Effects of inoculation of plant growth promoting rhizobacteria to minimize panicle grain shattering habit for increased yield of rice (Oryza sativa L.)

Locally isolated Bacillus subtilis (UPMB10) as nitrogen fixer and B. pumillus (GM118) as phosphate solubilizer act as plant growth promoting rhizobacteria (PGPR). The objective of this study is to determine the response of single and co-inoculation of these PGPR on the panicle breaking strength to minimize panicle grain shattering to increase rice yield. The ability to fix nitrogen and solubilize phosphate, production of indole-3-acetic acid (IAA) and grain yield production of single and coinoculants of PGPR were studied. The result demonstrated that single and combined inoculations were able to fix nitrogen and solubilize phosphate. Besides, the ability of UPMB10 and mix inoculation to produce high concentration of IAA enables them to provide a high breaking force needed to detach grain from panicle compared to GM118 and in turn increase the rice grain yield. Hence, this study showed the potential single and combined inoculations as a biofertilizer to increase rice productivity in the granary areas of Malaysia.


INTRODUCTION
Plant growth promoting rhizobacteria (PGPR) is a free living soil born bacteria that colonize the rhizosphere.The interaction between plant root exudates and bacteria to support each other in term of symbiosis interaction for growth and development also enhanced the plant growth when applied to the seed crops (Yu et al., 2012).Besides, it can increase the nutrient status of host plants through the mechanism of biological N 2 fixation, phosphate (P) and potassium (K) solubilization.
When single PGPR inoculation of rhizobacteria were applied, they showed potential ability to enhance shoot growth, root density and yield of rice.However, recent studies showed that co-inoculation of PGPR are better than single inoculation as they provide more balanced nitrogen, phosphorus, and mineral nutrients (Mohammadi and Sohrabi, 2012).These improvements in growth attributes of plants caused by PGPR are due to the potential of fixing and solubilizing mineral fertilizer and nutrition for the plant and improved the absorption of production of phytohormones resulting in the increased availability of nutrients to plants and root permeability (Glick, 2012).Phytohormones produced by PGPR such as IAA is crucial as they are capable of increasing shoot growth, root hair density and length, enhance rice seed germination and improve growth (Saharan and Nehra, 2011).
Normally in wild species of rice, grain shattering habit is an adaptive trait for seed dispersal at maturity to prevent seeds from being eaten by predator and guaranteed the continuous propagation of rice seed (Lin et al., 2007).However, easy shattering causes considerable yield loss in cultivated grain rice.This opinion was supported by Sadeghi et al. (2010), which stated that the threshing force had significant difference in different rice varieties.Breaking tensile strength force is the force required to detach a grain from the panicle which describes the grain shattering habit, assumed to be one of the effective lossevaluating parameters during harvesting (Alizadeh and Allameh, 2011).It is an important criterion for evaluating rice variety in the interactive effect with PGPR inoculation programs.Hence, one way to combat this problem is by promoting the use plant growth promoting rhizobacteria (PGPR).
This approach of using PGPR to reduce grain shattering should be conducted in Malaysia to determine the effectiveness of locally isolated PGPR toward the increment of grain quality of local rice varieties and in turn increase the national rice yield production.Here, we report the characteristic and locally isolated effects of single and co-inoculation of PGPR namely Bacillus subtilis (UPMB10) and Bacillus pumillus (GM118) on the panicle breaking strength to minimize panicle grain shattering under controlled and field conditions.

Source of inoculation, culture conditions and treatments
The two PGPR strains, B. subtilis (UPMB10) and B. pumillus (GM118) were obtained from the culture collection of the Department of Agriculture Technology Faculty of Agriculture, Universiti Putra Malaysia, Serdang.B. subtilis was initially isolated from oil palm rhizosphere and B. pumillus from a paddy soil.Bacterial single colonies from stock were prepared by touching and streaking on nutrient agar plates in a third streak pattern.The plates were then incubated for 24-48 h at 30ºC.A single colony was inoculated into 700 ml nutrient broth in an Erlenmeyer flask.Inoculated flasks were shaken (200 rpm) in a rotary shaker for 10 h at 30ºC.The final concentration of the bacterial cultures was adjusted to about 10 8 cfu/ml (Kausar et al., 2011).Four treatments were conducted for the rice plant experiment which:

Phytohormones production
Phytohormones production or Indole-3-acetic acid (IAA) was determined by colorimetric method (Gordon and Weber, 1951).Fully grown bacterial culture were incubated in 100 mL tryptic soy broth (TSB) and shaken for 24 h.One mL of the bacteria culture was transferred into new 100 mL TSB with addition of 5 mL Ltryptophan as the pre-cursor of Indole-3-acetic Acid.TSB without bacterial inoculation served as control.1.5 mL of the bacterial culture were transferred into sterile Eppendorf tube and centrifuged at 7000 rpm for 7 min.One mL of the supernatant was mixed with two mL of Salkowsky reagent (2% of 0.5 M FeCl3 in 35% prechloric acid) and allowed to settle for 25 min for development of pink color which indicates IAA production.The absorbance values of each isolate was determined by using spectrophotometer at 535 nm and compared using the standard curve.The IAA standard curved was prepared using pure IAA stock as 0,5,10,15,20,25,30,35,50 and 45 µg/mL of IAA (Y=0.0186x;R 2 = 0.9802).Supernatants of uninoculated test tubes were used as control, where no visible color was observed.

Evaluation of PGPR application on rice
The experiment was conducted in a glasshouse, at Universiti Putra Malaysia and field study in IADA Kemasin Semerak, Kelantan, Malaysia to evaluate the effects of UPMB10 and M118 on rice grain yield MR263.For glasshouse experiment, seeds were sown in trays of 3.2 × 3.2 × 4.5 cm cell size in a medium of peat and sandy (1:1) soil.When the seedling reached 5 cm height (after two weeks), they were transplanted into buckets (4 plants per pot) 40 cm diameter (20 kg soil pot -1 ) with four replications.The soil used in this study was collected from Kemasin Semerak, Kelantan Malaysia.The used nitrogen fertilizer was urea (46% N), potassium fertilizer was muriate of potash (60% K2O) and triple super phosphate (50% P2O5) at application rate 0.76 g N plant -1 , 0.71 g K plant -1 and 0.42 g P plant -1 .For field study, they were transplanted into micro plot (4 m × 4 m) with three replications.Total of fertilizers per hectare per season applied were: 121.1 kg nitrogen, 67.3 kg phosphorus and 45.5 kg potassium.The fertilizers were applied manually three times per season after transplant.The plants were harvested at days 105 after transplant and air dried and stored in the laboratory for analysis.Three replicates with two single inoculations treatments plus one combination inoculation and control were studied in field experiments.

Plant inoculation with Bacillus sp.
An appropriate dilution of 10 8 /mL bacterial cell suspension of B. pumillus or/and B. subtillis were used to inoculate onto the plants by spraying with pump sprayer (ratio 1:100 ml H2O) in all experiments at 20, 45 and 65 days after sowing.

Grains physiological analysis
The rice yield parameters which include total grain number per panicle, fill grains per panicle, 1000 g weight, total spikelet per panicle, panicle length and number of primary branch per panicle were determined after harvest.The filled grains were separated from unfilled grains by using salt solution of 1.06 specific gravity (Seizo, 1980).Table 1.Total grain number per panicle, fill grains per panicle, 1000 g weight, total spikelets per panicle, panicle length and number of primary branch per panicle at 100 DAT.

Treatments
Total Grain Panicle were randomly selected and analyzed for breaking tensile strength (BTS), or shattering level.BTS is a measure of the maximum amount of weight, in grams; a single grain can hold or attached to pedicle before being detached (Thurber et al., 2011).The shattering level was measure for each panicle by separating the panicle into three portions; upper, middle and lower.An electronic force device (FGP-1 Nidec SHIMPON) with the resolution of ± 0.01 N was used for measurement of the shattering force.The measurement of the maximum force was recorded when the applied force breaks the grain from the panicle.Average BTS values for the measurements were recorded for each sample.

Statistical analysis
All data were statistically analyzed using the SAS Software Program (Version 9.3), and treatment means were compared using Duncan Multiple Range Test (P < 0.05).

