Initial Steps towards Biocontrol in Hops: Successful Colonization and Plant Growth Promotion by Four Bacterial Biocontrol Agents

Verticillium wilt, caused by Verticillium nonalfalfae and V. dahliae, is a devastating disease in hops that can cause considerable economic crop losses. The perennial use of hops combined with the long persistence of the pathogen in soil make it difficult to suppress the disease with conventional measures. Biological control agents (BCA) are the basis of an environmentally friendly plant protection strategy that uses plant promotion and antagonistic effects of microorganisms. We evaluated the effect of four selected beneficial bacterial strains, Burkholderia terricola ZR2-12, Pseudomonas poae RE*1-1-14, Serratia plymuthica 3Re4-18, and Stenotrophomonas rhizophila DSM14405 T for their use in hops. All strains were shown to be both rhizosphere and endorhiza competent, and their abundances ranged from log 10 3.0 to log 10 6.2 CFU g −1 root fresh weight in the endorhiza and from log 10 2.9 to log 10 4.7 CFU g −1 root fresh weight in the rhizosphere with B. terricola ZR2-12 showing the highest overall cell densities. Microscopic visualization of DsRed-labeled transformants with confocal laser scanning microscopy showed different colonization patterns and confirmed the rhizosphere competence. Growth promoting effects on seedlings treated with bacteria were found for 584 S. plymuthica 3Re4-18 and S. rhizophila DSM14405 T. Competent colonization and plant growth promoting effects are the most important prerequisites towards efficient biocontrol.


Rhizosphere and Endorhiza Competence by Measuring Bacterial Abundance
The competence to colonize root systems was demonstrated via the re-isolation of rifampicin-resistant mutants of B. terricola ZR2-12, P. poae RE*1-1-14, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T from the rhizosphere and endorhiza of hops. Four weeks after inoculation, bacteria from the rhizosphere and endorhiza of roots were re-isolated on nutrient agar. In general, the rhizosphere was colonized from log 10 2.9 to log 10 4.7 CFU g −1 root fresh weight (RFW) and the bacterial abundance ranged from log 10 3.0 to log 10 6.2 CFU g −1 RFW in the endorhiza ( Figure 1). In this study, higher abundances were assessed for the bacteria in the endorhiza than in the rhizosphere of hop roots. B. terricola ZR2-12 showed the highest cell density (log 10 4.7 ± 0.1 CFU g −1 RFW in the rhizosphere and log 10 6.2 ± 0.4 CFU g −1 RFW in the endorhiza), as the abundances were up to three orders of magnitudes greater than the cell numbers of the other bacteria. B. terricola ZR2-12 was first selected according to its high rhizosphere competence in sugar beets roots (up to log 10 10 CFU g −1 RFW) [31], which can now also be confirmed for hops. In this study, the density of S. plymuthica 3Re4-18 in the hop root endorhiza was log 10 3.0 ± 0.2 CFU g −1 RFW, and the colonization of the rhizosphere was log 10 2.9 ± 0.1 CFU g −1 RFW. In P. poae RE*1-1-14 treated plants, the bacterial colonization was log 10 3.0 ± 0.2 CFU g −1 RFW in the rhizosphere and log 10 4.3 ± 0.2 CFU g −1 RFW in the endorhiza. Furthermore, S. rhizophila DSM14405 T showed similar abundances of approximately log 10 3.8 ± 0.2 CFU g −1 RFW in the rhizosphere and log 10 4.1 ± 0.2 CFU g −1 RFW in the endorhiza. In other studies, similar bacterial densities in the rhizosphere of different crops were also found. The strain S. plymuthica 3Re4-18 is described as an effective rhizosphere colonizer [30], and had an average abundance of log 10 3.6 to log 10 4.2 CFU g −1 RFW in sugar beet depending on the root section [30]. Similarly, the strain S. plymuthica HRO-C48 reached population densities of log 10 3.5 ± 1.4 CFU g −1 RFW in oilseed rape [34]. Lower densities of S. plymuthica 3Re4-18 were shown in the root system of hops. The abundance of P. poae RE*1-1-14 in the rhizosphere of sugar beets ranged from log 10 3.7 to log 10 3.9 CFU g −1 RFW depending on root section [30], and showed superior colonization behavior in the hop rhizosphere. A large variety of life styles and interaction strategies are known for different endophytic bacteria and also rhizosphere bacteria can colonize the endorhiza [35,36]. The endophytic life style of all applied strains emphasizes their intimate plant-microbe interaction and suggests a positive role in hops. After outdoor hibernation, the plants sprouted under greenhouse conditions. Random samples were analyzed to see if the bacteria could survive at sub-zero temperatures. Again, B. terricola ZR2-12 was the best colonizer with a similar colony density as the colonization experiments above. The other three bacteria were found in the endorhiza with approximately log 10 3.0 CFU g −1 RFW. These results are important for further biological control measures and practical applications as they will help establish the bacteria in the hops roots.
Furthermore, the potential establishments of these bacteria on seeds were analyzed. Hops seeds were inoculated and the root system cell density was determined after seven weeks. The abundances of B. terricola ZR2-12, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T ranged from approximately log 10 3.1 to log 10 4.1 CFU g −1 RFW. However, P. poae RE*1-1-14 showed a very low population density.

