Introduction

Plants offer diverse habitats for numerous microorganisms. Parasites, mutualists and commensals live in plant rhizosphere, inside plant tissues as endophytes, and on the surface of the aerial organs. The plant-associated habitats provide resident microorganisms with nutrients and energy, which is likely to apply a selection pressure on the microorganisms. Microbes in the rhizosphere benefit from exudation of organic compounds from roots and from dead root material, microbes living on above ground plant surfaces can benefit from nutrients leaching from plant tissues, and microbes living inside plants can access nutrients directly1. Abiotic factors, including water availability, temperature and solar radiation may directly affect the growth and activities of microorganisms in these habitats. Many of the plant associated microbes may directly and indirectly contribute to plant growth and stress resistance by various mechanisms, including increased availability of minerals, nitrogen fixation, and production of hydrolytic enzymes and phytohormones.

Actinobacteria, Gram-positive bacteria with a high genomic G + C content, are widespread environmental organisms and found in both terrestrial and aquatic habitats. The diversity of ecologically important plant associated actinobacteria is significantly affected by plant tissue type and growth stage, and soil nutrient availability2,3,4,5. Plant-associated actinobacteria may affect plant growth and improve the stress resistance of their host plants6,7,8,9. In addition, plant associated actinobacteria are a potential source of novel bioactive metabolites10,11. Many plant associated actinobacteria produce antifungal or antibacterial agents, for example extracellular hydrolytic enzymes that lyse fungal cell walls12.

Liquorice (Glycyrrhiza spp.) is one of the most ancient herbal medicines. The root and rhizomes of Glycyrrhiza inflata Bat., Glycyrrhiza glabra L. and Glycyrrhiza uralensis Fisch. have been widely used as a flavoring agent and for a variety of pharmaceutical applications for thousands of years in southern Europe and parts of Asia13. G. inflata is found mainly in Xinjiang, China14. It is one of the main sources of liquorice in Chinese medicine due to the presence of a wide variety active ingredients, for example triterpenoids, flavonoids, and polysaccharides15,16. Glycyrrhiza spp. belong to Leguminosae, and they are nodulated by diverse rhizobia with plant growth promoting (PGP) activity17,18,19. Generally, liquorice grow in Central Asia, Mongolia, Iraq and the northwest of China, in regions characterized by harsh environmental conditions, including high temperatures and evaporation, high salinity, low precipitation, poor soil condition, and strong winds and UV irradiation. Liquorice plants have been applied to remediate saline soils20. Inoculation with plant growth promoting (PGP) bacteria may benefit the remediation process, since they can increase germination and seedling growth in saline conditions21,22.

In our previous study we characterized the diversity and antimicrobial activity of actinobacterial isolated from G. inflata and G. glabra23. However, to our knowledge the plant growth promoting activity of actinobacteria associated with Glycyrrhiza spp. has not been studied. Therefore, our aim was to assess the plant growth-promoting properties, salt tolerance and antimicrobial activities of actinobacteria isolated from G. inflata to find strains that could be applied in agricultural industry, for example in reclaiming saline soils. Since the endophytic communities in different plant organs may differ and change during plant growth24, we sampled bark, leaf, root, and stem from both young and mature plants to increase our possibilities to isolate strains with desired characteristics.

Materials and Methods

Sample collection

Healthy one year old and three year old Glycyrrhiza inflata Bat. plants were randomly collected from Tarim in Xinjiang, China. Plants were sampled in triplicates. The sampling area is arid desert characterized with low rainfall and high evaporation. The soil is classified as sandy soil. The plants were dug out and bulk soil was removed by gentle shaking. Plants were kept at 4 °C, brought to the laboratory and processed immediately. The bark, leaf, root, and stem were separated and surface sterilized as described previously25. Aliquots from the final rinse were incubated on ISP2 media at 28 °C for 3–4 weeks. The sterilization was regarded effective when there was no growth.

Isolation and preliminary identification of endophytic actinobacteria

Surface-sterilized plant samples were aseptically cut into small fragments using commercial blender. Subsequently, the fragments were plated onto five selective isolation media: Tap Water Yeast Extract Agar (TWYE)26, Starch Casein Nitrate Agar (SCNA), Chitin Agar, Humic-vitamine Agar (HV)27, and Oatmeal Agar (ISP3). Isolation media were supplemented with nalidixic acid and K2Cr2O7 (50 µg ml−1) to inhibit the growth of non-actinobacteria. Purified isolates were stored on ISP4 slope medium at 4 °C.

