Screening plant growth-promoting rhizobacteria for improving growth and yield of wheat

Transcription

1 Journal of Applied Microbiology 2004, 96, doi: /j x Screening plant growth-promoting rhizobacteria for improving growth and yield of wheat A. Khalid, M. Arshad and Z.A. Zahir Soil Microbiology and Biochemistry Institute of Soil and Environmental Sciences, University of Agriculture, Faisalabad, Pakistan 2003/0311: received 14 April 2003, revised 14 October 2003 and accepted 18 October 2003 ABSTRACT A. K H A LID, M. A R S H A D A N D Z. A. Z A H I R Aims: Plant growth promoting rhizobacteria (PGPR) are commonly used as inoculants for improving the growth and yield of agricultural crops, however screening for the selection of effective PGPR strains is very critical. This study focuses on the screening of effective PGPR strains on the basis of their potential for in vitro auxin production and plant growth promoting activity under gnotobiotic conditions. Methods and Results: A large number of bacteria were isolated from the rhizosphere soil of wheat plants grown at different sites. Thirty isolates showing prolific growth on agar medium were selected and evaluated for their potential to produce auxins in vitro. Colorimetric analysis showed variable amount of auxins (ranging from 1Æ1 to 12Æ1 mgl )1 ) produced by the rhizobacteria in vitro and amendment of the culture media with L-tryptophan (L-TRP), further stimulated auxin biosynthesis (ranging from 1Æ8 to24æ8 mgl )1 ). HPLC analysis confirmed the presence of indole acetic acid (IAA) and indole acetamide (IAM) as the major auxins in the culture filtrates of these rhizobacteria. A series of laboratory experiments conducted on two cv. of wheat under gnotobiotic (axenic) conditions demonstrated increases in root elongation (up to 17Æ3%), root dry weight (up to 13Æ5%), shoot elongation (up to 37Æ7%) and shoot dry weight (up to 36Æ3%) of inoculated wheat seedlings. Linear positive correlation (r ¼ 0Æ99) between in vitro auxin production and increase in growth parameters of inoculated seeds was found. Based upon auxin biosynthesis and growth-promoting activity, four isolates were selected and designated as plant growth-promoting rhizobacteria (PGPR). Auxin biosynthesis in sterilized vs nonsterilized soil inoculated with selected PGPR was also monitored that revealed superiority of the selected PGPR over indigenous microflora. Peatbased seed inoculation with selected PGPR isolates exhibited stimulatory effects on grain yields of tested wheat cv. in pot (up to 14Æ7% increase over control) and field experiments (up to 27Æ5% increase over control); however, the response varied with cv. and PGPR strains. Conclusions: It was concluded that the strain, which produced the highest amount of auxins in nonsterilized soil, also caused maximum increase in growth and yield of both the wheat cv. Significance and Impact of Study: This study suggested that potential for auxin biosynthesis by rhizobacteria could be used as a tool for the screening of effective PGPR strains. Keywords: auxins, plant growth-promoting rhizobacteria, wheat growth. Correspondence to: Zahir Ahmad Zahir, Soil Microbiology and Biochemistry, Institute of Soil and Environmental Sciences, University of Agriculture, Faisalabad 38040, Pakistan ( bio@fsd.paknet.com.pk). ª 2004 The Society for Applied Microbiology

2 474 A. KHALID ET AL. INTRODUCTION Net effect of plant microbe interactions on plant growth could be either positive, neutral or negative. All those bacteria inhabiting plant roots and influencing the plant growth positively by any mechanism are referred to as plant growth-promoting rhizobacteria (PGPR) (Kloepper et al. 1986; Frankenberger and Arshad 1995; Arshad and Frankenberger 1998). These bacteria significantly affect plant growth by increasing nutrient cycling, suppressing pathogens by producing antibiotics and siderophores or bacterial and fungal antagonistic substances and/or by producing biologically active substances such as auxins and other plant hormones. A diverse array of bacteria including species of Pseudomonas, Azospirillum, Azotobacter, Bacillus, Klebsiella, Enterobacter, Xanthomonas and Serratia have been shown to promote plant growth. During the last couple of decades, the use of PGPR for sustainable agriculture has increased tremendously in various parts of the world. Significant increases in growth and yield of agronomically important crops in response to inoculation with PGPR have been reported (Chen et al. 1994; Amara and Dahdoh 1997; Biswas et al. 2000a,b; Hilali et al. 2001; Asghar et al. 2002). Studies have also shown that the growth-promoting ability of some bacteria may be highly specific to certain plant species, cv. and genotypes (Nowak 1998). Poi and Kabi (1979) reported that inoculation with Azotobacter strains isolated from the rhizosphere