1 Introduction

Sugar beet is the most important sugar crop in the developing world. Fertilizers and pesticides are the most important inputs for beet production but using growth-promoting bacteria can contribute to making beet cultivation sustainable and less dependent on fertilizers. Among such bacteria are rhizobacteria, which colonize plant roots and consume root exudates and lysates (Pieterse et al., 2002; Antoun and Prevost, 2006). Some strains, the growth-promoting rhizobacteria (PGPR), can promote plant growth. They improve the cycling of nutrients and minerals such as nitrogen, phosphate and can be used as biofertilizers (Kennedy et al., 2004). The growth-promoting effects of PGPR are affected by environmental conditions and therefore are not stable. Endophytes including a variety of bacteria, fungi, and actinomycetes and can be isolated from cultivated crops (Liu and Tang 1996) as well as from wild plants (Brooks et al. 1994) both monocotyledons (Fisher et al. 1992) and dicotyledons (El-Shanshoury et al. 1996).

Healthy roots of sugar beet have been shown to harbour endophytic bacteria and healthy leaves to harbour endophytic fungi (Jacobs et al. 1985; Larran et al. 2004), little is known about the behaviour of these microorganisms during the growing season. Endophytes occupy microniches within plant tissues and some have been found to be growth-promoting endophytes (PGPE). These endophytes are unaffected by competition from other microorganisms or conditions in the rhizosphere. As a result, a new approach has been developed in recent years to reduce fertilizer use which consists of treating seeds and seedlings with PGPE. Thus, inoculation of pepper with endophytic Pseudomonas fluorescens significantly increased fresh weight, height, and stem diameter (Lucas-Garca et al. 2003). In an previous study, we isolated 221 endophytic bacterial isolates from tissues sterilized sugar beets from a widely planted beet cultivar (Beta vulgaris L. var. saccharifera) variety KWS2409 in a semi-arid region in east Urumuqi in Xinjiang,China. In the present study, we assess the effect of Acinetobacter johnsonii strain 3–1 on the growth of beet seedlings and on the ability of their root system to absorb mineral nutrients.

2 Materials and methods

2.1 Sugar beet variety, bacterial strain, and soil

The soil for the pot experiments was collected from a fertile field in northern Xinjiang. Chemical and physical properties of soil were analysed (Table 1). Total carbon, C, was determined by muffle furnace heating the soil to 500°C; total nitrogen, N, by the Kjeldahl method; and total phosphorus, P, by molybdenum blue colorimetry (Riehm 1985). Calcium, extractable magnesium, potassium, and sodium were measured using an atomic absorption spectrophotometer (Schachtschnabel and Heinemann 1974); pH was determined with a pH meter; and total salt content was assessed by measuring the electrical conductivity of soil.

Table 1 Chemical and physical properties of soila in the top (0–25 cm) layer
Full size table

Sugar beet (Beta vulgaris L. var. saccharifera) variety KWS2409 and Acinetobacter johnsonii strain 3–1 (GenBank accession number EU594557) were used, with Chryseobacterium indologenes strain 2–2 (GenBank accession number EU594563) as control for 3–1, in the study assessing the effect of bacteria on plant growth. Sugar beet seeds (Beta vulgaris L. var. saccharifera) variety KWS2409 were obtained from Toutunhe farm in a semi-arid region in east Urumuqi in Xinjiang,China.

2.2 Plant growth and inoculation in pots

Both strains 3–1 and 2–2 were grown in a liquid medium (3 g beef extract, 5 g peptone, 5 g NaCl, and 1 g glucose, in 1 L water, pH 7.0–7.2) shaken at 200 rpm at 30°C for 48 h, after which the bacterial cells were precipitated by centrifugation at 5,000×g for 5 min. The pellets were suspended in sterile distilled water and the concentration of the suspension adjusted to 108 cfu ml−1 with a spectrophotometer (Mayak et al. 2004). Sugar beet seeds were surface-sterilized with 70% alcohol for 8 min followed by 5% hydrogen peroxide solution for 2 h and then rinsed in sterile water (Caitriona et al. 2004). The disinfected seeds were immersed in the bacterial suspension for 6 h and then sown in pots, 10 seeds per pot. After germination, the stand was thinned to keep six plants in each pot. Ten replicates were maintained, with beet seeds immersed in distilled water instead of the bacterial suspension serving as control (CK). The pots were kept in a growth chamber with the day/night temperatures set at 30°C/25°C and 200 μmol photons m−2 s−1 of light supplied throughout the 12 h day. Soil moisture was maintained at 60% of the soil’s moisture-holding capacity. Germination percentage was recorded 7 days later and data on plant height, fresh weight and dry weight, and the contents of N, P, K, and Mg and of Vitamin B,Vitamin C and protein recorded 4 weeks after germination.