The isolates were preliminarily identified by cultural and morphological characteristics as described previously25 using light microscopy (Olympus CX31, Olympus Corp., Japan) to observe the spore chain morphology of isolates grown for 10 d on ISP4 media.

DNA extraction, PCR amplification and DGGE analysis

DNA was extracted from 100 mg of fresh tissue with Power PlantTM pro DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA) according to the manufacturer’s instructions. Extracts were stored at −20 °C. In the first round of a nested PCR 16S rRNA gene was amplified using the primers 243F (5′-GGATGAGCCCGCGGCCTA-3′)28 and 1186R (5′-CTTCCTCCGAGTTGACCC-3′)29 in a PCR mixture containing 10 µl MIX buffer (Premix TaqTM, TaKaRa, China), 1 µl template DNA, 1 µM each primer, and sterile distilled water to the final volume of 20 µl. In the second round a fragment was amplified using the primers 907F (5′-AAACTCAAAGGAATTGACGG-3′)30 with a GC-clamp and 1186 R in a PCR mixture containing 25 µl MIX buffer, 1 µl of the first PCR product as template, 1 µM each primer, and sterile distilled water to the final volume of 50 µl. The touchdown PCR was conducted as described previously31. Amplification of the approximately 270 bp target fragment was verified by electrophoresis in 2% agarose gel.

PCR products were loaded onto a 8% (w/v) polyacrylamide gel with a 30–60% denaturant gradient in Tris acetate EDTA32 buffer and separated for 8 h at 60 °C and 160 V using a Dcode Universal Mutation Detection System (Bio-Rad, USA). After electrophoresis, the gels were silver stained as described earlier33. Gel images were acquired using a Gel Doc imaging system (Bio-Rad) and analyzed using Quantity One version™ software. The predominant DGGE bands were excised and reamplified and sequenced at Suzhou GENEWIZ Biological Technology Co., Ltd. (Suzhou, China). The sequences were compared with sequences in the NCBI Genbank nucleotide database using BLASTN to find the closest matching sequences.

RFLP, sequencing, and phylogenetic analysis of cultivable actinobacteria

Genomic DNA was extracted and purified as described earlier34. The 16S rRNA genes were amplified with forward primer 27F (5′-CAGAGTTTGATCCT GGCT-3′) and reverse primer 1492R (5′-AGGAGGTGAT CCAGCCGCA-3′)35. The PCR products were digested with restriction endonucleases HhaI (TaKaRa, China) for 2 h. The digested fragments were separated in a 2% agarose gel by electrophoresis for 3 h at 60 V and visualized with an UV transilluminator. Isolates were grouped based on the restriction fragment patterns36. A phylogenetic tree was constructed using the Unweighted Pair Group with Arithmetic Mean (UPGMA) method in NTSYS 2.1 software37. Representative isolates were chosen for 16S rRNA gene sequencing in Suzhou GENEWIZ Biological Technology Co., Ltd. (Suzhou, China). Sequences were compared with NCBI GenBank database using BlastN to find the closest matching sequences. The sequences were pairwise aligned using Clustal X38. A phylogenetic tree was constructed under the Kimura two parameter model and bootstrap analyses with 1,000 resamplings using MEGA 6.039.

Physiological characteristics of the representative strains

Production of indole-3-acetic acid (IAA) and siderophore secretion were assessed as described earlier40,41. The phosphate solubilizing ability was evaluated by using insoluble Ca3(PO4)2 as sole P source in Pikovskaya’s medium42. Chitinase activity was estimated as recommended by Xiang et al.43. Salt resistance was tested by growing the isolates in ISP4 media with 0 mM, 100 mM, 200 mM, 300 mM, 400 mM, and 500 mM NaCl at 30 °C for 10 days.

Evaluation of antimicrobial activity

Representative isolates were tested for their antagonistic activity against seven indicator organisms: Mycogone perniciosa Magn [SCAU3216], Curvularia lunata Boedijn [SCAU3697], Alternaria alternata (Fries) Keissler [SCAU3471], Fusarium graminearum Sehw. [SCAU3741], Fusarium oxysporum [SCAU3221], Staphylococcus aureus [ATCC 25923], and Escherichia. coli [ATCC35218]. The antagonism was measured as the distance from the mycelium edge to the margin of actinobacterial colony. All strains were tested in triplicates.