soils of Cucurbita maxima, wheat and jute improved the grain yields but the strains were crop specific. Like other phytohormones, auxins are also synthesized endogenously by plants, however, their hormonal effects have been elucidated by their exogenous applications. There is also ample evidence that numerous soil micro-organisms are actively involved in the synthesis of auxins in pure culture and in soil (Arshad and Frankenberger 1998; Barazani and Friedman 1999; Biswas et al. 2000a,b). Generally, micro-organisms isolated from the rhizosphere and rhizoplane of various crops have revealed more potential of auxin production than those from the root free soil (Sarwar and Kremer 1995a,b; Arshad and Frankenberger 1998). L-tryptophan (L-TRP), an amino acid, serves as a physiological precursor for biosynthesis of auxins in plants and in microbes (Frankenberger and Arshad 1995). Root exudates are natural source of TRP for rhizosphere microflora, which may enhance auxin biosynthesis in the rhizosphere (Kravchenko et al. 1991; Martens and Frankenberger 1994). It is most likely that auxins of microbial origin in the vicinity of plant roots could evoke a physiological response in the host plant. Thus screening of the rhizobacteria for their in vitro potential of auxin production could provide a reliable base for selection of effective PGPR, particularly if this approach is used in combination with screening of rhizobacteria for their growth-promoting activity under gnotobiotic conditions. Thus, we used a combination of two approaches including in vitro auxin production and growth-promoting activity under gnotobiotic conditions to select effective PGPR strains isolated from wheat rhizosphere. The selected PGPR were tested for their growth and yield increasing potential under natural soil (nonaxenic) conditions by conducting pot and field experiments. MATERIALS AND METHODS Isolation and screening of PGPR Several bacterial strains were isolated from the rhizosphere of different varieties of wheat (LU-26S, Watan, Inqlab-91, Pasban-90, Parvaaz) crop grown at different sites. Plants of wheat were uprooted along with good amount of nonrhizosphere soil, brought immediately to the laboratory in polythene bags (2 kg size: cm) and were air-dried within 2 h. The nonrhizosphere soil was removed by gentle shaking leaving behind the rhizosphere soil only (strongly adhering to the roots). The rhizosphere soil was collected from roots by dipping and gentle shaking in sterilized water under aseptic conditions. The soil suspension obtained was used to inoculate glucose peptone agar medium (GPAM) and pure cultures were obtained by streaking three to four times in the fresh medium (Wollum-II 1982). Thirty bacterial isolates showing prolific growth and having different morphological appearance on agar medium were selected and stored for studying auxin biosynthesis in vitro. Auxin production by the rhizobacterial strains both in the presence and absence of L-TRP was determined by colorimetry. For this purpose, 20 ml of GPAM broth were added in 100-ml Erlenmeyer flasks, autoclaved and cooled. Five millilitre of filter sterilized (0Æ2 lm membrane filter, Whatmann) L-TRP solution (5%) were added to the liquid medium (GPAM) to achieve a final concentration of 1Æ0 gl )1. The flask contents were inoculated by adding 1Æ0 ml of 4-day-old bacterial broth adjusted to optical density of 0Æ5 ( CFU ml )1 ) measured at 550 nm by spectrophotometer (ANA-720W; Tokyo Photo-electric Company Limited, Tokyo, Japan). The flasks were plugged and incubated at 28 ± 1 C for 48 h at 100 rev min )1 shaking. Noninoculated/untreated control was kept for comparison. After incubation, the contents were filtered through Whatmann filter paper no. 2. Auxin compounds expressed as (IAA-equivalents) were determined by spectrophotometer (ANA-720W) using Salkowski colouring reagent as described by Sarwar et al. (1992). To keep a check on mutation due to subculturing, the ability of the isolates to produce auxins in vitro was repeatedly confirmed prior to

3 SCREENING OF PGPR 475 next experiment. Similarly, sterilized and nonsterilized soils were inoculated with four selected PGPR and L-TRP (2Æ0 gkg )1 soil)-dependent auxin biosynthesis was monitored. All auxin determinations were made in triplicate. The qualitative production of auxin compounds in the culture filtrates was further confirmed by HPLC-UV as described by Martens and Frankenberger (1991). Filtrates were examined on a reverse phase HPLC (10-A; Shimadzu Corporation, Kyoto, Japan) by using mixture of methanol and water (70 : 30 ratio) as a mobile phase (solvent) at 280 nm using the Shim-pack CLC-ODS C 18 column [4Æ6 mm internal diameter (ID), 250 mm long and 100 Å pore diameter]. Based upon in vitro auxin production, 30 isolates were categorized into three major groups: (i) low (L) auxin producers (a1 a10), (ii) medium (M) auxin producers (a11 a20) and (iii) highly (H) effective auxin producers (a21 a30). Plate and Leonard jar experiments were conducted on two cv. of wheat (Pasban 90 and Inqlab 91) under gnotobiotic (axenic) conditions to see the effectiveness of inoculation with these three groups of rhizobacteria on root and shoot growth, respectively. Wheat seeds were surface sterilized by momentarily exposing to 95% ethanol and immersing in 0Æ2% HgCl 2 solution for 3 min. The seeds were then subjected to six washings with sterile distilled water. Thoroughly washed seeds of the two cv. of wheat were sown on sterilized filter paper sheets placed in Petri plates. Six seeds were sown in each Petri plate with four repeats. Two millilitre of 4-day-old broth of rhizobacteria ( CFU ml )1,0Æ5 of O.D. 550 ) were applied on seeds present in each plate with the help of sterilized pipette. Sterilized distilled water (10 ml) was added to each Petri plate to wet the filter paper sheets and the seeds were covered with another sterilized filter paper sheet. Sterilized broth (free of bacterial population) was applied in case of control. The plates were incubated in a growth room at 28 ± 2 C. After 2 weeks, the sheets were removed and examined for root growth (root elongation and root weight). For Leonard jar experiments, plastic glasses (383 cm 3 ) were filled with sand and 1/2 strength Hoagland solution (Hoagland and Arnon 1950) was applied from jars (460 cm 3 ) through a wick to provide nutrition to the plants. The whole apparatus was autoclaved (25 min at 121 C) prior to the transplantation of seedlings. Surface-disinfected seeds were sown on sterilized filter sheet in Petri plates. Uniformly germinated seeds were transplanted to the glass containing sand under aseptic conditions to eliminate the variation in growth contributed by different endogenous germination rate/potential of the seeds. Five millilitre of 4-day-old inocula were applied to the seedlings growing in sand, 2 days after transplanting. The jars were incubated in the growth room at 28 ± 2 C and 16 h of light (1600 lmol m )2 s )1 ) were supplied daily. Two weeks after transplanting, the plants were uprooted and length and weight of the seedling shoots were measured. Pot and field experiments Based upon the performance of rhizobacteria in the Plate and Leonard jar experiments, four effective auxin-producing PGPR isolates (Ha21, Ha22, Ha23 and Ha30) were selected and used in pot and field trials. Inocula were prepared by growing the selected PGPR in GPAM broth and incubated at 28 ± 1 C with 100 rev min )1 shaking. Four-day-old inoculum ( CFU ml )1,0Æ5 of O.D. 550 ) was injected into sterile peat at 100 ml kg )1 peat and incubated for 24 h at 28 ± 1 C prior to seed inoculation. Seeds were inoculated by mixing with peat and 10% sugar solution at 100 ml kg )1 peat while controls consisted of the seeds treated with peat having nutrient broth and sugar solution without PGPR. Treated seeds were dried under shade for 6 8 h. For pot experiment, soil sample was collected, air-dried, sieved (2-mm/10-mesh) and analysed for physico-chemical characteristics before filling the pots. The soil was clay loam having ph 7Æ9; electrical conductivity of saturated soil extract (ECe), 1Æ6 dsm )1 ; cation exchange capacity (CEC), 6Æ8 cmol(+) kg )1 and organic matter, 0Æ72%. Eight inoculated and uninoculated seeds of two wheat cv. (Pasban- 90 and Inqlab-91) were sown in soil filled pots (12 kg soil per pot) receiving nutrient inputs of NPK at 120, 75 and 50 kg ha )1 as urea, diammonium phosphate and muriate of potash, respectively. All of PK and half of N were mixed with soil at the time of sowing while remaining N was applied in solution form at tillering. Four seedlings were maintained in each pot after germination. The pots were arranged randomly with four repeats at ambient light and temperature in a wire house. Good quality canal water [EC ¼ 0Æ03 ds m )1, sodium adsorption ratio (SAR) ¼ 0Æ26 (mmol l )1 ) 1/2 and residual sodium carbonates (RSC) ¼ 0] meeting the irrigation quality criteria for crops (Ayers and Westcot 1985) was used for irrigation. The plants were harvested after 5 months and data were recorded. Two years replicated field trials were conducted with the same treatments and agronomic practices as used in pot experiment, however, whole dose of PK and half N was broadcast at the time of soil preparation, and remaining half N was applied at tillering. Seeds of wheat were sown with single row seed drill keeping row to row distance of 25Æ0 cm. Each experiment was conducted in randomized complete block design (RCBD) with four repeats. The analysis of composite soil (typic haplocambids) samples collected from experimental field (ca 0 15 cm layer) revealed the following characteristics: texture, clay loam; ph 7Æ7; ECe, 1Æ9 dsm )1 ; CEC, 5Æ9 cmol(+) kg )1 and organic matter, 0Æ58%. Data on plant height, number of tillers, spike length, spikelets per

