Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Casavant, N. C. Articles by Halverson, L. J. Search for Related Content PubMed PubMed Citation Articles by Casavant, N. C. Articles by Halverson, L. J. Agricola Articles by Casavant, N. C. Articles by Halverson, L. J.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, July 2002, p. 3588-3596, Vol. 68, No. 7
0099-2240/02/$04.00+0     DOI: 10.1128/AEM.68.7.3588-3596.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Site-Specific Recombination-Based Genetic System for Reporting Transient or Low-Level Gene Expression

N. Carol Casavant,¹ Gwyn A. Beattie,² Gregory J. Phillips,² and Larry J. Halverson^1,2*

Departments of Agronomy,¹ Microbiology, Iowa State University, Ames, Iowa 50011²

Received 17 December 2001/ Accepted 11 April 2002

Top Abstract Introduction Materials and Methods Results Discussion References

 
We report here the construction, characterization, and application of a plasmid-based genetic system that reports the expression of a target promoter by effecting an irreversible, heritable change in a bacterial cell. This system confers strong repression of the reporter gene gfp in the absence of target promoter expression and utilizes the site-specific recombination machinery of bacteriophage P22 to trigger high-level reporter gene expression in the original cell and its progeny after target gene induction. We demonstrate the effectiveness of this genetic system by tailoring it to indicate the availability of arabinose to the biological control agent Enterobacter cloacae JL1157 in culture and in the barley rhizosphere. The presence of bioavailable arabinose triggered the production of P22 excisionase and integrase from the reporter plasmid pAraLHB in JL1157, and this led to excision of the cI repressor gene, which is flanked by att sites, and the subsequent irreversible expression of gfp in the original cell and in its progeny. In culture, nearly 100% of an E. cloacae JL1157(pAraLHB) population expressed gfp after exposure to 6.5 to 65 µM arabinose for 3 h. We used this biosensor to demonstrate that arabinose was released from the seeds of several legumes and grass species during germination and from roots of barley seedlings grown hydroponically or in soil. When introduced into microcosms containing barley, the biosensor permitted the localization of arabinose along the roots. Arabinose was present near the root-seed junction and on the seminal roots but was not detected at the root tips. This recombination-based reporter system should be useful for monitoring bacterial exposure to transient or low levels of specific molecules directly in the environment.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Transcriptional fusions between a target gene and a reporter gene are a powerful tool for investigating bacterial gene expression in natural habitats and for detecting specific signal molecules in the environment (1, 5, 18, 23). Transcriptional fusions that are used for these purposes are often limited in their sensitivity, however, since low levels of reporter gene expression can be difficult to detect in cells from natural habitats, particularly if reporter gene expression is not assessed immediately after induction (23). Furthermore, such fusions do not effectively report transient expression, such as would result from encountering the inducer for only a limited period of time. The ability to detect whether a target gene was ever expressed, even at a low level, in a natural habitat would often be useful. For example, knowledge of the presence or absence of specific nutrients and environmental pollutants, both of which may be ephemeral or be present in only low concentrations, is important for understanding the ecology and bioremediation potential of an organism. For the same reasons, exposure of microbes to plant- or animal-derived compounds that induce specific microbial functions, including traits related to virulence, symbiosis, or biocontrol, may be difficult to detect. A genetic system that strongly represses the expression of a reporter gene when the target promoter is not being expressed and irreversibly induces a high level of reporter gene expression when the target promoter is expressed, even at low levels, would facilitate the detection of signal molecules in the natural environment.

To this end, we developed a genetic system that reports the expression of a target promoter by effecting an irreversible, heritable change in the biosensor cell; this system confers strong repression of a reporter gene in the absence of target promoter expression and high-level reporter gene expression in the original cell and its progeny after even a low level of target gene induction in the original cell. This system used the site-specific recombination elements of the Escherichia coli bacteriophage and Salmonella enterica serovar Typhimurium bacteriophage P22 and is related in concept to the recombination in vivo expression technology (RIVET) that has been used as a promoter trap to identify bacterial genes involved in virulence (4, 5, 21). As currently designed, RIVET requires that cells be reisolated from a habitat and screened for loss of an antibiotic resistance marker. The genetic system described in this report used the reporter gene gfp, encoding the green fluorescent protein (GFP) from Aequoria victoria (7). GFP is particularly useful as a reporter since, in addition to reporting gene activity based simply on inspecting cells or colonies for fluorescence, it permits visualization of the spatial patterns of gene expression directly in the habitat.