Plant growth promotion assay

Eight strains that were resistant to 400 mM NaCl and produced IAA were selected to study their effects on seed germination under salt stress. To obtain enough spores, the pure cultures were spread on ISP4 agar plates and incubated for 5–6 days, after which the agar medium was cut into small pieces. The pieces were transferred on sterilized wheat grains, and incubated at 28 °C until the grains were completely covered with mycelia and spores. The spores were washed off the grains by sterilized distilled water to make a final concentration of 1.0 × 108 CFU mL−1 as described previously2. G. inflata seeds were surface sterilized in 1% HgCl (v/w) for 10 min, rinsed three times in sterile distilled water, and inoculated by soaking into the spore suspension for 8 h. Negative control seeds were soaked into sterile distilled water. Seeds were transferred aseptically on MS medium with 0 mM, 100 mM, 200 mM, 300 mM, and 400 mM NaCl with 30 seeds per plate. Treatments were done in three replicates. Germination rate was calculated after 5 day incubation at 28 °C.

Based on the results of the germination test, the isolates SCAU5283, SCAU5276, SCAU5201 and SCAU5207 were selected to test their plant growth promotion activity on G. inflata. The spore suspension and G. inflata seeds were prepared as above. G. inflata seeds were germinated on MS medium with 200 mM at 28 °C. After 3–5 days germination, three seedlings were planted into polypropylene cup filled with a sterilized mixture of washed sand, vermiculite, and ceramic gravel. The surface was covered with 1–2 cm sterilized quartz sand. Cups were put on glass jars filled with sterilized Hoagland’s solution44 supplemented with 200 mM NaCl. The seedlings were inoculated with 50 µl of spore suspension around the seedling root. Negative control seedlings were inoculated with 50 µl of sterile distilled water. The treatments were done in three replicates. Seedlings were grown for 45 days in an illuminating incubator using 18 h light period and 6 h dark period at 24 °C and 16 °C, respectively. After harvest, the dry weight, shoot and root length, and N, P, and K contents were measured to evaluate the effect of strains on plant growth. Total N, P, and K contents were determined as described by Liu et al.45.

Statistical analysis

Principal component analysis based on the presence/absence of physiological characteristics was done in Canoco 5.046 to visualize the grouping of strains from one year old and three year old plants. Differences between numbers of strains from one year old and three year old plants with antimicrobial activity were tested with Fisher’s exact test. Germination percentages were transformed using centered log ratio transformation (clr), and tested using one-way analysis of variance (ANOVA) and Tukey’s post hoc test. Plant properties and inhibition zone data were analyzed using ANOVA. Significant differences between means were compared using Duncan’s multiple range test at p < 0.05. The results were expressed as mean ± SD. Statistical analyses were performed using the SPSS version 20.0 software package for Windows, R statistical software47, and package compositions in R48.

Results

Isolation and identification of strains

Altogether we isolated 36 and 52 strains that showed morphological characteristics of actinobacteria from one year old and three year old G. inflata plants, respectively. Most of the strains were isolated from roots (n1Y = 17; n3Y = 25) followed by stem (n1Y = 8; n3Y = 12), leaf (n1Y = 8; n3Y = 10), and only three and five strains were isolated from bark of one year old and three year old liquorice plants, respectively.

The 36 strains isolated from one year old liquorice plants were assigned to six groups at 80% similarity level in the RFLP analysis (Fig. 1). The isolates formed one dominant group of 28 strains that were further separated into subgroups. The other five groups contained 1–2 strains. The 52 strains from three year old plants were assigned to ten groups at 80% similarity level (Fig. 2). The biggest groups contained 25 and 13 strains that were further divided into subgroups. The other eight groups contained 1–4 strains. The RFLP fingerprints of the strains from one year old plants were not detected among those from three year old plants and vice versa. Based on the RFLP, one to thirteen representative strains per group were selected for subsequent 16S rRNA gene sequencing and physiological analyses.

Figure 1
figure 1

UPGMA dendrogram based on the 16S r DNA PCR-RFLP fingerprints of strains isolated from one year old Glycyrrhiza inflata Bat.

Full size image
Figure 2
figure 2

UPGMA dendrogram based on the 16S r DNA PCR-RFLP fingerprints of strains isolated from three year old Glycyrrhiza inflata Bat.

Full size image

The 16S rRNA gene sequences of 13 representative strains from one year old liquorice plants were aligned with the 99.1–100% similar sequences of type strains retrieved from databases (Table 1). The strains belonged to the orders Streptomycetales, Corynebacteriales, Micromonosporales, and Micrococcales. Most of them belonged to genus Streptomyces (Table 1 and Fig. 3). Streptomyces, Micromonospora, and Rhodococcus strains were found in both root and stem, Streptomyces and Promicromonospora in leaf, and Streptomyces in fruit (Table 1).