4 476 A. KHALID ET AL. spike, straw and grain yields were recorded on maturity after 5 months. Statistical analysis The experimental data were analysed statistically according to Steel and Torrie (1980) and means were compared by Duncan s Multiple Range Test (Duncan 1955). Correlations between in vitro auxin production by PGPR and effect on growth parameter(s) were also calculated (Steel and Torrie 1980). RESULTS A series of laboratory, pot and field experiments were conducted to assess the potential of various rhizobacterial isolates for auxin biosynthesis and improving growth and yield of two cv. of wheat (Triticum aestivum L.). Biosynthesis of auxins Results of colorimetric analysis indicated that different isolates of rhizobacteria varied greatly in their efficiency for producing auxins in the broth medium (GPAM), both in the presence and absence of L-TRP (Table 1). Among 30 isolates tested, 73% (22 isolates) produced auxins (ranging from 0Æ6 to12æ1 mgl )1 IAA-equivalents) in the absence of L-TRP. Auxins produced by the 10 isolates belonging to rhizobacterial group H (effective auxin producers; Ha21,, Ha30) ranged from 5Æ1to12Æ1 mg IAA-equivalents per litre, with an average amount of ca 7Æ0 mgl )1. In the presence of L-TRP, bacterial efficiency for auxin synthesis was enhanced by several folds (ranging from 1Æ8 to24æ8mg IAA-equivalents per litre). L-TRP-derived auxin biosynthesis by group H bacteria varied between 13Æ8 and24æ8 mg IAA-equivalents per litre, which was almost two fold greater than the L-TRP unamended culture. All other isolates belonging to group L (La1,, La10) and M (Ma11,, Ma20) were also Table 1 In vitro auxin production by low (L), medium (M) and high (H) auxin producing rhizobacteria in glucose peptone agar medium. Data are average of three replications Group of rhizobacteria IAA-equivalents (mg l )1 ) Range Without L-TRP With L-TRP Mean ± S.E. Without L-TRP With L-TRP L (a1 a10) 0Æ0 1Æ1 1Æ8 13Æ4 0Æ17 ± 0Æ014 6Æ2 ± 1Æ155 M (a11 a20) 1Æ2 5Æ0 2Æ7 14Æ3 3Æ1 ± 0Æ199 10Æ3 ± 1Æ857 H (a21 a30) 5Æ1 12Æ1 13Æ8 24Æ8 6Æ9 ± 0Æ581 17Æ7 ± 1Æ108 IAA, indole acetic acid; L-TRP, L-tryptophan; S.E., standard error. able to derive auxins from L-TRP, however, they were relatively less effective in auxin biosynthesis. Auxin production in sterilized and nonsterilized soils, amended with L-TRP and inoculated with four selected PGPR isolates was also monitored (Table 2). Data revealed that inoculation with the PGPR substantially stimulated auxin synthesis in soil. Inoculation with selected PGPR strain Ha21 was found the most effective auxin producing rhizobacterium in sterilized soil (27Æ5 mg IAA-equivalents kg )1 soil). The magnitude of auxin synthesis was less in the inoculated-sterilized soil than the uninoculated nonsterilized soil both amended with L-TRP at 2Æ0 gkg )1 soil (Table 2) implying that the mixed indigenous microflora are comparatively more effective than the single strain inoculation. However, inoculation of nonsterilized soil with PGPR isolates further stimulated the auxin production indicating that inoculation with specific micro-organisms is even effective in the presence of indigenous soil microflora. HPLC analysis demonstrated the presence of IAA and indole-3-acetamide (IAM) as major L-TRP-derived microbial products both in broth medium and in soil. Presence of IAM also indicated that TRP fi IAM fi IAA pathway of auxin biosynthesis was active in these rhizobacteria. All the isolates of rhizobacteria tested for auxin biosynthesis in vitro were further screened for their growthpromoting effects on wheat seedlings by conducting Plate and Leonard jar experiments under controlled (axenic) conditions. Different isolates of rhizobacteria had variable effects (both negative and positive) on root elongation and weight of roots in two wheat cv. tested (Table 3). Overall, inoculation with the bacterial isolates of group H resulted in maximum increase in root elongation (17Æ3%) and weight (11Æ4%) of cv. Pasban-90 compared with uninoculated control. Similarly in case of cv. Inqlab-91, the same group of bacteria promoted root elongation and weight by 11Æ3 and 13Æ4%, respectively compared with uninoculated control. Table 2 L-tryptophan-derived auxin biosynthesis in soil inoculated by four most active plant growth-promoting rhizobacterial isolates belonging to group H (high auxin-producing rhizobacterial isolates). The data are average of four replications PGPR isolate IAA-equivalents (mg kg )1 soil) Sterilized soil Uninoculated 1Æ70 d 35Æ7c Ha21 27Æ5 a 45Æ2 b Ha22 13Æ5 c 66Æ3 a Ha23 17Æ6 b 41Æ5 b Ha30 14Æ9 c 38Æ8 c Nonsterilized soil Mean values sharing similar letter do not differ significantly at P <0Æ05. IAA, indole acetic acid.

