Effect of Plant Growth Promoting Rhizobacteria on Ambrosia artemisiifolia L . Seed Germination

Soil bacteria are able either to stimulate or inhibit seed germination. If seed germination is stimulated, the seedlings of weed species emerge more uniformly, so that they could be killed in the next step of weed control. This investigation focused on testing the germination of Ambrosia artemisiifolia L. on several media: Pseudomonas fluorescens (B1), Azotobacter chroococcum (B2), Bacillus licheniformis (B3), B. pumilus (B4), B. amyloliquefaciens (B5). In control, seeds germinated in water. Seed germination varied depending on bacterial media. Germination was inhibited by bacterial treatments B1 and B3, treatments B2 and B4 stimulated germination, while germination in treatment B5 was similar to control.


INTRODUCTION
Common ragweed (Ambrosia artemisiifolia L.) is an Asteraceae originating in North America (Basset & Crompton, 1975).During the last two centuries, it is spreading in Europe because of international trade and wars.This annual broadleaf weed species was first observed in France and Germany in the middle of the 19 th century (Heckel, 1906).It was also reported in Yugoslavia and Hungary during the 1920s (Be´res & Hunyadi, 1984).A. artemisiifolia presents a two-sided problem.First, its pollen causes allergies that often develop into asthma (Plavšić, 2007).Secon-dly, it is also a weed that can cause substantial yield losses.Its weediness is particularly conspicuous in spring crops such as maize, sunflower, soya bean, sugar beet and, in some respects, pea and cereal crops (Chollet et al., 1999).As a result of its late emergence, it can also grow during the inter-crop period in rapeseed or cereal stubbles, as well as on fallow or set-aside land.However, few studies have been conducted on the control of A. artemisiifolia during the inter-crop period.In many crops A. artemisiifolia can be controlled by different herbicides (glyphosate, isoxaflutole, acetochlor + isoxaflutole, isoxaflutole + terbuthylazine, dicamba, fluroxypyr, bentazone + dicamba, prosulfuron, foramsulfuron + iodosulfuron-methyl-Na, topramezone, topramezone + dicamba, tritosulfuron + dicamba, triasulfuron + dicamba, S-metolachlor+ mesotrione + terbuthylazine, lactophen, fomesafen, clopyralid, clopyralid + phenmedipham + desmedipham, triflusulfuron + phenmedipham + desmedipham + clopyralid, etc.) (Janjic et al., 2011).Mechanical control often gives poor results, because A. artemisiifolia can produce new stems after being cut (Delabays et al., 2005).
In most cases, a second mechanical operation is thus necessary.Also, this species is increasing in importance due to development of herbicide resistance (Zheng et al., 2005).
In Serbia, A. artemisiifolia is on the list of invasive weed species.Vrbničanin et al. (2008a) confirmed its widespread all around Serbia, especially in Vojvodina, Mačva and Šumadija (Figure 1) row crops, orchards, vineyards and rarely grain crops, as well as based grasslands, ruderal, settlement and areas along roads are infected.
Modern crop production technology include increasing concern about the environmental impact of arable farming, and of the effects of herbicides in particular.This concept lead to the development of more integrated approaches in weed management.Use of PGPR as stimulants for seed germination of weed species can provide more uniform weed emergence.Killing of these plants in the next step of weed control can be a good way for reducing the seed bank in the soil.Before that, it is necessary to investigate basic ecological interactions between different bacterial media and seed germination of economically important, especially invasive weed species.Therefore, the main object of this work was to determine the effects of bacterial media on germination of seeds of A. artemisiifolia.The results obtained will show potential for using studied bacterial media in reduction of seed bank of those species in the future.

MATERIAL AND METHODS
Seeds of A. artemisiifolia were collected from arable fields in 2009 for testing effects of five different rhizobacteria on germination.Immediately before germination, seeds were sterilized with 1% (v/v) sodium hypochlorite solution for 10 min, and then seeds were rinsed three times with distilled water.The next rhizobacteria were selected: Pseudomonas fluorescens (B 1 ), Azotobacter chroococcum (B 2 ), Bacillus licheniformis (B 3 ), B. pumilus (B 4 ), B. amyloliquefaciens (B 5 ).In control, seeds germinated in water.Bacterial strains used in the study were isolated from rhizosphere of wheat (P.fluorescens and A. chroococcum) or from manure (B.licheniformis, B. pumilus and B. amyloliquefaciens).Incubation of weed seeds was done with 24h old inocula with cell concentration of 10 8 ml -1 .
Twenty seeds of A. artemisiifolia were selected and placed in each Petri dish and treated with solutions containing the above mentioned bacterial media.In control, only water was added.In treatments, 5 ml of solution containing different bacterial media was added.Three dishes were used for each treatment and control.Germination took place in an incubator (Binder CE) at 25 o C, in the dark.The seeds were considered to be germinating at the moment of radicle emergence.The number of germinated seeds was recorded daily (germination rate), and the final percentage of germination was measured after 8 days.Germination rate (sum of germinations per day) was calculated using the formula described by Maguire (1962): M=n 1 /t 1 +n 2 /t 2 +….+n x /t x, where n 1 , n 2 , ....n x are the numbers of the germinated seeds at times t 1 , t 2 , ....t x in days.Each experiment was conducted three times.
All data were processed by analysis of variance (ANOVA) and means were separated by least significant differences (LSD) test using statistical software Statistica 5.0.