Table 1 Actinobacteria strains isolated from Glycyrrhiza inflata Bat., and the closest type strains based on 16S rRNA gene sequence similarity.
Full size table
Figure 3
figure 3

Neighbour-joining tree based on 16S rDNA sequences of actinobacteria closely associated with Glycyrrhiza inflata Bat. The numbers at the nodes indicate the level of boot strap support (%) based on 1000 resamplings; only values above 50% are given. The scale bar corresponds to 0.02 substitutions per nucleotide position. Numbers in parentheses are the NCBI GenBank accession numbers. The strains isolated in this study are highlighted in bold.

Full size image

The 23 representative strains from three year old plants belonged to Streptomycetales, Micromonosporales, Micrococcales, Propionibacteriales, and Streptosporangiales with 98.5–100% similarity to the closest matching type strains (Table 1). The strains were more diverse than those from one year old plants, and represented ten genera: Streptomyces, Micromonospora, Actinokineospora, Arthrobacter, Actinomadura, Oerskovia, Cellulomonas, Nocardioides, Promicromonospora, and Rhodococcus (Fig. 3). Strains belonging to six genera were isolated from root (Table 1). Streptomyces strains were isolated from all organs, Actinokineospora strains were isolated from both root and stem, and an Arthrobacter strain from leaf.

To estimate if the isolated strains were representative of the actinobacterial diversity in G. inflata, twenty DGGE bands were excised for sequencing. The sequences were affiliated with nineteen genera, out of which four were identified among the isolated strains, suggesting that the isolation methods had captured less than half of the endophytic genera (Table 2).

Table 2 Identification and distribution of actinobacteria excised and sequenced from DGGE bands derived from Glycyrrhiza inflata Bat.
Full size table

Physiological characteristics of the strains

To further characterize the representative strains, their plant growth promoting (PGP) activity and salt tolerance were tested (Table 3). Nine out of thirteen (69.2%) and fifteen out of 23 (65.2%) strains isolated from one year and three year old plants, respectively, produced IAA at levels ranging from 11.3–71.8 mg L−1 and 2.3–46.2 mg L−1. SCAU5283 (71.8 mg L−1) and SCAU5215 (46.2 mg L−1) produced the highest amount of IAA among strains isolated from one year and three year old plants, respectively. Five (38%) and thirteen (56%) strains from one year and three year old plants, respectively, produced siderophores in an iron-deficient culture medium. Two (15.4%) and six (26.1%) strains isolated from one year and three year old plants, respectively, showed a clear halo zone around colony on Pikovskaya’s medium, indicating phosphate solubilization ability. Three (23.1%) and nine (39.1%) strains from one year and three year old plants, respectively, produced chitinase.

Table 3 Salt tolerance, antimicrobial activities and plant growth promoting properties of actinobacteria strains isolated from Glycyrrhiza inflata Bat.
Full size table

All strains grew in media with 100 mM NaCl. Streptomyces strains SCAU5201, SCAU5210, and SCAU5283 tolerated 500 mM NaCl. Nine (69.2%), six (46.2%), and three (23.1%) strains from one year old plants tolerated 200 mM NaCl, 300 mM NaCl, and 400 mM NaCl, respectively. Eighteen (78.3%), thirteen (56.5%) and seven (30.4%) strains from three year old plants tolerated 200 mM NaCl, 300 mM NaCl, and 400 mM NaCl, respectively. In the principal component analysis, the strains from one year old and three year old liquorice plants were not separated based on physiological characteristics (Fig. 4).

Figure 4
figure 4

Principal component analysis based on the presence/absence of antimicrobial and PGP activities of actinobacteria strains isolated from one year old (1Y) and three year old (3Y) Glycyrrhiza inflata Bat. Only isolates that showed at least one activity are included.

Full size image

Analysis of antimicrobial activities

The antimicrobial activities of representative strains were tested against seven indicator organisms (Table 3). Differences between numbers of strains with antimicrobial activity from one year old and three year old plants were not statistically significant. Nine out of thirteen (69.2%) strains from one year old plants and twenty out of 23 (86.9%) strains from three year old plants showed antogonistic activity against at least one of the seven indicator organisms.