We used the rhizosphere as a model habitat to evaluate the efficacy of this recombination-based biosensor for detecting low amounts of a transiently available compound in a complex environment. The spermosphere and rhizosphere are nutrient-rich environments relative to the bulk soil (9). Seeds, as well as living and dying root tissue, release relatively labile sugars, amino acids, organic acids, and polysaccharides (16, 32, 33). The amount and composition of seed and root exudates are likely to differ depending on the age and physiology of the plant and on the sites that are examined on the root (16), particularly relative to the site of seed germination. Thus, specific components in these exudates may be produced only transiently or may be used by the colonizing microflora at such a rapid rate that their availability is only transient or there are only low levels available. Because the temporal and spatial distribution of specific nutrients may be complex, identifying whether an organism has access to a specific nutrient in the rhizosphere is challenging. The ability of a microorganism to access and utilize specific nutrients, however, may be critical to both its fitness in the rhizosphere and its relative competitiveness within the rhizosphere community.

We developed a bacterial sensor to detect available arabinose by using the biocontrol agent Enterobacter cloacae JL1157 (8, 24). Arabinose is a component of seedling exudates (31) and is a constituent of many cell wall polymers of grasses and thus may be present in root-derived compounds through exudation, secretion, or, indirectly, plant cell wall damage (6). It is likely to be one of the many carbohydrates that can be used by E. cloacae during spermosphere and rhizosphere colonization. We constructed an arabinose-inducible, recombination-based bacterial biosensor by using the E. coli L-arabinose-inducible araBAD promoter (P_BAD), the associated AraC regulator (14, 19, 27), and various site-specific recombination elements from bacteriophage and P22. We describe here the construction of this system and demonstrate its use in detecting the availability of arabinose to E. cloacae in the barley rhizosphere.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Bacterial strains, plasmids, and growth conditions.
Bacterial strains and plasmids used in this study are described in Table 1. E. coli and E. cloacae were routinely cultivated on Luria agar or Trypticase soy agar (TSA), respectively. Media were supplemented with antibiotics when appropriate at the following concentrations: ampicillin, 50 µg/ml; chloramphenicol, 50 µg/ml; kanamycin, 50 µg/ml; and rifampin, 20 µg/ml. For the arabinose dose-response curves and inhibitor studies, E. cloacae was grown in a semidefined medium, yeast extract (YE)-succinate broth or YE-glycerol broth containing 0.5 g of NH₄Cl, 1.75 g of Na₂HPO₄ · 7H₂O, 1.38 g of KH₂PO₄, 0.2 g of YE, and 40 ml of Huntner mineral solution (37)/liter and either 7 mM succinate or 26 mM glycerol, and the pH was adjusted to 6.8. Solid media contained 15 g of agar/liter.

View this table: [in this window] [in a new window]

TABLE 1. Bacterial strains and plasmids

Genetic techniques.
Standard methods were used for DNA isolation, restriction digests, ligations, and transformation (34). Restriction site modifications were performed by treating 10 pmol of oligonucleotide adapters with T4 kinase, heating the mixtures to 80°C, and then slowly cooling them to room temperature prior to ligation. PCR amplification targets, primers, and amplification conditions are described in Table 2. Plasmids were mobilized into E. cloacae by triparental mating using the mobilization function of plasmid pRK2073 (11).