2.3 Re-isolation of endophytic bacteria

Endophytic bacteria from the seedling plant tissues were re-isolated 4 weeks after germination. To do this, fresh samples of sugar beet plants were washed thoroughly with tap water to remove adhering soil and debris, rinsed in 95% ethanol, and then flamed. This minimised carry-over of external bacteria when cross-sectioning at various locations using a sterile knife. Samples were aseptically removed from the core, periphery, and secondary root emergence zone (crease) areas. Each sample of root tissue was aseptically weighed, then added to 100 mL of sterile saline (0.85%), and blended for 2 min in a Waring blender. The blended samples were initially diluted to standardise all preparations. This was followed by additional serial decimal dilution in sterile saline. Endophytic bacteria were isolated using an aerobic spread-plate method. Dilution volumes of 0.1 mL were plated in triplicate on nutrient agar (NA) supplemented with 2.0 g/L sucrose. The spread plates were incubated at room temperature for 7 d and colony counts recorded by standard methods (Kodaka et al. 2005). Pure cultures were subsequently isolated and bacterial isolates were stored on agar slants.

2.4 Identification of endophytic bacteria

Pure cultures of the bacterial isolates were prepared for identification purposes. The bacterial re-isolates 3–1 and 2–2 were identified using the Vitek AutoMicrobic System (Vitek AMS; Vitek Systems, Inc., Hazelwood, MO.). The Vitek test was repeated twice. Isolates were characterized by Vitek AMS, colony morphology, catalase production, oxidase test, and gram stain.

2.5 Phylogenetic analysis of the isolated bacteral strains

The 16S rRNA gene of genomic DNA isolated was extracted from pure bacterial colonies following standard protocols. The primers used for PCR were pF (5 -AGA GTT TGATCC TGG CTC AG-3 ) and pR (5 -AAG GAG GTG ATC CAG CCG CA-3).The PCR products with the expected size (about 1,500 bp) were purified using a DNA Gel Extraction Kit and cloned into pMD18-T vector followed by sequencing. Sequence analysis was performed using the BLAST algorithm (http://www.ncbi.nlm.nih.gov). Bacterial identifications were based on 16S rRNA gene sequence similarity. For further characterization of isolates 3–1 and 2–2, a neighbor-joining phylogenetic tree was constructed with the MEGA 4.0 program.

2.6 Measuring the ability of strain 3–1 to solubilize phosphates

Strains 3–1 and 2–2 were grown in PKO (Ca3(PO4)2) medium Pikovaskaia’s medium, Yao 2004, per liter: Ca3(PO4)2 3.0 ~ 5.0 g, sucrose 10.0 g, (NH4)2SO40.5 g, NaCl 0.2 g, MgSO4·7H2O 0.1 g, KCl 0.2 g, yeast extract 0.5 g, MnSO41 mL (0.004 g/L), FeSO4 (Fe-EDTA) 0.1 mL (0.002 g/L), agar 15 g, pH value 7.0 ± 0.2 () and incubated at 30°C for 12 h with vigorous shaking. The concentration of the bacterial suspension was adjusted to 108 cfu ml−1 and the suspension used as the primary inoculum. Two millilitres of the primary inoculum were added to 200 ml PKO medium whereas 2 ml distilled water added to 200 ml PKO medium served as control. Each treatment was repeated three times. The phosphates-solubilizing capability of Strains 3–1 and 2–2 were measured every 24 h for 7 days. Each time, a sample of the culture medium was centrifuged at 10,000 rpm for 20 min, and pH of the supernatant determined with a pH meter. Dissolved P (H2PO −4 ) was determined using molybdenum blue colorimetry (Zhao et al. 2003; Zhang 2008). The pellet was suspended in 0.1 mol·L−1 HCl, rinsed three times in sterilized water, and optical density (OD660nm) of the suspension was measured.

