TEXT

Top Abstract Text References

Auxins constitute a class of phytohormones that play important roles in the coordination of plant growth and development. Indole-3-acetic acid (IAA), the most abundant naturally occurring auxin, has been implicated in regulating a variety of developmental and cellular processes such as cell extension, cell division, vascular differentiation, root formation, apical dominance, and tropisms (18). Regulation of these processes by auxin is believed to involve auxin-induced changes in gene expression (1, 31). Over the past 10 years, a number of plant genes that are transcriptionally induced by auxin, and that may play roles in one or more of these processes, have been cloned and further characterized (28, 31). The signal transduction pathways leading to the auxin-mediated gene induction in plants, however, are still not well understood (13, 26).

Besides plants, many soil and rhizosphere bacteria, including phytopathogenic, epiphytic, and plant growth-stimulating bacteria, also produce IAA. IAA biosynthesis in these bacteria has been shown to occur through different biosynthetic pathways (6, 20). By means of feeding experiments with tritiated putative IAA precursors, we have previously demonstrated the existence of multiple routes for IAA biosynthesis in the plant growth-promoting rhizobacterium Azospirillum brasilense (24). One of the pathways was identified as the indole-3-pyruvic acid (IPyA) pathway (L-tryptophan [Trp]IPyAindole-3-acetaldehydeIAA) by cloning of the A. brasilense ipdC gene encoding an IPyA decarboxylase (5). In addition, two amino acid aminotransferases that catalyze the transamination of Trp, the first step of this pathway, were also purified (29). An A. brasilense ipdC knockout mutant was found to synthesize less than 10% of the level of wild-type IAA production, indicating that the IPyA decarboxylase is a key enzyme for IAA biosynthesis in this bacterium (24). In this study, we show that the expression of the A. brasilense ipdC gene is upregulated by IAA and other auxins.

Construction of the translational ipdC-gusA fusion pFAJ64. To study the expression of the A. brasilense ipdC gene, we constructed a translational ipdC-gusA fusion. First, the promoter region of the A. brasilense ipdC gene was amplified by means of PCR with primers annealing approximately 350 bp upstream of the ipdC start codon (avbgltx1, 5'-CGCCGGATCCAAAGACGCCCATCAGGCGTC-3') and at codons 1 to 7 of the ipdC gene (avbipdC1, 5'-AGGGAGATCTCACCCGGAATCCCGAACATG-3'). The proofreading Vent DNA polymerase (New England Biolabs, Beverly, Mass.) was used to increase the fidelity of DNA synthesis. The amplified fragment was flanked by BamHI (avbgltx1) and BglII (avbipdC1) recognition sites (underlined in the sequences of the primers). Following digestion with BamHI and BglII, the fragment was then cloned into the BamHI site of pFAJ1171, a pUC18 derivative containing the promoterless gusA gene from pBI101.3 (15) on a 2-kb BamHI-EcoRI fragment. The orientation of the insert in the transformants was determined by PCR with the primer avbgltx1 and an internal gusA primer (5'-GATTTCACGGGTTGGGGTTTCT-3'). The DNA sequence of the 350-bp promoter fragment was verified by DNA sequence analysis using the primers avbgltx1 and avbipdC1. Nucleotide sequence analysis using the internal gusA primer confirmed that the fusion was in frame. Finally, the BamHI-EcoRI restriction fragment of approximately 2.4 kb, carrying the entire ipdC-gusA fusion, was inserted in the corresponding sites of the broad-host-range plasmid pLAFR3 (30), yielding pFAJ64. pFAJ64 was then transferred to Escherichia coli S17.1, from which it was mobilized to A. brasilense Sp245 and to the Sp245 IpdC derivative FAJ009.

Expression of ipdC is cell density dependent. It was observed previously that the level of IAA biosynthesis in A. brasilense varies during bacterial growth and that the highest IAA production level is obtained during the stationary-growth phase (24). In a first experiment, expression of the ipdC-gusA fusion, pFAJ64, in the wild-type strain Sp245 and in the IpdC mutant FAJ009 was measured quantitatively during growth in complex medium (Luria-Bertani medium supplemented with 2.5 mM MgSO₄ and 2.5 mM of CaCl₂ [LB* broth]). As shown in Fig. 1, expression of the ipdC-gusA fusion in both strains clearly increases with cell density and reaches its maximum when the cells enter the stationary-growth phase. Strikingly, the maximum level of gus expression in FAJ009 was found to be reduced to approximately 60% of that of the wild-type level. Growth phase and cell density-dependent induction of bacterial genes have often been shown to be mediated by small diffusible signal molecules. In E. coli, for instance, accumulation of weak acids in stationary-phase cultures serves as a signal to activate the expression of the sigma factor, RpoS, which in turn is required for the activation of other growth phase-specific sets of genes (27). Additionally, a variety of cellular processes in bacteria, including bioluminescence in Vibrio fischeri (4, 7, 9), production of exoenzymes and antibiotics in Erwinia carotovora (2, 16), and plasmid conjugal transfer in Agrobacterium tumefaciens (22, 34), are switched on by diffusible compounds, termed autoinducers and identified as N-acyl homoserine lactones (AHLs). Activation of the specific target genes occurs when a required threshold of AHLs is attained and in this way is only achieved at high cell densities. Therefore, the observed growth phase-dependent expression of the A. brasilense ipdC gene prompted us to investigate whether an extracellular product accumulating in spent medium is involved in ipdC induction. To test this possibility, different amounts of filter-sterilized culture supernatant (0.22-µm-pore size) obtained from a stationary-growth-phase culture of A. brasilense Sp245 were added to exponentially growing cells of Sp245(pFAJ64), and -glucuronidase activity was determined after 2.5 h of additional growth. As can be concluded from Fig. 2, ipdC expression increases with increasing amounts of the stationary-phase culture supernatant. A twofold dilution of the reporter strain (2 ml of culture plus 2 ml of supernatant) resulted in a fivefold increase in -glucuronidase activity compared to the activity detected in control cultures to which no supernatant was added.