View this table: [in this window] [in a new window]

TABLE 2. PCR protocols

Biosensor construction.
The components of the pAraLHB biosensor were assembled into pPROBE-KT (26), a broad-host-range promoter probe vector, by using amplified products as the source of key elements (see Table 2 for a description of the PCR protocols) and standard recombinant DNA techniques. The strategy employed in pAraLHB construction is shown in Fig. 1 . First, we generated a fusion between the arabinose-responsive promoter P_BAD and a promoterless copy of the P22 operon containing the integrase (int) and excisionase (xis) genes (Fig. 1A). This was accomplished by amplifying the xis,int operon from pXINTP22 (29), which included a ribosomal binding site, and cloning it into the EcoRI-SalI site of pBAD18 (14) to generate pBAD-xis,int. Second, the bacteriophage promoter P_L, including the operator sites, was amplified from EMBL3 (Table 2) and cloned into the BamHI site of pPROBE-KT to generate pVP_Lgfp (not shown). Third, we amplified a 3.2-kb portion of pBAD-xis,int that included the transcriptional activator araC, the P_BAD-xis,int fusion, and the rrnB transcriptional terminators, and cloned this fragment into the KpnI site of pVP_Lgfp to create pAra (Fig. 1B). Fourth, we generated a cassette in which the cI repressor gene was flanked by attachment (att) sequences that are recognized by bacteriophage P22 excisionase and integrase proteins (Xis and Int, respectively) (Fig. 1C). This was accomplished by amplifying attP22, which is recognized by the bacteriophage P22 integrase, and introducing it into pGEM-T to form pT-attP22. Cointegration of pT-attP22 and plasmid pCDP22SK, which contained attP, was facilitated by pINTKanP22 by the technique described by Platt et al. (30). Plasmid pINTKanP22 mediated cointegrate formation via its constitutive production of integrase, and it was lost via growth at a nonpermissive temperature (42°C). The 1.1-kb XbaI fragment containing the R6Kori in the pT-att-CD-R6K cointegrate was replaced with a 1.2-kb EcoRV-XbaI fragment containing the cI repressor gene from pCDP22SK-cI (29) to create pT-att-cI. Fifth, a 4.5-kb ClaI fragment containing the cI cassette was excised from pT-att-cI and was inserted into the ClaI site of pAra to generate pAraLHB (Fig. 1D).

View larger version (14K): [in this window] [in a new window]

FIG. 1. Schematic illustration of the arabinose life history biosensor (pAraLHB) construction. (A) Construction of PBAD-xis,int fusion. The amplified 1.4-kb fragment containing xis,int (Table 2) was ligated into the EcoRI and SalI sites of pBAD18 adjacent to the PBAD promoter. (B) Diagram of pAra. A 0.2-kb amplification product (Table 2) containing the promoter PL was inserted into the BamI site of pPROBE-KT. A 3.2-kb amplification product (Table 2) containing araC and PBAD-xis,int and an rrnB terminator, which was generated by using pBAD-xis,int as a template, was inserted into the KpnI site of pPROBE-KT. (C) Construction of the cI cassette. The attP22 and attP sites were obtained from pT-attP22 and pCDP22SK, respectively, to generate the pT-att-CD-R6K cointegrate by using P22 integrase. pT-att-cI was derived by deleting a 1.1-kb XbaI fragment that included the R6Kori, filling in the XbaI sites, and inserting a 1.2-kb fragment containing the cI gene from pCDP22SK-cI. (D) Diagram of pAraLHB. A 4.2-kb ClaI fragment from pT-att-cI, which contained the attL-cI-Cmr-attR cassette, was inserted into the ClaI site of pPROBE-KT. Abbreviations: RBS, ribosomal binding site; R, EcoRI; S, SalI; C, ClaI; X, XbaI; CmR, chloramphenicol resistance gene; R6K, R6K origin of replication; T(1) and T(4), E. coli rrnB T1 terminator(s); rrnBT1,2, E. coli rrnB T1 and T2 terminators.