2.7 Measuring the ability of strain 3–1 to secrete IAA

The IAA assay was performed using the method described by Patten and Glick (2002) with some modifications. Both the strains, namely strains 3–1 and 2–2, were multiplied overnight in 5 ml of DF salts minimal medium (Dworkin and Foster 1958) which had the following composition (for 1 L): 4.0 g KH2PO4, 6.0 g Na2HPO4, 0.2 g MgSO4·7H2O, 2.0 g glucose, 2.0 g gluconic acid, and 2.0 g citric acid; trace elements: 1 mg FeSO4·7H2O, 10 mg H3BO3, 11.1 mg MnSO4·H2O, 124.6 mg ZnSO4·7H2O, 78.22 mg CuSO4·5H2O, and 10 mg MoO3; pH 7.2; and 2.0 g (NH4)2SO4 as the nitrogen source. The next day, 20 μL of the primary inoculum was transferred into 5 ml of the DF medium supplemented with 100 μg ml−1 L-tryptophan (Sigma Chemicals) taken from a filter-sterilized 2 mg/ml stock prepared in warm water. After incubation for 48, 72, 96, and 120 h, the density of each culture was measured spectrophotometrically at 600 nm. The bacterial cells were then removed from the culture medium by centrifuging at 6,000×g for 15 min; a 1 ml aliquot of the supernatant was mixed vigorously with 4 ml of Salkowski’s reagent (Gordon and Weber 1951) consisting of 150 ml of concentrated H2SO4, 250 ml of distilled H2O, and 7.5 ml of 0.5 M FeCl3·6H2O, and allowed to stand at room temperature for 20 min before the absorbance at 535 nm was measured. The concentration of IAA in each culture medium was determined by comparing with a standard curve.

2.8 Estimation of internal root colonization

Pot trials were designed to assess the internal colonization of sugar beet roots after seedling inoculation through the seed soaking method described above. Rifampicin-resistant mutants of isolates 3–1 and 2–2 were prepared as described by He et al. (2004). The obtained mutants were compared with their wild types in relation to their ability to produce auxins. None of these mutants differed morphologically from their parental strains, and all mutants had identical growth rates and auxin-production capability with their parental strains. Sugar beet seedlings were prepared as described above and inoculated with isolates 3–1 and 2–2. Every week after planting (1–8), roots were sampled from the soil, washed thoroughly in tap water, surface disinfested as described above and the population densities of isolates 3–1 and 2–2 (log10 CFU g − 1 fresh root weight) were determined using NA amended with rifampicin (300 μgml−1 , Sigma). Each treatment was replicated five times with two plants in each replicate for each sampling.

2.9 Field experiments to evaluate performance of strain 3–1

Field experiments were conducted at the Toutunhe Farm in Urumuqi, Xinjiang province, China, from 2009 to 2010. Strain 3–1 inoculation was made the same way as pot experiment, inoculated by using the root drench method. Beets were harvested shortly after the 7th week of testing, at which time the plants were about 15 cm at the crown. Beets were fertilized, irrigated, and weeds controlled following typical production practices in northwest, China. The sucrose and fructose assay was performed using a spectrophotometric method (Taylor, 1995). Yield comparisons among treatments were made at harvest date, and sugar content was determined by refractive method.

2.10 Statistical analysis

All experiments were arranged in completely randomised block designs. Population data were transformed into log10 CFU g−1 fresh weight. Data were subjected to analysis of variance, and treatment means were compared using Fisher’s Protected LSD Test at P < 0.05.

3 Results

3.1 Effect of strain 3–1 on plant growth

Results of seed inoculation are shown in Figs. 1 and 2. Germination percentage was significantly higher in seeds inoculated with Acinetobacter johnsonii strain 3–1 (3–1-W) and so were the two other growth parameters, namely dry weight and plant height. Germination rate of inoculated seeds was 92%, 28.6% (in the text just give percentages to the nearest whole percentage i.e. 92% and 29%) greater than that of seeds that had been soaked in water (CK-W). Dry weight of seedlings from the inoculated seeds was about 3-fold greater, and the seedlings were 18.3% taller, than those from seeds that had been soaked only in water before sowing. Plants from seeds inoculated with strain 3–1 grew significantly better than those growing from seeds that had been soaked in water (W) or in Hoagland solution (N). Plants from seeds inoculated with strain 3–1 were 19.4% taller and 69.8% heavier than those from uninoculated seeds. In comparison, the 2–2 strain did increase the dry weight of beets, and it had little effect on plant height, however.