5 SCREENING OF PGPR 477 Table 3 Effects of low (L), medium (M) and high (H) auxin producing rhizobacterial isolates on root growth of two wheat cv. grown under gnotobiotic conditions in Plate experiments (average of 40 values; four repeats 10 isolates) Group of rhizobacteria Cutivar Pasban-90 Root elongation (cm) Dry root weight Cultivar Inqlab-91 Root elongation (cm) Uninoculated 9Æ63 b 0Æ070 b 10Æ6b 0Æ052 b L (a1 a10) 8Æ16 c 0Æ064 c 10Æ4b 0Æ048 c M (a11 a20) 9Æ72 b 0Æ071 b 11Æ2ab 0Æ052 b H (a21 a30) 11Æ3 a 0Æ078 a 11Æ8a 0Æ059 a Dry root weight Values followed by different letters in a column were significantly different (P < 0Æ05), using Duncan s multiple range test. Table 4 Effects of low (L), medium (M) and high (H) auxin producing rhizobacterial isolates on shoot growth of two wheat cv. grown under gnotobiotic conditions in Leonard Jar experiment (average of 40 values; four repeats 10 isolates) Group of rhizobacteria Cultivar Pasban-90 Shoot elongation (cm) Dry shoot weight Cultivar Inqlab-91 Shoot elongation (cm) Uninoculated 15Æ7c 0Æ113 c 14Æ6d 0Æ113 c L (a1 a10) 17Æ2b 0Æ117 c 16Æ3c 0Æ115 c M (a11 a20) 18Æ8 a 0Æ140 b 17Æ2 b 0Æ131 b H (a21 a30) 19Æ5a 0Æ167 a 20Æ1a 0Æ154 a Dry shoot weight Values followed by different letters in a column were significantly different (P < 0Æ05), using Duncan s multiple range test. Results of Leonard jar study revealed that shoot growth (length and weight) of cv. Pasban-90 and Inqlab-91 was significantly improved by bacterial inoculation (Table 4). Once again, maximum increases in shoot length (24%) and weight (47%) in cv. Pasban-90 were recorded in case of inoculation with isolates belonging to H group. For cv. Inqlab-91, increases in shoot length and weights observed in response to inoculation with rhizobacteria of H group were up to ca 37% in both parameters compared with uninoculated control. When increases in root and shoot growth were regressed against in vitro auxin production by the rhizobacteria, significant linear correlations were found. Significant positive linear correlation (r ¼ 0Æ99**) was observed between auxin (IAA-equivalents) production in vitro in the absence of L-TRP by rhizobacteria and root elongation of cv. Pasban-90 and Inqlab-91. Also a significant correlation (r ¼ 0Æ99**) was found between auxin production in the presence of L-TRP and root elongation of cv. Pasban-90. Similarly, a highly significant correlation (r ¼ 0Æ99**) was found between in vitro auxin production in the absence of L-TRP and root weight of both the cv. while in the presence of L-TRP, only root weight of cv. Inqlab-91 was significantly correlated. A significant linear correlation (r ¼ 0Æ99**) was observed between the shoot weight of both the cv. and production of auxins (IAA) in vitro by rhizobacterial isolates in the absence and in the presence of L-TRP, however, in case of shoot length, correlation was nonsignificant in both the cv. Pot and field trials Seed inoculation with four selected PGPR isolates significantly affected the growth and yield of two wheat cv. under wire house conditions (Table 5). PGPR inoculation significantly enhanced number of tillers of cv. Pasban-90 while response was nonsignificant in case of cv. Inqlab-91. Maximum number of tillers were recorded in cv. Pasban- 90 in case of inoculation with isolate Ha30 (23Æ2% more than uninoculated control). In case of cv. Inqlab-91, all the PGPR isolates except Ha22 and Ha23 increased the tillers up to 6Æ1% over uninoculated control. Inoculation with PGPR also caused significant increases in straw yields of both the cv. compared with their respective uninoculated controls (Table 5). In case of cv. Pasban-90, isolate Ha30 gave the most promising results and caused an increase of 12Æ4% in straw yield while in cv. Inqlab-91, straw yield was increased up to 10Æ1% in response to inoculation with Ha23 compared with respective uninoculated control. In case of cv. Pasban- 90, PGPR Ha23 and Ha30 significantly increased the grain yield (by 13Æ7 and 14Æ7%, respectively) but none of the PGPR significantly increased grain yield of cv. Inqlab-91 (Table 5). Data of 2 years repeated field experiments indicated that inoculation with selected PGPR isolates had significant but variable effects on growth and yield of both the cv. of wheat (Table 6). During the first year, PGPR inoculation significantly increased number of tillers of two wheat cv.