In vitro testing of the biosensor.
E. cloacae JL1157(pAraLHB) was cultivated on TSA supplemented with antibiotics. Exponential-phase cultures were prepared by inoculating single colonies into fresh YE-succinate broth and incubating the cultures for 1 to 2 h at 28°C in an orbital shaking incubator set at 200 rpm prior to the addition of arabinose. The cultures were exposed to 6.6 x 10^-7 to 6.6 x 10^-3 M arabinose for various periods of time (1, 3, or 6 h), and subsamples were serially diluted in phosphate-buffered saline prior to plating onto TSA-kanamycin by using a spiral plater (Spiral Biotech, Bethesda, Md.). Plates were incubated at 28°C for 1 to 2 days, the colonies were enumerated, and the proportion of the colonies producing GFP was determined visually by using a UV handheld lamp (_ex = 365 nm) or a Dark Reader handheld lamp (_ex = 420 to 500 nm; Clare Chemical Research, Denver, Colo.) and filtered glasses that collected emission wavelengths of >520 nm, which facilitated the visual detection of green fluorescence. The same procedure was used to evaluate the responsiveness of stationary-phase JL1157(pAraLHB) biosensor cells to arabinose except that the cells were cultivated for 24 h prior to the addition of arabinose.

In situ application of the biosensor.
Clarion soil (Typic Hapludoll; fine-loamy, mixed mesic) was sieved through a 2-mm mesh and stored at 4°C until use. JL1157(pAraLHB) was grown overnight on 1/5-strength TSA amended with kanamycin and rifampin; a single, nonglowing colony was restreaked, and the cells were resuspended in phosphate-buffered saline for inoculation of either soil or roots. The soil microcosm (no plants) consisted of 30 g of soil in a 50-ml screw-cap test tube that was gently packed to generate a bulk density of ca. 1.6 g/cm². In the soil-only experiments, an aliquot of JL1157(pAraLHB) was added directly to the soil to achieve ca. 10⁷ to 10⁸ CFU/g of soil with a sufficient inoculum volume to achieve a final soil water content of 15% (wt/wt). Preliminary experiments showed that E. cloacae maintained population sizes of between 10⁶ to 10⁷ CFU/g in nonrhizosphere soil over a 14-day period. For the rhizosphere experiments, barley seeds (Hordeum vulgare cv. Manchurian) were surface disinfested by soaking them in 0.5% sodium hypochlorite (vol/vol) for 20 min, were rinsed extensively with sterile water, and were germinated in the dark for 48 h on moistened filter paper. Barley seedlings were dipped in 10¹⁰ CFU of JL1157(pAraLHB)/ml cell suspensions and were grown in soil microcosms (600 g of soil in Styrofoam cups), in rhizotrons for microscopy, or in a hydroponic system. The rhizotrons for microscopy consisted of microscope slides placed over a 3-mm spacer, and the chamber was filled with soil. The edges of the chambers were sealed with masking tape. Rhizotrons were placed at a 45° angle to ensure that the roots grew along the surface of the microscope slide. The hydroponic system consisted of wrapping the seedling in sterile Whatman filter paper and placing it in a flask of half-strength plant nutrient solution (3). Plants were grown in a growth chamber set at 21°C and 99% humidity with a 12-h (light)-12-h (dark) photoperiod for 3 to 5 days.

Microscopy.
Samples were viewed on a Nikon EFD-3 epifluorescence microscope, and images were taken with a charge-coupled-device camera (SPOT Camera; Diagnostic Instruments, Sterling Heights, Mich.). GFP fluorescence was imaged by using a filter set with an excitation wavelength of 480 ± 15 nm and an emission wavelength of 535 ± 20 nm. DsRed fluorescence was imaged by using a (470 ± 20)-nm excitation and a 515-nm long-pass emission filter set. Images were cropped by using Adobe Photoshop (Adobe Systems, Mountain View, Calif.).