Fig. 1
figure 1

Effect of Strain 3–1 on growth of sugar beet. From left CK-W, seeds soaked in water; 3-1-N, seeds soaked in a suspension of Strain 3–1 in nutrient solution; CK-N, seeds soaked in nutrient solution

Full size image
Fig. 2
figure 2

Effect of seed inoculation on growth of beet seedlings. CK-W, seeds soaked in water; CK-N, seeds soaked in Hoagland solution; 3-1-W, seeds soaked in a suspension of Strain 3–1 cells in water; 3-1-N, seeds soaked in a suspension of Strain 3–1 cells in Hoagland solution; 2-2-W, seeds soaked in a suspension of Strain 2–2 cells in water; 2-2-N, seeds soaked in a suspension of Strain 2–2 cells in Hoagland solution. The results are representative of 48 plants. Error bars show the standard deviation

Full size image

3.2 Effect of strain 3–1 on plant nutrient uptake

The content of N, P, K, and Mg and of vitamin B, vitamin C, and protein was determined 4 weeks after germination (Table 2). Inoculation with Strain 3–1 significantly increased the uptake of all the elements that were analysed: total N and P following inoculation with strain 3–1 were greater by 10.8% and 26.4% respectively than their levels in CK-W, and the levels of K and Mg were also higher. It was also found that salt, that is soaking the seeds in a nutrient solution, affected the uptake of nutrients from soil: levels of N and P in CK-W were lower by 2.1% and 6.2% (In the test just give the % to the nearest whole percentage, i.e. 2% and 6%) respectively than the levels in CK-N. Inoculation with strain 3–1, however, counteracted this effect, levels of K+ and Mg2+ in strain 3–1-N being 7.4% and 23.6% higher respectively than those in CK-N and levels of total Vitamin B,Vitamin C, and protein being 13.1%, 10.4%, and 35.2% higher respectively in strain 3–1-N than those in CK-N. Strain 2–2 also influenced the ability of sugar beet to accumulate ions, although not to the same extent as strain 3–1: levels of N, P, and K+ in 2–2-W were 1.9%, 11.9%, and 0.6% higher respectively than those in CK-W.

Table 2 Effect of inoculation with Strain 3–1 on the content (mg/kg fresh weight) of N, P, K, and Mg and of Vb2,Vc, and protein in beet
Full size table

3.3 Re-isolation and identification of endophytic bacteria

Endophytic bacteria were re-isolated from in vitro inoculated beet tissues. The strain 3-1-1 strain had the phenotypic properties of Acinetobacter johnsonii strain 3–1. Thus the cells were Gram-negative rods measuring 0.5–1.0 × 0.9–5.8 μm. They were motile at a growth temperature of 4–41°C. The cytochrome oxidase test was positive. Nitrate reduction was negative. Phenotypic data of morphological, physiological, and biochemical characteristics combined with 16S rRNA gene (GenBank accession number EU594563) analysis (Fig. 3) allowed the identification of the strain 3-1-1 as Acinetobacter johnsonii. Strain 3-1-1 presented high similarity values with Acinetobacter johnsonii (100%) in the BLASTn search, and were recovered in a tight cluster with Acinetobacter johnsonii in the phylogenetic tree (Fig. 3).

Fig. 3
figure 3

Phylogenetic analysis of partial 16S rDNA sequence of isolates Acinetobacter johnsonii strain 3–1 obtained from sugar beet. Numbers in parentheses represent the sequencesaccession number in GenBank. The number at each branch points is the percentage supported by bootstrap

Full size image

3.4 Effect of strain 3–1 on the level of solubilized phosphates

The relationship between plant growth, measured by the number of days after germination, and (a) the levels of dissolved P in the culture medium and (b) the pH of the culture medium is shown in Figs. 4 and 5 respectively. Without strain 3–1, the mean P content of the culture medium was very low and the pH remained at 8.0. However, with strain 3–1, the levels of dissolved P increased (reaching 180.5 mg L−1 on the sixth day) and the pH decreased to 5.5. However, in the first 144 h, as the strain multiplied, the level of dissolved P decreased significantly from 180.5 mg L−1 to 150.0 mg L−1, a decrease that could be attributed to the use of P by the bacteria for their growth.