RESULTS
The interaction effects of single inoculation, B. subtilis (UPMB10) and B. pumillus (GM118) and also combined inoculations (Mixture) on production of IAA is given in Figure 1.In the presence of L-tryptophan, application of UPMB10 and Mixture indicated the highest IAA production which were 63.8 and 50.1 µg/mL, respectively, while GM118 produced only 1.3 µg/mL (Figure 1a).Besides, UPMB10 and Mixture still showed the highest IAA production which were 8.74 and 7.04 µg/mL, respectively even without the addition of L-tryptophan as compared to GM118 (4.07 µg/mL) (Figure 1b).The result showed that IAA production by bacteria inocula were significantly increased by the single inoculation of UPMB10 and Mixture compared to single inoculation of GM118.
Generally, all of the single inoculations, GM118 and UPMB10 and combined inoculation (Mixture) treatments indicated significant higher result in total number of grain per panicle, fill grains per panicle, 1000 grains weight, total spikelets per panicle, panicle length and number of primary branch per panicle at 100 DAT compared to control as represented in Table 1.The application of UPMB10 showed the highest total number of grain per panicle (98.83), number of filled grain per panicle (97.92), total number of spikelets per panicle (129.04) and panicle length (23.7 cm).However, GM118 indicated the highest number of primary branch (12.9) and showed same panicle length (23.7 cm) result with UPMB10, while Mixture showed the highest 1000 grains weight which was 22.53 g.Besides, most of parameter data showed insignificant difference between inoculations treatment except on data of 1000 grains weight and number of primary branch.
Figure 2 represent the average shattering force value measured after application of single inoculation, GM118 and UPMB10 and combined inoculation (Mixture) with different positions of grain panicle; upper, middle and lower.A significant difference was shown in lower position of grain panicle.Detaching the grain from the lower position of the panicle required a force of 160.4 gf but the forces required by those of the upper and middle positions were 152.8 and 151.3 gf, respectively (Figure 3).Meanwhile, single inoculation of UPMB10 indicated the highest breaking force to detach grain from pedicle (162.2 gf), followed by mixture (158.6 gf) and GM118 (151.9 gf) and control treatments required only 146.7 gf to detach grain from pedicle (Figure 4).The effect of single and combine inoculations on the rice grain yield in  IADA Kemasin Semerak granary is given in Figure 5.The highest grain yield (t/ha) was shown by single inoculation, UPMB10 and combined inoculation (mixture), which 5.53 and 5.50 t/ha, respectively, and followed by single inoculation, GM118 (4.80 t/ha) as compared to control treatment.Both of UPMB10 and mixture indicated 61% difference yield/ha from control, meanwhile, GM118 indicated only 40% difference yield/ha from control.both