RESULTS
In this experiment, seeds of invasive species A. artemisiifolia were incubated on media with different bacterial species.Dynamic and final percentage of seed germination (Figure 2) and germination rate (Figure 3) differed depending on the treatment.Generally, seed germination of A. artemisiifolia was very low in all treatments, as well as in control.The highest percentage of germinated seeds was recorded for treatment B 2 (10.67%), while the lowest germination was scored for treatment B 3 (1.33%) .In the treatment B 1 seeds did not germinate.Different bacterial treatments had diverse effects (stimulative or inhibitory) on seed germination of A. artemisiifolia.For example, treatments B 2 and B 4 stimulated seed germination, while opposite results were obtained for treatments B 1 and B 3 .Seed germination in treatment B 5 was similar to control.Obtained effects were statistically significant (P<0.05)only for percentage of germination (Table 1).Germination rate for treatment B 1 statistically significant differed from germination rate in control, while differences in germination rate between other treatments and control were not found (Table 1).A one-way ANOVA was performed for seed treatments as the main effect to compare samples of seeds incubated on bacterial media with samples germinating in distilled water.N.S. (P>0.05)-no significant differences; *(P<0.05)-significant differences

DISCUSSION
In continuation of our systematic studies of the PG-PR on seed germination of weed species (Vrbničanin et al., 2008b,c;Sarić and Božić, 2009) we reported effects of different bacterial media on seed germination of several weed species.Besides this, many studies showed that microorganisms can promote germination and growth of different plant species (Gutierrez-Manero et al., 2001;Egamberdiyeva, 2007).In order to evaluate effects of this group of bacteria, we have studied effects of five different bacterial media on germination of A. artemisiifolia.P. fluorescens was sometimes classified as deleterious rhizobacteria (DRB) (Zdor et al., 2005) and sometimes as PGPR (Jaleel et al., 2007).This species (B 1 ) totally inhibited germination of A. artemisiifolia, although Ahmad et al. (2008) confirmed plant growth promoting traits of this bacteria.Carrillo-Castaneda et al. (2002) showed that effect of P. fluorescens on alfalfa germination depend on conditions in which bacterial cultures develop.
It was found that Bacillus species have variable effects on the seed germination.Namely, Egamberdiyeva (2007) reported positive effects of Bacillus on seed germination and plant growth as a result of their production of plant growth-promoting substances.On the other hand, Sarić and Božić (2009) found Bacillus species to have inhibitory effect on germination of C. campestris and alfalfa.In this experiment only effect of B. pumilus (B 4 ) on germination of A. artemisiifolia were stimulative.Opposite results were obtained for B. licheniformis (B 3 ), which inhibited seed germination and for B. amyloliquefaciens (B 5 ), which had no effect.Results for B. pumilus are consistent, while results for B. licheniformisa are oposite with those reported by Gutierrez-Manero et al. (2001) showing that both promoted growth of alder trees.
In conclusion, results from this study showed different PGPR have diverse (stimulative or inhibitory) effects on germination of A. artemisiifolia, but it may not be possible to extrapolate the results of these in vitro studies to soil or rhizosphere conditions.The reason is influence of different conditions in the soil (pH, microelements, salinity) on excretion of plant growth-promoting supstances by PGPR (Narula and Gupta, 1986).Some (A. chroococcum and B. pumilus) bacteria tested have potential as promoters of seed germination, while some of them (P.fluorescens, B. amyloliquefaciens and B. licheniformis) have not.So, it is necessary to do screening of effects of many bacteria on seed germination of A. artemisiifolia, and based on that to evaluate possibility of their use in this weed species management.Uticaj zemljišnih bakterija na klijanje semena korovske vrste Ambrosia artemisiifolia L.