Statistical analyses.
Statistical analyses were performed by using JMP version 4.04 (SAS Institute, Cary, N.C.). Analysis of variance for comparing the proportion of JL1157(pAraLHB) cells that expressed gfp at different incubation times was performed on the arcsine square-root transformation of the proportion of cells that expressed gfp. The Tukey-Kramer honestly significant difference (HSD) means test (P = 0.01) was calculated by JMP software for comparisons among treatment means.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Construction of the biosensor system.
Plasmid pAraLHB (Fig. 1) was derived by inserting genes or genetic elements into the pPROBE-KT (26) vector as outlined in Materials and Methods and shown in Fig. 1. The phage promoter P_L, which was inserted into pPROBE-KT to form a P_L-gfp fusion, included the operator sites O_L1, O_L2, and O_L3 that bind the cI repressor. The arabinose-inducible promoter P_BAD of E. coli was fused ca. 25 bases from the ribosomal binding site of a promoterless copy of the bacteriophage P22 xis,int operon. The gene encoding the regulator of P_BAD, araC, was expressed constitutively under the control of its cognate promoter. In the resulting biosensor strain JL1157(pAraLHB), arabinose-mediated induction of the P_BAD-xis,int transcriptional fusion led to Xis- and Int-mediated site-specific recombination at the att sequences, causing excision of the cI cassette. Since the cI cassette does not contain an origin of replication, it is subsequently lost from the cell.

Arabinose induction of E. cloacae JL1157(pAraLHB) in vitro.
Throughout these studies, the proportion of cells in a population that was exposed to sufficient levels of arabinose to induce P_BAD-xis,int expression was determined based on the proportion of fluorescent colonies that arose after plating a sample of the population. This assay for P_BAD induction, and thus arabinose availability, was possible due to the irreversible genetic change that occurred in the induced cells. In preliminary studies, we observed that a single chromosomal copy of the cI gene blocked P_L-mediated gfp expression from a pVS1-based plasmid containing the P_L-gfp fusion (data not shown). Therefore, for high-level gfp expression to occur, sufficient quantities of both Xis and Int proteins must be produced to remove the cI cassette from all copies of the vector.

The P_L promoter was not sufficiently leaky to cause visible green fluorescence in colonies of uninduced cells. During growth in a nutrient rich medium (e.g., Luria broth and YE-glycerol) in the absence of exogenous arabinose, excision of the cI cassette occurred in only 0.5% of the population per generation, which generally resulted in excision of the cI cassette in 1 to 13% of a culture population based on colony fluorescence. The leakiness from the P_BAD promoter was reduced by a carbon source that functions as a catabolite repressor of the P_BAD promoter (14, 27). We found that this leakiness was, in fact, reduced by cultivating cells with glucose, as well as with chloramphenicol, which selected for cells that had not lost the cI cassette (data not shown). A chloramphenicol- and glucose-containing nutrient rich medium (i.e., TSA) was thus used to prepare inocula that contained a minimum number of cells that had lost the cI cassette and were irreversibly green fluorescent. We also determined that gfp expression in the fluorescent colonies after arabinose exposure was due to removal of the cI cassette since 100% of the colonies examined were chloramphenicol sensitive. Furthermore, we determined that cells obtained from non-green fluorescent colonies that had previously been exposed to arabinose exhibited loss of the cI cassette after subsequent arabinose exposure, which indicates that the failure of a cell to lose the cI cassette upon arabinose exposure was not due to a genetic change that prevented excision of the cI cassette. Doubling times (mean ± the standard error of mean [SEM]; n = 3) of JL1157(pAraLHB) cultures derived from cells that constitutively expressed gfp (42.5 ± 1.4 min) were comparable to those of JL1157(pAraLHB) cultures that did not express gfp (45.4 ± 1.0 min), indicating that high-level gfp expression did not impose a fitness cost during growth in Trypticase soy broth (TSB), and similar results were obtained when YE-succinate was the growth medium (data not shown).