6 478 A. KHALID ET AL. PGPR Isolaters No. of tillers per plant Straw yield Grain yield Pasban-90 Inqlab-91 Pasban-90 Inqlab-91 Pasban-90 Inqlab-91* Table 5 Effect of inoculation with selected plant growth-promoting rhizobacteria (PGPR) on growth and yield of two wheat cv. grown in pots (average of six repeats) Uninoculated 4Æ31 c 4Æ43 NS 5Æ73 bc 5Æ82 b 3Æ07 b 3Æ03 NS Ha21 4Æ30 c 4Æ70 5Æ34 c 6Æ24 a 3Æ14 ab 3Æ13 Ha22 5Æ0 ab 4Æ10 6Æ09 ab 5Æ79 b 3Æ15 ab 2Æ96 Ha23 4Æ90 b 4Æ40 6Æ01 b 6Æ41 a 3Æ49 a 3Æ24 Ha30 5Æ31 a 4Æ50 6Æ44 a 6Æ18 ab 3Æ52 a 3Æ06 *Wheat cv. NS, not significant. Values followed by different letters in a column were significantly different (P < 0Æ05), using Duncan s multiple range test. Table 6 Effect of inoculation with selected plant growth-promoting rhizobacteria on growth and yield of two wheat cv. grown in field (average of four replications) No. of tillers (m )2 ) Straw yield (t ha )1 ) Grain yield (t ha )1 ) Pasban-90 Inqlab-91 Pasban-90 Inqlab-91 Pasban-90* Inqlab-91* PGPR Isolates I-year II-year I-year II-year I-year II-year I-year II-year I-year II-year I-year II-year Control 361 c 212 c 373 c 259 bc 6Æ98 c 5Æ97 c 8Æ11 bc 8Æ60 bc 3Æ60 d 2Æ91 c 4Æ18 cd 3Æ77 b Ha b 292 a 399 b 294 a 8Æ18 a 7Æ94 a 8Æ12 bc 10Æ1 a 4Æ14 c 3Æ48 a 4Æ25 c 3Æ90 b Ha b 288 a 427 a 255 bc 7Æ93 b 8Æ04 a 8Æ94 a 8Æ90 b 4Æ40 b 3Æ21 b 4Æ97 a 4Æ23 a Ha b 234 b 368 c 265 b 8Æ16 a 7Æ30 b 7Æ58 c 8Æ50 bc 4Æ37 b 3Æ07 bc 3Æ97 d 3Æ67 b Ha a 280 a 397 b 244 c 8Æ20 a 7Æ97 a 7Æ72 c 7Æ80 d 4Æ59 a 3Æ18 b 4Æ55 b 3Æ61 b *Wheat cv. Values followed by different letters in a column were significantly different (P < 0Æ05), using Duncan s multiple range test. tested, and isolate Ha30 produced the highest number of tillers in case of cv. Pasban-90, which was 15Æ5% greater than uninoculated control. Three isolates (Ha21, Ha22 and Ha30) produced significantly more number of tillers (ranging from 6Æ4 to 14Æ5%) in case of cv. Inqlab-91 compared with uninoculated control. Data regarding straw yield revealed that all the four PGPR isolates significantly increased straw yield of cv. Pasban-90 ranging from 13Æ6 to 17Æ2% over uninoculated control. In case of cv. Inqlab-91, the isolate Ha22 gave the highest straw yield (10Æ2% over uninoculated control). Inoculation also caused significant increases in grain yield of cv. Pasban-90, which ranged from 15Æ0 to 27Æ5%, over uninoculated control. In cv. Inqlab-91, increases in grain yield were up to 18Æ9% over uninoculated control in response to PGPR inoculation. The most effective isolates in promoting grain yields of cv. Pasban-90 and Inqlab-91 were Ha30 and Ha22, respectively. During the second year, there were significant differences in tillering of both the cv. due to inoculation with PGPR isolates. The most effective isolate was Ha21, which produced the highest tillers (up to 37Æ0% increase over uninoculated control) in cv. Pasban-90 and Inqlab-91. Effects of inoculation on straw yields were also significant in both cv. Cultivar Pasban-90 responded more promisingly as all the isolates applied as inocula caused increase in straw yields ranging from 22Æ3 to 34Æ7% over uninoculated control. In case of cv. Inqlab-91, only one isolate (Ha21) gave significantly higher straw yield (17Æ4%) than uninoculated control. PGPR isolate Ha21 caused maximum enhancement in the grain yield of cv. Pasban-90 (19Æ6%), while isolate Ha22 was the most effective promoter of grain yield (12Æ2% greater than uninoculated control) in cv. Inqlab-91. DISCUSSION In this study, two approaches were simultaneously employed to select effective PGPR for wheat to be used as inocula in pot and field trials. The approaches include screening of rhizobacteria for in vitro auxin biosynthesis and for their growth-promoting activity under gnotobiotic (axenic) conditions. In general, it is observed that isolates belonging to group H (more effective auxin producers) had more promising effects on the inoculated plant seedlings, compared with group L and M (least and medium auxin

