The effect of plant growth promoting rhizobacteria on Datura stramonium L., Abutilon theophrasti Med., Onopordon acanthium L. and Verbascum thapsus L. seed germination

1University of Belgrade, Faculty of Agriculture, Nemanjina 6, Belgrade, Serbia 2 Educons University, Faculty of Ecological Agriculture, V. Putnika bb., 21208 Sremska Kamenica, Serbia 3Institute for Plant Protection and Environment, Teodora Drajzera 9, Belgrade, Serbia 4 Institute of Pesticides and Environmental Protection, Banatska 31b, Belgrade, Serbia (*sava@agrif.bg.ac.rs) Received: February 26, 2014 Accepted: July 16, 2014


INTRODUCTION
Increasingly frequent development of weed resistance to herbicides and an increase in environmental concerns and pressure to reduce pesticide use have encouraged a search for non-chemical alternatives in weed control.Stimulation of seed germination by plant growth promoting rhizobacteria (PGPR) can be a component in weed control programmes.Namely, seedlings of weed species emerge more uniformly when seed germination is stimulated, so that they can be killed in a next step of weed control.That is a way of reducing seedbanks in soil and seedling growth.
Abutilon theophrasti Med.(velvetleaf) and Datura stramonium L. (jimsonweed), are widespread and very important weeds in many row crops in Serbia and worldwide (especially problematic in corn, sunflower, soybeans, sugar beet or cotton as major crops), causing serious yield and economic losses (Vrbničanin et al., 2008a,b).Both of them are annual weed species that reproduce by seeds.Most seeds fall near the parent plant, but some disperse to greater distances with water, mud, soil movement, and especially through agricultural operations.Seed production can be high with A. theophrasti and D. stramonium producing up to 50,000 and 45,000 seeds per plant, respectively.Buried seeds are long-lived.A large proportion of buried jimsonweed seeds can survive for 40 years or more, while velvetleaf may survive for 50 years or more.Once established, velvetleaf and jimsonweed are difficult to control because of their long-lived seeds and sporadic germination pattern.In addition to chemical control, manual removal of individual plants before seeds develop can be of great help in controlling small or sparse populations.For more troublesome infestations, integrated weed management may substantially reduce soil seedbanks as soon as in the next couple of years.
Onopordon acanthium L. (Scotch thistle) and Verbascum thapsus L. (common mullein) spread along roadsides, annual grassland, pastures, rangeland, and other disturbed habitats.Sometimes both of them penetrate into forest clearings and arable fields, as well as orchards, vineyards, etc.Both of them reproduce by seeds.Most seeds fall near the parent plant.Soil disturbance facilitates germination and seedling growth.Under field conditions, some common mullein seeds are able to survive for up to 35 years or more, while Scotch thistle can remain viable in soil seedbank for at least 7 years and possibly up to 20 years or more.Establishing or encouraging perennial grasses can increase Scotch thistle and common mullein seedlings mortality due to increased competition for moisture.Both of these plants are difficult to control even with systemic herbicides because of their very hairy foliage.
The present study was conducted to determine the ecological interaction between different bacterial media and four weed species: D. stramonium, A. theophrasti, O. acanthium and V. thapsus.The main objective of this study was to determine the effects of those media on seed germination and assess whether the bacteria tested can be used as a means of seedbank reduction.

MATERIAL AND METHODS
Seed Source.Seeds of D. stramonium, A. theophrasti, O. acanthium and V. thapsus were collected from arable fields in 2010.The collected seeds were cleaned and stored at room temperature (approximately 20-25 o C) and were put on +4 o C for 30 days before studying.Immediatelly before imbibition, seeds were sterilized with 1 % (v/v) sodium hypochlorite solution for 10 min, and then rinsed three times with distilled water.
Imbibition.A hundred seeds of each species were selected and their dry mass was measured.During imbibition the seeds were put in Petri dishes and treated with solutions of different bacterial strains and humates.Only water was added to control seeds.Water uptake (imbibition) by seeds was measured after 24 h by drying their surface on filter paper and then re-weighing.
Germination.The experiments were carried out in an incubator (Binder CE).Imbibed seeds were germinated in Petri dishes in the dark at 25 o C. Twenty seeds were placed in each of three dishes per treatment.In treatments, 5 ml of solution containing different bacterial strains at a concentration of 10 8 cells ml -1 were added.In the control, only water was added.Seeds were considered to be germinated after the emergence of radicles.Germinated seeds were counted and the percentage of germination was calculated daily over a period of eight days.Germination rate was calculated using the formula described by Maguire (1962): where n 1 , n 2 , ....n x represent the number of germinated seeds at times t 1 , t 2 , ....t x in days.Each experiment was conducted three times.
Statistics.The results were processed using the software Statistica 5.0 by descriptive statistics and LSD test.