Altogether 23 out of the 36 strains inhibited the growth of fungus Alternaria alternate (Table 3). At the other end, the growth of Fusarium oxysporum was inhibited by only ten strains. The growth of bacteria Staphylococcus aureus and Escherichia coli were inhibited by twelve and fifteen strains, respectively.

All 23 Streptomyces strains except SCAU5204 inhibited the growth of at least two indicator organisms (Table 3). Streptomyces SCAU5201 and SCAU5202 exhibited broad spectrum antimicrobial activities by inhibiting all the seven indicator organisms. In addition, four other Streptomyces strains (SCAU5212, SCAU5220, SCAU5270 and SCAU5281) inhibited the growth of six indicators. Seven out of thirteen rare actinobacteria strains did not inhibit any of the indicator organisms. Out of the rare actinobacteria, Actinokineospora SCAU5231 inhibited the widest range of indicator organisms, altogether four.

The plant growth promotion activity of selected actinobacterial strains

The eight Streptomyces strains that grew with 400 mM NaCl and produced indoleacetic acid (IAA) were selected for assessing their effect on G. inflata seed germination under salt stress. In line with most of the isolates being from roots, six of the strains were from roots, and the other two from leaf and stem. The seed germination rate decreased with the increasing NaCl concentration (Table 4). The higher the NaCl concentration, the more there were strains that did not differ from the non–inoculated control treatment. At 400 mM NaCl, the germination rate of the seeds inoculated with Streptomyces SCAU5283 were the highest (Table 4). Compared to the non–inoculated treatment, strains SCAU5201, SCAU5207, SCAU5276, and SCAU5283 increased the seed germination rate under all NaCl concentrations tested.

Table 4 Germination rates of un–inoculated and inoculated Glycyrrhiza inflata Bat. seeds at different NaCl concentrations. The inoculants were Streptomyces sp. strains isolated from G. inflata. The values are percentage ± standard deviation (n = 3). Different superscript letters in a column indicate statistically significant differences (p < 0.05, Tukey’s post hoc test).
Full size table

The above mentioned four strains were selected for assessing their effect of G. inflate seedling growth in a greenhouse experiment. Compared to the non–inoculated treatment, all the four strains increased plant shoot length, root length, dry weight, and N, P and K contents significantly (Fig. 5). All the measured parameters were greatest in plants inoculated with strain SCAU5283.

Figure 5
figure 5

The shoot length (a), root length (b), plant dry weight (c), and nutrient contents (d) of Glycyrrhiza inflata Bat. inoculated with Streptomyces sp. strains. Control = un-inoculated Glycyrrhiza inflata Bat. The values are mean ± standard deviation (n = 3). Different superscript letters on a column indicate statistically significant differences (p < 0.05, Duncan’s multiple range test).

Full size image

Discussion

Liquorice is known as “the king of Chinese medicine” that is widely applied in pharmaceutical and food industry due to its medicinal value and sweet taste. The liquorice plants (Glycyrrhiza spp.) tolerate harsh environmental conditions, and they may be applied for example in reclaiming saline soils20. Glycyrrhiza spp. are nodulated by rhizobial bacteria that fix atmospheric nitrogen and thus promote the growth of the host plant17,18,19. Like numerous plant species5, Glycyrrhiza spp. host endophytic actinobacteria23. Many endophytic bacteria have plant growth promoting (PGP) ability, and they can increase germination and growth of their host plants under environmental stress8,9,21.

Since the endophytic communities change over time24,49,50, sampling plants at different growth stages may increase possibilities to isolate strains with desired characteristics. We isolated actinobacteria from the roots, stems, leaves and bark of one year old and mature three years old liquorice plants, and tested their PGP and antimicrobial activities. In line with the observation that endophytic bacteria enter through roots and then migrate to other organs, most of the strains were isolated from roots. In addition to the genera Streptomyces, Micromonospora, and Rhodococcus isolated in our previous study23, in this study seven genera more were isolated from G. inflata. Most of them have been previously reported as endophytes of medicinal or other plants25,51,52,53,54. Actinokineospora spp. have been isolated from soil, plant litter and sponges55,56, but, to our knowledge, not from inside a plant. More diversity was revealed by DGGE, highlighting the need to develop cultivation methods to isolate rare actinobacteria species for assessing their PGP and antimicrobial activities.

