation affected the plant uptake of P, N, Fe, Cu, Zn, Na, Ni, Mg, Mn, and Ca. It was revealed that the inoculation with Streptomyces sp. KA12 and MA34 and their consortia significantly (p < 0.05) increased the uptake of these elements. The shoot and root biomass, number of leaves did not differ between treatment at harvest, but they significantly increased as compared to the untreated control tea plants, thereby influencing plant growth promotion ( Table 6 ; Figure 7 ).