1021Bp isolation

1021Bp used for this study was subcultured from an isolation plate spread with the preparation of three randomly selected leaves of English boxwood (Buxus sempervirens ‘Suffruticosa’) after surface sterilization. The leaves were taken from 10-year-old healthy plants in a public garden to study cultural endophyte communities [16]. Cultural endophytes were obtained as described previously. Briefly, the leaves were surface sterilized with 70% ethanol and 10% bleach, then rinsed three times in a large volume of sterilized distilled water (SDW) to remove epiphytes and the associated DNA residues. The sterilized leaves were then cut into small pieces and placed in a presterilized 2 mL microtube containing 1 mL SDW, 0.3 mL 1.4 mm Zirconium beads (MP Biomedicals, Santa Ana, CA, USA), and a 0.35 mm ceramic sphere (OPS Diagnostics, Lebanon, NJ, USA) and homogenized for 3 × 30 s at speed 6 on the FastPrep-24™ Classic bead beating grinder and lysis system (MP Biomedicals, Santa Ana, CA, USA). To grow the microbes, a 0.3-mL aliquot of the resultant suspension was plated on potato dextrose agar (PDA, Sigma-Aldrich, St. Louis, MO, USA) in each 9-cm replicate Petri dish. One of two tan colonies (Additional file 1) named 1021Bp was isolated after 3 days at 23 °C and grown on PDA at 28 °C for 48 h in an incubator (Percival, Boone, IA, USA). To store the culture, its single colonies were grown in Difco™ nutrient broth (NB, Becton, Dickinson, and Company, Sparks, MD, USA) on a shaker (New Brunswick Scientific Inc., Edison, NJ, USA) at 180 rpm overnight, and the culture with a 50% glycerol solution was placed in a -80 °C freezer until use.

1021Bp liquid culture, cell suspension, and cell-free supernatant (CFS)

To test 1021Bp in vitro and in planta activities, 1021Bp was retrieved from storage and grown on PDA plates. For liquid culture, a single colony from the plate was placed in a 5-mL NB, and shaken at 180 rpm at 28 °C overnight to make a seed culture, and 1 mL of the seed culture was transferred into 100 mL NB and incubated for an additional 18 h or until OD600 = 0.6 to 0.8 under the same conditions to obtain 108 to 109 colony-forming units (CFU) per mL (UFC/mL). To separate the cells and CFS in the culture, the liquid culture was centrifuged at 14,210× g for 15 min with a Sorvall® RC 5 C Plus Superspeed Centrifuge (Mashall Scientific, NH, USA). The supernatant that passed through a 0.22-µm Stericup® and Steritop® Vacuum Driven Disposable Bottle Top Filter (EMD Millipore, MA, USA) was used as CFS. The pellet of bacterial cells was resuspended in 0.01% Tween 20 (Sigma-Aldrich, MO, USA).

Analyses of plant growth promotion (PGP) traits

PGP characters of 1021Bp, including nitrogen fixation, phosphate solubilization, siderophore, and indole-3-acetic acid (IAA) production, were analyzed as described previously for the boxwood bacterial endophyte Burkholderia sp. SSG [25]. Briefly, IAA production was measured using the colorimetric method [26] with a minor modification as described previously for SSG [25]. The nitrogen fixation ability of 1021Bp was determined by growing the bacterium on the nitrogen-free agar medium for 4 days as described previously (Liaqat and Eltem, 2016).

The ability of 1021Bp to solubilize phosphate was determined qualitatively and quantitatively with the National Botanical Research Institute’s Phosphate (NBRIP) agar and broth medium [27, 28], with minor modifications as described previously for SSG [25]. Specifically, in the qualitative assay, a 10-µL aliquot of the bacterial seed culture was pipetted onto each of three sterilized Whatman filter paper disks that were placed on NBRIP agar at the points of an equilateral triangle and checked for inducing halo development around the disks after 7 days of incubation at 27 °C. In the quantitative assay, 0.3 mL of an overnight NB culture of 1021Bp was added to 30-mL NBRIP broth containing 150 mg Ca3(PO4)2 as an insoluble form of phosphate. After shaking at 27 °C for 7 days, the mix was centrifuged at 13,416 g for 10 min to obtain the supernatant to quantify soluble phosphate by the bacterium. For measurement, the supernatant was autoclaved for 20 min at 4 °C, and 1 mL of the supernatant or its dilution was added to 2 mL of 2.5% ammonium molybdate and 0.5 mL of 10 mol/L sulfuric acid, mixed with 1 mL of 0.5 mol/L hydrazine hydrate solution, then brought to 25 mL with SDW. An aliquot of the diluted mix was blanked with the control solution without the bacterium and measured for absorbance at 840 nm on a DU800® spectrophotometer (Beckman Coulter, CA, USA).