Colonization Patterns Observed by Microscopic Monitoring
Microscopic analyses were prepared using Confocal Laser Scanning Microscopy (CLSM) to visualize the colonizing behavior in the root system. Exemplarily, the colonization patterns for B. terricola ZR2-12, P. poae RE*1-1-14, and S. plymuthica 3Re4-18 were monitored six to seven days after inoculation. A high density of B. terricola ZR2-12 cells was found in the rhizosphere (Figure 2a

Effect of Bacterial Treatment on Plant Growth
Seeds and plants were dipped in bacterial suspensions to assess the effect of B. terricola ZR2-12, P. poae RE*1-1-14, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T treatments on plant growth. Seven weeks after inoculation of the seeds, the length of the plantlets without the roots and the length of the fourth leaves were measured. Additionally, to evaluate an additive effect, seeds were treated with a mixture of S. plymuthica 3Re4-18 and S. rhizophila DSM14405 T (1:1). S. plymuthica 3Re4-18 showed significant PGP effects in both experimental sets (Figure 3), and S. rhizophila DSM14405 T and the combined strains promote plant growth as shown from the fourth leaves (Figure 3a). However, no significant PGP effects could be observed for plant length (Figure 3b), even though both strains already showed plant growth promotion in other crops [9,23,29,30]. In addition, the growth effect on taller cuttings was assessed. The weight gain four weeks after treatment was measured, but no differences in PGP could be found.

In vitro Antagonistic Activity against V. nonalfalfae and V. dahliae
The antagonistic activity of the four bacteria against V. nonalfalfae and V. dahliae was assessed using a dual culture test. The strain S. plymuthica 3Re4-18 showed an inhibition zone which has been confirmed with other studies for V. dahliae [25,30]. P. poae RE*1-1-14 did not have any antagonistic activity against these two Verticillium species in this or other studies [30]. An inhibition by S. rhizophila DSM14405 T and B. terricola ZR2-12 could also not be found, although it was described in other studies for V. dahliae [21,24].

Hop Cultivars
The cultivar "Hallertauer Tradition" was used in the experiments due to its wide spread cultivation and its increased susceptibility in fields towards current pathotypes of V. nonalfalfae. To assess the PGP effect, seeds of the cross breed of Cascade and 2007/005/504 were treated.

Microorganisms
Four strains of bacteria, previously isolated from diverse microhabitats and crops, were used in this study. For greenhouse-experiments, spontaneous mutated isolates of B. terricola ZR2-12, P. poae RE*1-1-14, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T resistant to rifampicin (100 µg mL −1 ; Roth, Karlsruhe, Germany) were used. No differences in growth parameters (colony morphology and growth rate) and traits (antifungal properties towards V. nonalfalfae and V. dahliae, proteolytic activity, Box PCR pattern) were found between the mutant and wild type. The strains were stored in nutrient broth (10 g of peptone, 5 g of yeast extract, Roth and 5 g of NaCl, Merck, Darmstadt, Germany in 1 L of distilled water, pH 7) containing 12.5% glycerol at −80 °C. The wildtype strains were maintained in the Strain Collection of Antagonistic Microorganisms (SCAM) at Graz University of Technology in LB medium containing 15% glycerol at −70 °C. The used V. nonalfalfae and V. dahliae strains for dual culture tests were isolated from infected hop bines. The fungi were maintained as monospore cultures on prune agar at 20 °C [37], and the species identity was verified by specific primers [38,39].