The actinobacteria closely associated with plants have a long-held relationship with host plants, and they may play an active role in plant development and also protect the hosts against pathogens57,58. In our work, we assessed four PGP characteristics: production of indole acetic acid (IAA), siderophore, chitinase, and phosphate solubilization activities. All Streptomyces strains showed at least one activity, whereas over half of the rare actinobacteria strains did not show any. IAA is a plant growth promoting hormone, produced not only by plants themselves but also by many plants associated bacteria. As in earlier studies6,8,59, most of the IAA-producing strains belonged to genus Streptomyces. Siderophores chelate Fe (III), and siderophores secreted by actinobacteria contribute to plant protection by competing with potential pathogens for iron60. Many Streptomyces spp. produce siderophores61,62, and in our study all of the siderophore producing strains were affiliated with Streptomyces.

Phosphorus is one of the most important nutrients for plant growth and development. Phosphate solubilizing bacteria are effective in releasing P through solubilization and mineralization, and have been used as inoculants to improve the growth and yield of crop plants. A considerable number of bacterial species associated with plant rhizosphere have a high capacity in solubilizing P63. Among the endophytes, 19% of isolates from the medicinal plant Ferula songorica, half of the actinobacteria strains from seven medicinal plant species, and four out of nineteen isolates from Jatropha curcas solubilized P6,59,64. In our work, the proportion of P solubilizing strains was within the same range: eight Streptomyces strains solubilized P.

Actinobacteria isolated from various plant tissues inhibited pathogens by producing active compounds and chitinase65. Endophytic actinobacteria that produced chitinase protected plants against phytopathogenic fungi66. Endophytes with chitinase activity suppressed fungal pathogens by degrading cell wall and thus bursting spores and hyphal tips, thereby inhibiting spore germination and germ tube elongation32. In our study, all the strains with chitinase activity were able to inhibit pathogens. However, most of the antifungal strains did not produce chitinase, suggesting that those strains have alternative mechanisms to inhibit the growth of fungi.

Actinobacteria closely associated with terrestrial and marine plants are considered vital sources of secondary metabolites with potential antimicrobial activity67,68. Similar with our previous research23,25, almost all of the Streptomyces strains showed antimicrobial activity against at least one of the tested indicator organisms. In addition, some of the Actinokineospora, Cellulomonas, Actinomadura, Nocardioides, and Rhodococcus strains inhibited the growth of indicator organisms, indicating that rare actinobacteria are a potent storehouse that should not be ignored when searching for natural products.

In general, the Streptomyces strains tolerated higher concentrations of NaCl and inhibited the growth of greater number of indicator organisms than the rare actinobacteria. However, it should be noted that the difference between Streptomyces and rare actinobacteria in vitro does not necessarily indicate a difference in vivo. Streptomyces strains are relatively easier to cultivate than the rare actinobacteria69. Possibly the PGP, salt tolerance, and antimicrobial activities of the Streptomyces strains are also more strongly expressed than those of the rare actinobacteria.

Salt tolerant actinobacteria with plant growth promoting as well as antagonistic activity against pathogens could alleviate the deleterious effect of salinity6,59. We selected the eight strains that tolerated high level of salt and produced IAA to evaluate if the strains could promote G. inflata seed germination under salt stress in vivo. All the eight strains belonged to genus Streptomyces. In the germination assay at 200 mM and higher concentrations of salt, inoculation with the four strains that had produced highest amounts of IAA and solubilized P resulted in highest germination rates. Concluding that the strains affected germination through IAA would require further analyses. Exogenous IAA and IAA producing bacterial strains have increased germination rate under salt stress21. However, IAA is not thought to affect germination directly, yet it may interact with gibberellins and ethylene and indirectly affect germination70.

Phytohormone producing strains have been proposed to alleviate salt stress and facilitate plant growth in harsh environment21. We assayed the effect of the abovementioned four strains on the growth of G. inflata under salt stress in a greenhouse experiment. The growth of all the inoculated plants was significantly better than that of the un–inoculated plants. The growth promotion in vivo was not directly related to the degree of IAA production in vitro; strains SCAU5215 and SCAU5201 outperformed SCAU5276 that produced higher amount of IAA. The best promoters of G. inflata growth, strains SCAU5283 and SCAU5215, inhibited a wide range of indicator organisms, and may thus be considered as promising candidates to be applied in inoculating G. inflata in reclaiming saline soils.

In summary, the actinobacteria strains isolated from G. inflata represented ten genera. Most of the strains had plant growth promoting characteristics in vitro, tolerated 200 mM NaCl and inhibited the growth of at least one indicator organism. The eight selected Streptomyces strains increased the germination rate of G. inflata seeds under salt stress. In addition, the four best seed germination promoters promoted the growth of G. inflata in vivo.