To determine the rapidity with which site-specific recombination took place upon exposure to arabinose, we examined the kinetics of the acquisition of green fluorescence in cultures exposed to various concentrations of arabinose in YE-succinate broth. After exposure to 65 µM arabinose, nearly 100% of the colonies exhibited green fluorescence after 3 h, and a significant subpopulation (40 to 50%) of cells was fluorescent after exposure to only 6.5 µM arabinose for 3 h (Fig. 2A). Shorter incubation times required higher arabinose concentrations to achieve 100% of the population expressing GFP. Similar results were obtained in replicate experiments where the carbon source was glycerol (i.e., YE-glycerol; data not shown). In experiments with stationary-phase JL1157(pAraLHB) cells, a higher arabinose concentration and a longer incubation period were required for site-specific recombination to take place (Fig. 2B). The requirement for higher arabinose concentrations and prolonged incubation periods (3 h) indicates that the cell's physiological state affects its ability to respond to arabinose. This result suggests that a low-energy state, such as occurs in stationary cells, increases the threshold concentration and reduces the rate at which the arabinose-induced genetic change occurs in JL1157(pAraLHB) cells.

View larger version (21K): [in this window] [in a new window]

FIG. 2. Dose response of JL1157(pAraLHB) in YE-succinate broth amended with various concentrations of arabinose and incubated for 1, 3, or 6 h. (A) Exponential-phase cultures. (B) Stationary-phase cultures. The symbols represent the duration of arabinose exposure as follows: •, 1 h; , 3 h; , 6 h. All points represent the mean of three replicates, and the error bars represent the SEM. The background green fluorescent cells in the exponential- and stationary-phase populations were 13.4% ± 4.3% and 6.8% ± 1.2%, respectively (mean ± the SEM).

Arabinose induction of JL1157(pAraLHB) by plant exudates or in planta.
In the 2-week period after introduction of JL1157(pAraLHB) into soil microcosms, a similar percentage of fluorescent cells was recovered as in the original inoculum (12% ± 3.3% for each [mean ± SEM]; n = 3) (P < 0.05). This percentage increased to 73 to 86% of the cells exhibiting green fluorescence after 24 h of exposure to 2 mM arabinose, whereas only 12.4% ± 3.3% exhibited fluorescence in the unamended controls. JL1157(pAraLHB) residing in soil for up to 14 days after inoculation before arabinose amendment responded to arabinose by exhibiting a similar percentage of fluorescent cells (83% ± 7.5%) as a population that had been residing in soil for only 1 day before arabinose amendment (74% ± 3.5%).

We examined various seed exudates for the presence of arabinose by incubating JL1157(pAraLHB) for 6 h in a 1:1 YE-glycerol-seed exudate mixture before the exudates were plated to determine the proportion of the population that exhibited fluorescence. The proportions of E. cloacae cells that expressed gfp in the presence of seed exudates from the barley, maize, soybean, radish, cucumber, alfalfa, and the negative control treatments were 42.5% 177 ± 7.5%, 42.8% ± 1.1%, 44.7% ± 2.3%, 23.6% ± 1.1%, 35.2% ± 7.6%, 20.6% ± 1.5%, and 2.7% ± 0.8% (mean ± standard deviation; n = 2), respectively. Although it is possible that other sugars, such as glucose, which is a known component of many seed exudates (31), inhibited arabinose-mediated induction in some of the cells, the significant increase in the percentage of gfp-expressing cells in the presence of each of the seed exudates relative to the negative control provides strong evidence that arabinose was present in each of the seed exudates tested.

We examined the availability of arabinose in the barley rhizosphere by dipping 30- to 40-h-old barley seedlings in a JL1157(pAraLHB) cell suspension prior to planting them in a hydroponic system. The background population of JL1157(pAraLHB) cells that expressed gfp was determined by immediately plating excised roots that had been dipped in the cell suspension (Fig. 3). At the time of inoculation (day 0) only 2.1% ± 0.1% of the population expressed gfp, but within 24 h ca. 50% of the population expressed gfp (Fig. 3). There was no statistical difference in the proportion of JL1157(pAraLHB) cells that expressed gfp among the different sampling times based on a Tukey-Kramer HSD test (P = 0.01), except between days 3 and 4. It is possible that the decrease between days 3 and 4 resulted from population decreases in regions containing arabinose with simultaneous population increases in areas devoid of arabinose. The population sizes of JL1157(pAraLHB) in the barley rhizosphere of hydroponically grown plants 1 to 4 days after inoculation did not change significantly over time (Fig. 3). In a parallel experiment with plants grown in soil, we determined that the percentage of rhizosphere JL1157(pAraLHB) cells that expressed gfp increased from 1.3% ± 0.2% immediately after inoculation to 56.5% ± 14.6% 5 days after inoculation (mean ± SEM; n = 3). The higher variability in the soil-grown plants was likely a consequence of including in the rhizosphere samples soil particles that contained uninduced JL1157(pAraLHB) cells. The population sizes of JL1157(pAraLHB) in the barley rhizosphere 1 to 5 days after inoculation were typically 6.5 ± 0.2 log₁₀ (CFU/root) for the soil-grown plants.