7 SCREENING OF PGPR 479 producers, respectively). There was significant linear correlation between auxins produced by rhizobacteria in vitro and growth of wheat seedlings (particularly root and shoot weights) under gnotobiotic conditions. This may imply that auxins produced by PGPR isolates caused improvement in root system, resulting in more biomass production. However, other mechanisms of action through which PGPR influence plant growth can not be ruled out. Okon and Vanderleyden (1997) suggested that the secretion of plant growth-promoting substances by the bacteria could be responsible for the beneficial effects of PGPR. Glick (1995) also viewed that the mechanism most commonly invoked to explain the various effects of PGPR on plants is the production of phytohormones, and IAA may play the most important role in plant growth promotion. Under gnotobiotic conditions, Noel et al. (1996) demonstrated the direct involvement of the plant growth regulators including, IAA modifying the growth of canola and lettuce. The results of our study suggest that simultaneous screening of rhizobacteria for in vitro auxin production and growth promotion under axenic conditions is a good tool to select effective PGPR for bio-fertilizer development biotechnology. In our pot and field experiments, it was observed that inoculation with selected isolates of PGPR significantly promoted growth and yield of different cv. of wheat under nonaxenic conditions. In general, inoculation resulted in early seedling growth and development in pots. Our results are in line with the findings of Dobbelaere et al. (2001) who assessed the inoculation effect of PGPR Azospirillum brasilense on growth of spring wheat. They observed that inoculated plants resulted in better germination, early development and flowering and an increase in dry weight of both the root system and the upper plant parts. There was a positive correlation between the increase in yield and the improvement of root development. Similarly, promotion in plant height, number of tillers, plant dry weight and grain yields of various crop plants in response to inoculation with PGPR were reported by other workers (Chen et al. 1994; Khalid et al. 1997; Biswas et al. 2000a,b; Hilali et al. 2000, 2001). It is noteworthy that the strain, which produced higher amount of auxin in nonsterilized soil also significantly promoted yield of cv. Inqlab-91, in both the field trials. This may imply that this strain had more competitive ability to survive and affect the growth of inoculated plants in the presence of indigenous microflora. In some of the cases, rhizobacterial inoculation had negative effects on different growth and yield parameters of wheat. This might be due to production of some kind of phytotoxins that inhibited the growth of inoculated plants (Brown and Rovira 1999). In this study, both the cv. tested responded differently to inoculation with different rhizobacterial isolates. This variation in response to inoculation might be due to genetic make-up of different varieties/cv. or plant species. Different crops and varieties or species might produce different types of root exudates, which could support the activity of inocula and/or serve as substrate(s) for the formation of biologically active substances by the inocula (Frankenberger and Arshad 1995; Dazzo et al. 2000). Nowak (1998) reported that the benefits of bacterization depended on plant species, cv. and growth conditions. The degree to which the inoculation imparts benefits to plant growth can vary with variety, cultural conditions and PGPR strains. Work on identification of selected PGPR is in progress. As all the four selected PGPR had promising positive effects on growth and yield parameters of wheat grown under natural conditions, this supports the premises of simultaneous use of both approaches including in vitro auxin production and growth promotion under axenic conditions for selection of effective PGPR. REFERENCES Amara, M.A.T. and Dahdoh, M.S.A. (1997) Effect of inoculation with plant growth promoting rhizobacteria (PGPR) on yield and uptake of nutrients by wheat grown on sandy soil. Egyptian Journal of Soil Science 37, Arshad, M. and Frankenberger, W.T. Jr. (1998) Plant growth regulating substances in the rhizosphere: microbial production and functions. Advances in Agronomy 62, Asghar, H.N., Zahir, Z.A., Arshad, M. and Khaliq, A. (2002) Relationship between in vitro production of auxins by rhizobacteria and their growth promoting activities in Brassica juncea. L.Biology and Fertility of Soils 35, Ayers, R.S. and Westcot, D.W. (1985) Water Quality for Agriculture. FAO Irrigation and Drainage Paper 29 (Rev. 1). Rome, Italy: FAO. Barazani, O. and Friedman, J. (1999) Is IAA the major root growth factor secreted from plant-growth-mediating bacteria? Journal of Chemical Ecology 25, Biswas, J.C., Ladha, J.K. and Dazzo, F.B. (2000a) Rhizobia inoculation improves nutrient uptake and growth of lowland rice. Soil Science Society of America Journal 64, Biswas, J.C., Ladha, J.K., Dazzo, F.B., Yanni, Y.G. and Rolfe, B.G. (2000b) Rhizobial inoculation influences seedling vigor and yield of rice. Agronomy Journal 92, Brown, G.D. and Rovira, A.D. (1999) The rhizosphere and its management to improve plant growth. Advances in Agronomy 66, Chen, Y., Mei, R., Lu, S., Liu, L. and Kloepper, J.W. (1994) The use of yield increasing bacteria as plant growth promoting rhizobacteria in Chinese agriculture. In Management of Soil Born Diseases ed. Gupta, V.K. and Utkhede, R. pp New Dehli, India: M/S Narosa Pub. House. Dazzo, F.B., Yanni, Y.G., Rizk, R., de Bruijn, F.J., Rademaker, J., Squartini, A., Corich, V., Mateos, P. et al. (2000) Progress in multinational collaborative studies on the beneficial association between Rhizobium leguminosarum bv. trifolii and rice. In The Quest