RESULTS AND DISCUSSION
Continuing our systematic studies of PGPR and seed germination of weed species (Vrbničanin et al., 2008(Vrbničanin et al., , 2011;;Sarić & Božić, 2009) we have reported the effects of different bacterial media on seed germination of several weed species.The reports have shown that different PGPR (A. chroococcum, B. megatherium, B. circulans and B. pumilus) and their combinations have a great potential as promoters of seed germination of the weed species Iva xanthifolia Nutt., Amaranthus retroflexus L., Sorghum halepense L.(Pers.) and Ambrosia artemisiifolia L. (Vrbničanin et al., 2008(Vrbničanin et al., , 2011)).On the other hand, Sarić and Božić (2009) found Bacillus species to have an inhibitory effect on germination of Cuscuta campestris Yunck seeds.Many other researchers have also reported stimulative effects of this group of bacteria on seed germination or plant growth of various crops and weed species (Shishido et al., 1996, Gutiérrez-Mañero et al., 2001;Ryu et al., 2003).In this experiment, seeds of four weeds were incubated on media with different Bacillus species and humates.The effect of the tested bacterial media on seed mass increase varied depending on the species and media after 24 hours of imbibition (data not shown).Seed germination of D. stramonium, A. theophrasti, O. acanthium and V. thapsus differed in responses to different media.The effects of bacterial inoculation on seed germination of weed species are presented in Figure 1.The results show an obvious lack of uniformity in effects of the tested bacterial media on seed germination.For example, B. subtilis (MO 3 ) increased germination of D. stramonium and O. acanthium and at the same time decreased germination of A. theophrasti.Some of these results agree with the findings of Bhat and Alagawadi (1998), Harper and Lynch (1980) and Vrbničanin et al. (2008), who had found PGPR to stimulate seed germination and seedling growth.However, the results indicating contrary effects of bacterial media on seed germination contradict the results of the mentioned studies, but agree with the findings of Miché et al. (2000), who had found the seed germination of Striga hermontica to be inhibited by a product of the plantpromoting bacterium Azopirillum brasilense known to be beneficial for sorghum growth.Generally, the highest germination (100.0%) in all treatments was recorded for V. thapsus seeds.Therefore, it was not possible to analyze the bacterial effect on germination of that species as all its seeds germinated, control included.Comparing all treatments, we found germination to vary among the three other species.For instance, the lowest germination in treatments MO 1 and MO 2 (two populations of B. licheniformis) was scored by D. stramonium seeds (5.0% and 6.7%, respectively).On the other hand, the highest germination in treatments with the two populations of B. licheniformis was recorded for A. theophrasti (41.7% in MO 1 and 40.0% in MO 2 ).In MO 3 and MO 4 treatments, A. theophrasti had the highest percentage of germinated seeds (28.3% and 43.3%, respectively).In MO 5 treatement, A. theophrasti germinated the best (65.0%), while the lowest germination was recorded for D. stramonium (5.0%).
The treatments had diverse effects (stimulative or inhibitory) on seed germination, depending on the type of media and weed species, in that order.The effect was positive in some treatments and species and negative in others (Figure 1).For example, B. licheniformis (MO 1 and MO 2 ) inhibited A. theophrasti and D. stramonium seed germination, while the results were opposite for O. acanthium.
Many researchers have reported stimulative effects of Bacillus on seed germination and plant growth as a result of their production of plant growth-promoting substances such as gibberelins, indoleacetic acid, ammonia, hydrogen cyanide, etc. (Gutiérrez-Mañero et al., 2001;Ahmad et al., 2008).In our study, both populations of B. licheniformis exerted stimulative effects on O. acanthium seed germination.On the other hand, the effect was negative on the germination of D. stramonium and A. theophrasti seeds, which is in line with the effect of that bacterial species on A. artemisiifolia (Vrbničanin et al., 2011).Except for B. licheniformis, contradictory results have been reported for B. subtilis (as mentioned above) and B. megatherium, which had stimulated germination of D. stramonium and O. acanthium, while seed germination of A. theophrasti was not effected.
Both the germination rate (Table 1) and the final percentage of germination varied similarly for all species tested.In contrast to the final percentage of germination, differences in germination rates were found among treatments of V. thapsus.Thus, the highest germination rate regarding all studied microorganisms was shown by V. thapsus seeds (in H 2 O -48.7, MO 1 -53.7,MO 2 -52.4,MO 3 -51.4,MO 4 -48.0, and MO 5 -51.0 seeds day -1 ).The results showed that statistical differences in germination rates between treatements of this species were not as prominent as for the other three species, but they were significant (p<0.05) and very significant (p<0.01) in some treatments (Table 2).The germination rate of A. theophrasti varied considerably in some treatments (in H 2 O -18.6, MO 1 -17.6,MO 2 -3.2, MO 3 -6.4,MO 4 -13.8, and MO 5 -25.6 seeds day -1 ).The highest value was recorded in treatment MO 5 , and it very significantly differed (p<0.01)from all other treatments.The germination rates of D. stramonium and O. acanthium seeds (0.7-2.0 seeds day -1 and 1.1-2.7 seeds day -1 , respectively) were much lower than the rates of the other two species.All treatments had very significant (p<0.01)effects on seed germination of both species, compared to the control (Table 2).Treatments MO 1 and MO 2 decreased and all other treatments (MO 3 , MO 4 and MO 5 ) increased the germination rate of D. stramonium, while all treatments had stimulative effects on O. acanthium germination rate.Aqueous suspensions of different bacteria were used in this experiment but the results cannot be extrapolated to soil or rhizosphere conditions because various conditions in soil (pH, microelements and salinity) can influence the excretion of plant growth-promoting substances by PGPR strains and their effect on seed germination (Narula & Gupta, 1986;Egamberdiyeva, 2007).

CONCLUSIONS
In conclusion, the results of this study indicate that: 1) the bacteria tested may have opposite effects (stimulative or inhibitory) on different weed species; 2) the bacteria had both promoting and inhibiting potential for seed germination, depending on plant species.Therefore, screening tests for bacterial effects on many weed species are needed.Based on such results it would be possible to evaluate the use of PGPR bacteria in weed management practices.Interesting new perspectives have opened, helping us to implement the acquired knowledge on PGPR-seed germination interactions as a biological weed control method in sustainable agricultural practices.Uticaj zemljišnih bakterija na klijanje semena Datura stramonium L., Abutilon theophrasti Med., Onopordon acanthium L. i Verbascum thapsus L.