Fig. 4
figure 4

Relationship between growth of Strain 3–1 and level of dissolved phosphorus. The results are representative of three experiments, with distilled water added to PKO medium as control. Error bars indicate standard deviation

Full size image
Fig. 5
figure 5

Relationship between dissolved phosphorus (in the culture medium inoculated with Strain 3–1) and pH of the culture medium. The results are representative of three experiments. Error bars indicate standard deviation

Full size image

3.5 Secretion of IAA by strain 3–1

When cultured in a medium supplemented with 100 μg ml−1 tryptophan for approximately 48–168 h, strain 3–1 produced roughly up to twice as much IAA as that produced by strain 2–2 (Table 3). The rate of IAA production increased with time up to Day 6 (up to 144 h) but slowed down on Day 7 (at the end of 168 h).

Table 3 Production of IAA by Strain 3–1 in a medium supplemented with tryptophan
Full size table

3.6 Estimation of internal root colonisation

The rifampicin-resistant mutants of isolates 3–1 were isolated from the in vitro inoculated beet tissues, indicating that the strain was an endophytic bacteria. Isolate 3–1 maintained their endophytic colonising abilities and were isolated from healthy beet roots in all samplings until week 8 (Table 4). An initial increase in colonies after 1, 2 and 3 weeks was followed by a decrease at the 4th week. However, at and after week 5 and up to week 8, the populations were relatively stable (Table 4).

Table 4 Average quantities of bacterial isolate 3–1 at different times of sampling of beet roots grown under glasshouse conditions
Full size table

3.7 Impact of strain 3–1 on yield

In field experiments, the content of sucrose, fructose, and the biomass were determined at harvest. Sugar beets were harvested by hand in October 2006. Inoculation with Acinetobacter johnsonii strain 3–1 significantly increased the content of sugar that were analysed: the content of sucrose and fructose following inoculation with strain 3–1 were greater by 9.0% and 41.2% respectively than their levels in CK. Inoculation with strain 3–1, levels of biomass and sugar content being 16.65% and 23.81% higher respectively than those in CK (Table 5).

Table 5 Effect of indophytic bateria inoculation on sugar increasing of sugar beet
Full size table

4 Discussion

Endophytes are known to promote the growth of plants (Compant et al. 2005a, b), primarily by increasing the supply of nutrients to the host plant. The results of the present study support this as inoculation of sugar beet seeds with Acinetobacter johnsonii strain 3–1 increased the levels of N, P, K+, Mg, VB2,Vc and the protein content of the plants (Table 2). The growth may also be enhanced by the ability of strain 3–1 to solubilize phosphate. Phosphorus is very important for normal plant growth and development. PGPR, which have the ability to solubilize P, increased the phosphate content and plant yield of maize and lettuce (Chabot et al. 1996; Yao 2004). Our study confirms this and revealed that strain 3–1 was alabe to solubilize Ca3(PO4)2 a major type of phosphorus. Further studies are needed using other forms of phosphorus that are difficult to solubilize.

Bacteria that can produce IAA and siderophores that can stimulate plant growth and help plants to acquire sufficient iron and phosphate for optimal growth (Glick et al. 1995; Chabot et al. 1996; Rajkumar et al. 2006). For example, Trianthema aestivum secretes IAA, which enhances root development and improves the uptake of mineral nutrients (Ali et al. 2009). Bacteria of the genus Methylobacterium can benefit plants by promoting nitrogen metabolism, producing vitamin B12 (Basile et al. 1985; Holland and Polacco 1992, 1994; Holland 1997). Indeed, many beneficial effects on plant growth have been attributed to endophytes including osmotic adjustment, stomatal regulation, modification of root morphology, enhanced mineral uptake of minerals, and altered nitrogen accumulation and metabolism (Compant et al. 2005a, b). However a great deal remains to be done to understand the symbiosis between endophytes and their host plants. Only when this research has been completed will we be able to use endophytes optimally in agriculture, forest regeneration or phytoremediation of contaminated soils.