Determination of Root Colonization
Bacterial overnight culture (12 mL) in nutrient broth (100 mL containing 100 µg mL −1 of rifampicin, 28 °C, 120 rpm) was transferred to 500 mL of a new culture (28 °C, 120 rpm) and diluted with 0.85% NaCl solution to a final cell concentration of 10 8 CFU mL −1 . Roots of the cultivar "Hallertauer Tradition" were dipped in the bacterial suspension for 15 min and planted in unsterilized soil (Lorenzer potting soil, Einheitserde special, Einheitserdewerke Patzer, Sinntal-Jossa, Germany). Control plants were dipped in 0.85% NaCl solution. The experiment was done in twelve replicates for each bacterium as well as for the negative control and repeated three times under greenhouse conditions with minimum of 13 h light (artificial light between 6 am and 7 pm, if the daylight is under 40 kLux) with a minimum temperature of 22 °C at day and 19 °C at night. After four weeks, twelve plants were divided into four parts (containing three plants) for each BCA. 2.5 g of roots (soil adhering to roots) were incubated in 15 mL of 0.85% NaCl solution for 20 min and at 300 rpm to determine bacterial density in the rhizosphere. To define the colonization number in the endorhiza, 2 g of roots were cleaned, surface sterilized with sodium hypochlorite (3% active chlorine, 5 min) and washed three times with sterile water. For sterilization control, roots were dipped on a nutrient agar (nutrient broth added 15 g L −1 agar agar, Roth) containing rifampicin (100 µg mL −1 ; Roth) and nystatin (20 µg mL −1 ; Roth). The roots were then crushed with 5 mL 0.85% NaCl solution by mortar and pestle. The resulting suspensions of the rhizosphere and endorhiza were serial diluted, and 100 µL were plated on selective nutrient agar as described above. After seven days of incubation (28 °C, in dark), the number of colonies (CFU g −1 fresh root weight) was determined. After the outdoor hibernation in sub-zero temperatures, six random samples of the roots, which were divided into two parts, were taken and re-isolated as described above.

Plant Growth Promotion (PGP) in the Greenhouse
The weight of the plants before and four weeks after the bacterial treatment was measured to determine the PGP. The same dipping procedure for the determination of cell densities was used. In addition, seeds of the cross breed of Cascade and 2007/005/504 were dipped for 15 min in bacterial suspension (10 8 CFU mL −1 ) and the control seeds were treated with sterile 0.85% NaCl solution. The seeds were planted in potting compost (Lorenzer potting soil) and grew for seven weeks. The growth was determined by the size of the plantlets without the roots and the fourth leaves (including the cotyledons). The roots were assembled and separated into two parts of 1.5 g to 2 g to control successful root colonization. The roots were then ground with 15 mL 0.85% NaCl with mortar and pestle and 100 µL plated on nutrient agar (nutrient broth added 15 g L −1 agar agar, Roth) containing rifampicin (100 µg mL −1 , Roth) and nystatin (20 µg mL −1 ; Roth). After seven days, the colonization was determined. The growth experiments were done in twenty replicates and independently repeated two times.

Screening for Antagonistic Activity
The in vitro antagonism was determined in a dual culture assay. 200 µL of seven day old mycelia suspension V. nonalfalfae isolated from infected hops was plated onto PDA (potato extract glucose agar; Roth) and Waksman agar containing 5 g tryptone/peptone ex casein (Roth), 10 g glucose (Roth), 3 g yeast extract (Roth), 20 g agar (Roth) and filled up to 1 L with distilled water. After 30 min, the bacteria were spotted on the plate. Antagonistic activity and the inhibition zone were assessed after seven days of incubation at 20 °C.

Statistical Analysis
Root colonization data was log 10 transformed before statistical analysis, and the package SPSS (SPSS Inc., Chicago, IL, USA) was used for statistical data analysis. For the determination of PGP, the significance towards the control plants was analyzed using the Scheffé's test and Tukey's test (P ≤ 0.05). The outlier test DIXON was used in the PGP experiments and the outliers were excluded from the statistical analysis. The deviation was indicated as standard error.

Conclusions
This study investigates the preliminary requirements of four bacteria, B. terricola ZR2-12, P. poae RE*1-1-14, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T to act as BCAs to suppress Verticillium wilt in hops. Many previous studies have already demonstrated the ability of the four bacteria for biological control in other crops. Regarding their rhizosphere competence and PGP effect, S. plymuthica 3Re4-18 and S. rhizophila DSM14405 T are promising candidates for BCA on hops, as well as B. terricola ZR2-12 that showed exceptionally high cell densities. Due to this rhizosphere competence even after hibernation in sub-zero temperatures, this strain can be a suitable BCA in hops. The bacterial treatment of seeds with S. plymuthica 3Re4-18 and S. rhizophila DSM14405 T was also shown to benefit plant growth. Antagonistic activity against V. dahliae is known for B. terricola ZR2-12, S. plymuthica 3Re4-18, and S. rhizophila DSM14405 T and an antagonistic effect has also been shown via dual culture test in this study for S. plymuthica 3Re4-18 against V. nonalfalfae and V. dahliae. Furthermore, the beneficial traits of S. plymuthica 3Re4-18 are well-known and contribute to its biotechnological potential in hops. Nevertheless, to objectively assess the ability of these beneficial bacteria strains towards V. nonalfalfae in hops, artificial infection tests and further experiments under field condition are necessary, also to assess the effects on crop yield. Furthermore, the consistent efficacy of these beneficial bacteria must be verified in the field.