8 480 A. KHALID ET AL. for Nitrogen Fixation in Rice ed. Ladha, J.K. and Reddy, P.M. pp Philippines: Los Banos. Dobbelaere, S., Croonenborghs, A., Thys, A., Ptacek, D., Vanderleyden, J., Dutto, P., Labendera-Gonzalez, C., Caballero-Mellado, J. et al. (2001) Response of Agronomically important crops to inoculation with Azospirillum. Australian Journal of Plant Physiology 28, Duncan, D.B. (1955) Multiple range and multiple F- test. Biometrics 11, Frankenberger, W.T. Jr. and Arshad, M. (1995) Phytohormones in Soil: Microbial Production and Function, pp NY, USA: Marcel Dekker Inc. Glick, B.R. (1995) The enhancement of plant growth by free-living bacteria. Canadian Journal of Microbiology 41, Hilali, A., Przrost, D., Broughton, W.J. and Antoun, A. (2000) Potential use of Rhizobium leguminosarum bv. trifolii as plant growth promoting rhizobacteria with wheat. Abstract. 17th North American Conf. on Symbiotic Nitrogen Fixation. Laval University, Quebec, Canada, July Hilali, A., Przrost, D., Broughton, W.J. and Antoun, A. (2001) Effects de I inoculation avec des souches de Rhizobium leguminosarum bv. trifolii sur la croissance du bl e dans deux sols du Marco. Canadian Journal of Microbiology 47, Hoagland, D.R. and Arnon, D. (1950) The Water Culture Methods for Growing Plants Without Soil. Circular. No. 347, pp. 39. Davis: California Agriculture Experiment Station. Khalid, A., Arshad, M., Zahir, Z.A. and Khaliq, A. (1997) Potential of plant growth promoting rhizobacteria for enhancing wheat yield. Journal of Animal and Plant Sciences 7, Kloepper, J.W., Scher, F.M., Laliberte, M. and Tipping, B. (1986) Emergence-promoting rhizobacteria: description and implications for agriculture. In Iron, Siderophores and Plant Disease ed. Swinburne, T.R. pp Plenum, NY: CRC Press Inc. FL, USA: Boca Raton. Kravchenko, L.V., Borovkov, A.V. and Pshikvil, Z. (1991) The possibility of auxin biosynthesis in wheat rhizosphere by associated nitrogen-fixing bacteria. Mikrobiologiya 60, Martens, D.A. and Frankenberger, W.T. Jr. (1991) On-line solid-phase extraction of soil auxins produced from exogenously-applied tryptophan with ion-suppression reverse phase HPLC analysis. Chromtaographia 32, Martens, D.A. and Frankenberger, W.T. Jr. (1994) Assimilation of exogenous 2 14 C-indole acetic acid and 3 14 C-tryptophan exposed to the roots of three wheat varieties. Plant and Soil 166, Noel, T.C., Cheng, C., Yost, C.K., Pharis, R.P. and Hynes, M.F. (1996) Rhizobium leguminosarum as a plant-growth rhizobacterium: direct growth promotion of canola and lettuce. Canadian Journal of Microbiology 42, Nowak, J. (1998) Review benefits of in vitro ÔÔbacterizationÕÕ of plant tissue cultures with microbial inoculants. In vitro Cellular and Developmental Biology of Plant 34, Okon, Y. and Vanderleyden, J. (1997) Root associated Azospirillum sp. can stimulate plants. American Society for Microbiology News 63, Poi, S.C. and Kabi, M.C. (1979) Effect of Azotobacter inoculation on growth and yield of jute and wheat. Indian Journal of Agriculture Sciences 49, Sarwar, M. and Kremer, R.J. (1995a) Determination of bacterially derived auxins using a microplate method. Letters of Applied Microbiology 20, Sarwar, M. and Kremer, R.J. (1995b) Enhanced suppression of plant growth through production of L-tryptophan-derived compounds by deleterious rhizobacteria. Plant and Soil 172, Sarwar, M., Arshad, M., Martens, D.A. and Frankenberger, W.T. Jr. (1992) Tryptophan-dependent biosynthesis of auxins in soil. Plant and Soil 147, Steel, R.G.D. and Torrie, J.H. (1980) Principles and Procedures of Statistics. 2nd edn. NY, USA: McGraw Hill Book Co. Inc. Wollum-II, A.G. (1982) Cultural methods for soil microorganisms. In Methods of Soil Analysis: Chemical and Microbial Properties, 2nd Edn. ed. Page, A.L., Miller, R.H. and Keeny, D.R. pp Madison-Wisconsin: American Society of Agronomy and Soil Science Society of America.