View larger version (21K): [in this window] [in a new window]   FIG. 1.   Growth and expression of ipdC in A. brasilense Sp245 and in the IpdC mutant FAJ009. Cultures of Sp245(pFAJ64) and FAJ009(pFAJ64) were grown in LB* broth. Periodically, samples were taken to determine the expression of the ipdC-gusA fusion, pFAJ64. Quantitative analysis of -glucuronidase activity was carried out in microtiter plates with p-nitrophenyl--D-glucuronide as substrate (14). Growth was monitored by measuring the OD600. The data presented are from one representative experiment, which was confirmed twice in additional independent experiments. Triangles, Sp245(pFAJ64); squares, FAJ009(pFAJ64); open symbols, cell density (OD600); solid symbols, -glucuronidase activity in Miller units (17).

View larger version (20K): [in this window] [in a new window]   FIG. 2.   Induction of the ipdC-gusA fusion by spent culture supernatants. An exponential-phase culture of Sp245(pFAJ64) (OD600 = 0.6) grown in LB* broth was centrifuged and resuspended in fresh LB* broth. Two-milliliter samples of this culture were then transferred to sterile test tubes, supplemented with different amounts of spent culture supernatant (ranging from 0 to 2 ml), and diluted up to a total volume of 4 ml with fresh LB* broth. After 2.5 h of growth at 30°C, the cultures were assayed for -glucuronidase activity as detailed in the legend of Fig. 1. The spent culture supernatants were obtained from late-stationary-phase cultures (OD600 = 2.0) of Sp245 (solid bars) or FAJ009 (stippled bars) grown in LB* broth. Open bar, control culture to which no supernatant was added. All data are the means from three replicates. Error bars denote the standard deviations.

Expression of ipdC is upregulated by IAA. The difference in ipdC expression levels between the wild type and the IpdC mutant and the observation that spent culture supernatant of the IpdC mutant does not efficiently induce ipdC expression led us to speculate that the end product of the biosynthetic pathway, IAA, could be the inducing compound. It has been shown previously that IAA accumulates in the supernatant of wild-type culture (up to 200 µM in a stationary-growth-phase culture in LB* broth) but is reduced to 10% of the wild-type level in the supernatant of the IpdC mutant (24). The idea of IAA as inducing compound was further tested by examining the effect of exogenously added IAA on ipdC expression. Sp245(pFAJ64) and FAJ009(pFAJ64) cultures at different growth phases (starter cultures) were supplemented with 1 mM IAA and, after 3.5 h of additional growth, compared for ipdC-gusA induction with cultures to which no IAA was added. As can be concluded from Fig. 3, in all growth phases, both for Sp245 and FAJ009, a higher expression level of pFAJ64 was found in cells exposed to exogenously added IAA compared to the level in untreated cells. As expected, upregulation by exogenously added IAA (i.e., the difference in the expression levels between a culture supplemented with IAA and an untreated culture) was most obvious in the IpdC mutant background and was growth phase dependent. The difference in the extent of upregulation at the various growth phases and between the mutant and the wild type may be due to differences in endogenous IAA levels in the starter cultures. For Sp245, a threefold upregulation was detected in early-exponential-phase cells (optical densities at 600 nm [OD₆₀₀] of 0.1 and 0.4). Upregulation then gradually decreased in the mid and late-exponential phases (1.4-fold increase at OD₆₀₀ = 1.1 and 1.4) and stationary-phase cultures (1.1 fold increase at OD₆₀₀ = 1.9) because of the higher endogenous IAA production in the corresponding starter cultures. In contrast to Sp245 and concomitant with the lower IAA production capacity of the IpdC mutant, a fourfold upregulation could still be detected in late-exponential-phase FAJ009 culture (OD₆₀₀ = 0.9). The lowest concentration of exogenously added IAA for which significant ipdC induction was observed in an exponential-phase culture of FAJ009 was 10 µM (Table 1).