Siderophore production by 1021Bp was determined by streaking the bacterium on the blue agar medium containing chrome azurol S with an indicator hexadecyltrimethylammonium bromide and observing color change after 48 h at 28 °C [29]. All the assays included three replicate plates and were done twice.

Evaluation of 1021Bp for fungal pathogen growth suppression

Five important fungal plant pathogens collected from Virginia, USA (Table 1) were tested in the dual culture assay on PDA plates, as described previously [18]. A mycelial plug from the edge of one week-old culture of a target fungus was placed in the center of each of three 90-mm PDA plates and an 18 h-old liquid culture of 1021Bp was streaked equidistantly on the sides of the plug. Another set of three plates with fungal plugs was streaked with NB as controls. All the plates were incubated at 25 °C and checked for fungal growth after 5 and 10 days. The diameters of fungal growth in the control plates without the bacterium were compared with those cocultured with the bacterium to calculate the suppression of fungal growth by 1021Bp. Two independent experiments were performed.

Table 1 Test fungal pathogens for growth suppression by 1021Bp in the dual culture assay Calonectria pseudonaviculata (Cps) inoculum preparation

Conidia from the culture of Cps were used for plant inoculation. Production of conidia was done using fresh potato broth, as described previously [10]. Briefly, before inoculation, conidia were washed off with 30 mL of 0.01% Tween 20 from the mycelial mats that were exposed to fluorescent lights for 3 days after growing in the broth for 4 days in the dark at 23 °C and collected in a 50 mL tube. The suspension was vortexed to break conidia clumps and counted for conidia with a hemocytometer under a microscope (EXC-350 series, ACCU-SCOPE®, NY, USA), then diluted with 0.01% Tween 20 to have a concentration of 5 × 104 conidia/mL for inoculation.

Evaluation of 1021Bp for boxwood blight control

The in planta experiment was conducted in the laboratory at 23 °C in a diurnal setting of 16 h dark and 8 h day. Plants of the blight-susceptible boxwood cultivar Buxus sempervirens ‘Justin Brouwers’ at age five years were used. They were grown in 2.8-L containers in the greenhouse before being transferred to the laboratory under the conditions described previously [8, 9]. The resuspended cells at 108 to 109 CFU in 0.01% Tween and the CFS from the same concentration of the cell culture were used for plant pretreatment at 1 and 7 days before inoculation (dbi) with Cps conidial suspensions at 5 × 104. Additionally, the 18-h cell culture, including the cells and CFS of 1021Bp, was used for plant pretreatment at 1, 2, 3, 6, or 11 weeks (wbi) before inoculation with Cps. In these experiments, NB was used as the control of CFS and the cell culture, and 0.01% Tween 20 was used as the control of resuspended cells. Individual experiments were set up for each treatment time, in which three replicate container plants were used for a treatment. After pretreatments, a completely randomized experimental design was used to arrange the container plants in plastic boxes for inoculation with Cps inoculum. Both bacterial pretreatments and inoculation with Cps were done on the plant canopy by hand spray until runoff. To maintain the moisture facilitating bacterial cell hydration or Cps conidia germination and plant infection, treated/inoculated plants were placed in closed large plastic bins within the first 24 h of the pretreatment or inoculation. Each experiment containing three container plants per treatment was conducted twice.

The evaluation of 1021Bp for boxwood blight control was done 7 days after inoculation, when leaf blight symptoms were apparent. The performance was assessed with resuspended bacterial cells and CFS as well as the whole culture without separating cells and CFS. In the first case with resuspended cell and CFS, the total plant leaves and the infected plant leaves, including those that were defoliated, were counted to calculate the leaf infection rate. In the second case with the whole culture, disease severity was estimated using a 1–10 scale representing ascent percentage by 10, as was done in a field experiment [30]: 1 = 1–10%, 2 = 11–20%,. 10 = 91–100% leaves blighted. To be consistent, these disease severity rating data were converted to percentage of leaves blighted at each pretreatment lead time using the formula: (scale level × 10) – 5 before statistical analysis.

1021Bp identification

DNA of 1021Bp cells was extracted using the Quick-DNA™ Fungal/Bacterial Miniprep Kit (Zymo Research, CA, USA) and amplified with the universal primers 27 F and 1410R for the 16 S rRNA gene through PCR as described previously [31]. The amplicon was sequenced at Eton Bioscience (Research Triangle Park, NC, USA), and the processed sequence containing a large part of the 16 S rRNA gene was blasted against the GenBank repository at http://blast.ncbi.nlm.nih.gov and deposited into GenBank with Accession: SUB12930864 1021BPseq OQ565619. Colony morphology was compared with the morphology of the taxon that produced the most significant alignment.

Statistical analyses

All presented data are means of replicates from at least two experiments, expressed as mean ± standard error (SE). Analysis of variance was conducted using the Statistical Analysis Software (SAS Institute, Cary, NC, USA) after log10 transformation of the disease assessment data - % leaves blighted. Treatment means were separated according to the least significant difference (LSD) test at P = 0.05.