DISCUSSION
Plant growth promoting rhizobacteria (PGPR) used in this study were single inoculation; Bacillus subtilis (UPMB10) and B. pumillus (GM118) and also combine inoculation of UPMB10 and GM118 (Mixture).UPMB10 act as nitrogen fixer while GM118 act as phosphate solubilizer (Supplementary result: Plates 1 and 2).The presence of Plate 1. Nitrogen fixing ability of single and combine inoculation on N-free solid malate (Nfb) medium agar plates.Source: Day and Döberenier (1976).
Plate 2. P solubilization of single and combine inoculation on Pikovskaya agar media plates.Source: Pikovskay (1948).UPMB10 and GM118 in Mixture that caused it has the potential to fix nitrogen and solubilize phosphate.Both characteristics are very useful in ensuring that these isolates have potential to promote plant growth.
From Table 1, total number of grain per panicle, fill grains per panicle, 1000 grains weight, total spikelets per panicle, panicle length and number of primary branch per panicle were significantly different between inoculated treatments and control.With presence of bacteria inoculation its improvement grain physiological quality and showed positive symbiosis interaction between plant and bacteria.Rice growth performance is exposed to environmental factors which affect the physiological processes inside rice plant cells.Improving rice physiological characteristic is considered to be desirable because of its agronomic importance towards the high rice yield achievement (Li et al., 2012).This finding is in agreement with Alam et al. (2008) and Islam et al. (2012) who obtained increased rice grain yield in bacteria inoculation over the uninoculated control.
UPMB10 produced the highest breaking force to detach grain from panicle and this finding was complementary to the in-vitro assay, with showed that UPMB10 produce the highest concentration of IAA either with or without Ltryptophan as represented in Figure 1.The presence of UPMB10 in Mixture who contributed to the higher concentration of IAA compared to GM118 and in turn requires high breaking force to detach grain from panicle.IAA, one of the phytohormones, can be considered as the most physiologically active auxin in plants that influences root and shoot dry matter partitioning, root and stimulate both rapid (e.g., increases in cell elongation) and longterm (e.g., cell division and differentiation) responses in plants (Cleland, 1990;Subba, 1999).Biswas et al. (2000) suggested a possibility of interrelation between increased tiller production and yield of rice with a change in hormonal equilibrium, particularly IAA, due to rhizobial inoculation.Hence, it can be assumed that IAA inhibits or prolongs abscission process in the panicle via reduction of sensitivity of abscission zone (AZ).This abscission process is triggered by ethylene hormone and interplay with IAA.Meir et al. (2010), showed the role of IAA in early inhibition of organ abscission.A basipetal through the AZ inhibited abscission by rendering the AZ insensitive to ethylene.Several researchers have reported that in abscission process, the interplay between IAA and ethylene is well established (Roberts et al., 2002).Furthermore, it has been reported that elevated levels of IAA can "protect" AZ cell from abscission inducing stimulate on Arabidopsis (Basu et al., 2013).This finding can be assumed that the UPMB10 was the key trigger of hard detachment of grain panicle.
Besides, previous studies also indicated that IAA production could be a practical characteristic to choose endophytic and rhizosphere competent bacteria for rice growth promoting agents (Etesami and Alikhani, 2016;Etesami et al., 2015).The dynamic responsive pattern of IAA distribution within PGPR is a key factor to support plant growth, its reaction toward root rhizosphere, and specifically for stimulating development of plant organs.
It is clearly shown that the application of both single and combined inoculations of PGPR significantly increased the non-shattering of grain panicle (Figure 2).The result also indicated a significant difference between the lower position of grain panicle but not in the upper and middle positions and in turn showed that the lower position requires a higher force to detach grain from the panicle (Figure 3).Besides, only lower grain position of the panicle showed significant effect toward nonshattering habit.This may be due to non-uniform grain maturity across the panicle, which affects the required detaching force of a grain from the panicle (Lee and Huh, 1984;Szot et al., 1998).
This shattering habit result was also supplementary with the enhancement of total grain number and total spikelet number per panicle pattern in a time.So, the number of grains remain on the panicle can be preserved until harvesting time and reduce post-harvest losses.Hence, the national SSL percentage target to achieve yield in average 6 ton/ha can be done by applying the bacteria PGPR.

Conclusion
The data in this study showed that the positive effect of applying single inoculation; B. subtilis (UPMB10) and B. pumillus (GM118) along with combined inoculations (Mixture) of PGPR give significant increase in yield production compared to control.Besides that, the ability of UPMB10 and Mixture inoculation produce indole-3acetic acid (IAA) and gives dissimilar result in grain shattering habit compared to GM118 inoculation.Again, single inoculation of UPMB10 and Mixture showed that both of them produced high breaking force energy to prevent detachment of grain from panicle.Moreover, increased number of total spikelet per panicle and total grain per panicle data also can be observed.Hence, PGPR that produce IAA can increase non-shattering of grain on panicle and in turn increase rice yield production.
Figure 1.IAA production of single and combine inoculation with and without addition of L-tryptophan.
the mean of three replications and values in each column with different letters are significantly different according to Duncan`s Multiple Range Test at p ≤ 0.05.

Figure 2 .Figure 3 .
Figure 2. The average shattering force value measured after application single inoculation of GM118 and UPMB10, and Mixture (GM118 + UPMB10) inoculation with different position of grain panicles.Bar graphs with different letters are significantly different according to Duncan`s Multiple Range Test at p ≤ 0.05

Figure 4 .
Figure 4.The average shattering force value measured after application single inoculation of GM118 and UPMB10, and Mixture (GM118 + UPMB10) inoculation of the grain panicle; Bar graphs with different letters are significantly different according to Duncan`s Multiple Range Test at p ≤ 0.05.

Figure 5 .
Figure 5. Grains yield (t/ha) in the field of experiment in IADA Kemasin Semerak granary measured after application of single inoculation of GM118 and UPMB10, and Mixture (GM118 + UPMB10) inoculation.Bar graphs with different letters are significantly different according to Duncan`s Multiple Range Test at p ≤ 0.05