View larger version (13K): [in this window] [in a new window]   FIG. 3.   Effect of exogenously added IAA on ipdC expression in Sp245 and the IpdC mutant, FAJ009. To obtain cultures at various cell densities of Sp245(pFAJ64) (A) and FAJ009(pFAJ64) (B) different flasks containing 100 ml of LB* medium were inoculated with different volumes of preculture and grown overnight at 30°C. From each culture, six samples of 3 ml were transferred to sterile test tubes. Three test tubes of each series were supplemented with IAA to a final concentration of 1 mM (solid bars); to the remaining three test tubes no IAA was added (open bars). After 3.5 h of additional growth at 30°C, -glucuronidase activity was assayed as detailed in the legend of Fig. 1. Data are the means from the three replicates. Error bars denote the standard deviations.

View larger version (67K): [in this window] [in a new window]   FIG. 4.   Northern blot analysis of RNA extracted from cells of A. brasilense wild-type and IpdC mutant strains. (A) Blot of total RNA (10 µg per lane) isolated from A. brasilense IpdC mutant (lane 1) and wild-type strains (lanes 2 to 5). RNA was visualized by epi-illumination with UV light. Cells were harvested at an OD600 of 0.9 (lanes 1 and 3), 0.5 (lane 2), or 0.4 (lanes 4 and 5). To measure the effect of IAA on the transcription of the ipdC gene, the wild-type culture at OD600 of 0.4 was divided into two and further incubated for 3 h at 30°C in the absence (lane 4) or presence of exogenously added IAA at 1 mM (lane 5). The positions of the 16S and 23S ribosomal RNAs are indicated at the right. (B) RNA blot analysis of the same samples as in panel A after hybridization with a digoxygenin-labeled ipdC-derived riboprobe. The size of the ipdC transcript (in kilobases) is indicated at the right and was determined relative to RNA standards that were electrophoresed in the same gel.

Different auxins induce ipdC expression. IAA-inducible genes in higher plants have been shown to respond to, besides IAA, synthetic auxins such as naphthaleneacetic acid (NAA) and chlorophenoxy acids (28). To evaluate whether the A. brasilense ipdC gene could also be induced by other auxins, exponential-growth-phase cultures of FAJ009(pFAJ64) (OD₆₀₀ = 0.4) were supplemented with the naturally occurring auxin indole-3-butyric acid (IBA) or with the synthetic auxins NAA, 2,4-dichlorophenoxypropionic acid (2,4-DP), or p-chlorophenoxyacetic acid (4-CPA) and compared for ipdC induction. As indicated by the data in Table 1, the three tested synthetic auxins, NAA, 2,4-DP, and 4-CPA, were found to upregulate ipdC expression. Under the conditions tested, induction of the ipdC-gusA fusion by all three synthetic auxins was as strong as the induction observed with IAA. In contrast, addition of IBA to the cultures did not affect ipdC expression. In plants, the conversion of IBA to IAA has been reported (10, 11), and it is not yet clear whether IBA is itself an auxin or whether it exerts its auxin activity through its conversion to IAA (3). The observation that the A. brasilense ipdC gene is not upregulated by IBA might therefore be attributed to the absence of enzymes catalyzing the conversion of IBA into IAA in Azospirillum. The addition of the two IAA conjugates, IAA ethyl ester and IAA-L-phenylalanine, to the FAJ009(pFAJ64) culture had no effect on ipdC expression (Table 1). This might be due to the lack of intracellular transformation of these compounds into active auxins, although it cannot be excluded that these conjugates fail to induce ipdC gene expression because they are not taken up by A. brasilense cells. Treatment of the FAJ009(pFAJ64) culture with similar concentrations of Trp, acetic acid, and propionic acid also did not enhance ipdC induction (Table 1), indicating that the upregulation of ipdC expression by IAA and the synthetic auxins is auxin specific and not only caused by the acid group of the auxin molecules. In control experiments, incubation of FAJ009 containing pFAJ31.13, a pLAFR1 derivative carrying a fusion between a constitutive A. brasilense promoter and gusA (33), with IAA or NAA did not affect -glucuronidase activity (Table 1).

General conclusion. Evidence is presented here for the upregulation of expression of the A. brasilense IAA biosynthetic gene, ipdC, by IAA and synthetic auxins. This positive feedback regulation by IAA is responsible for the increasing ipdC transcription levels during growth of an A. brasilense culture, with the highest expression level of the ipdC gene observed in the stationary-growth phase. Reduced IAA accumulation in stationary-phase cells of the IpdC mutant strain results in a lower ipdC expression maximum in this strain compared to that of the wild type. In contrast to negative transcription control of biosynthetic genes by the end product formed (19, 23), autoinduction of a biosynthetic pathway is rather exceptional. Other genes, besides the A. brasilense ipdC gene, that have been demonstrated to be autoinduced are the AHL biosynthetic genes found in various gram-negative bacteria (12) and the genes involved in the production of the siderophores yersiniabactin in Yersinia enterocolitica (21) and pyochelin in Pseudomonas aeruginosa (25). Until now auxin-responsive genes have only been identified in plants (1, 28, 31). The role of many of these auxin-inducible genes in plant developmental processes is not yet fully understood. This study presents data indicating for the first time the induction of a bacterial gene by auxin. This finding suggests that proteins involved in IAA perception and signal transduction are present not only in higher plants but also in bacteria.