View larger version (18K): [in this window] [in a new window]

FIG. 3. Detection of available arabinose in the barley rhizosphere of hydroponically grown plants with the JL1157(pAraLHB) biosensor. Symbols: •, percentage of JL1157(pAraLHB) cells that expressed gfp; , log10 CFU/root. Points represent the mean ± the SEM of four replications per sampling time.

We also examined the spatial distribution of available arabinose around barley roots by introducing a mixture of JL1157(pAraLHB) and JL1157(pVLacRed) cells onto plants either by inoculating soil with a cell suspension prior to planting a 2-day-old seedling or by dipping a barley seedling in a cell suspension prior to planting. Strain JL1157(pVLacRed), which constitutively expressed the red fluorescent protein DsRed, was included to facilitate visualizing the distribution of the inoculated bacteria. Green fluorescent bacteria were rarely seen in the soil; in contrast, red fluorescent bacteria were distributed throughout the soil (Fig. 4A). Green fluorescent bacteria were frequently observed on the rhizoplane of the seminal (primary) roots (Fig. 4B, C, and D), less frequently on or near the root hairs of any root (Fig. 4C), and rarely on the adventitious (secondary) roots. The greatest number of green fluorescent cells and colonies were located on the seminal roots near the seed-root junction (Fig. 4B and C). Green fluorescent bacteria seldom covered large areas, nor were they usually associated with root tips. Given the abundance of the red fluorescent cells on the roots and in the soil, indicating widespread distribution of the introduced bacterial cells, the abundance of the green fluorescent cells on the roots compared to the occasional green fluorescent cell in the soil indicates that arabinose was available near the root but was limited, or absent, in the soil. Whether the origin of the arabinose was the seed or the seed and the roots could not be determined.

View larger version (118K): [in this window] [in a new window]

FIG. 4. Detection of available arabinose in the rhizosphere of barley growing in nonsterile soil with the JL1157(pAraLHB) biosensor. All panels show roots in microcosms inoculated with a mixture of JL1157(pAraLHB) (green fluorescence) and JL1157(pVLacRed) (constitutive red fluorescence), except for the roots in panel B, which were inoculated only with JL1157(pAraLHB). Green fluorescence indicates the presence of arabinose. Bars, 20 µm. (A) Red fluorescent bacteria in soil adjacent to a root hair. No green fluorescent bacteria were visible in the soil or along the root hair. (B) Green fluorescent bacteria forming microcolonies on the seminal root near the seed-root junction (the image obtained was taken 4 days after inoculation). (C) Red and green fluorescent bacteria on a seminal root close to the seed-root junction. (D) Red and green fluorescent bacteria on the same root as shown in panel C, but farther down the root axis.

Top Abstract Introduction Materials and Methods Results Discussion References

 
This report describes the construction of a genetic system that effectively reports the presence or absence of gene expression in response to transient and/or low levels of an inducing signal. In this system, the production of a site-specific recombinase is controlled by the promoter of a gene of interest, if gene expression itself is being studied, or a target signal-dependent promoter, if the presence of a target signal is being studied. Expression of this promoter, even at low levels, induces a site-specific recombination event, and this recombination results in the permanent and heritable loss of a repressor gene that is necessary for blocking expression of the reporter gene, gfp. This system consists of four components: (i) a fusion between the promoter of interest and a promoterless copy of the bacteriophage P22 operon containing the excisionase and integrase genes; (ii) a fusion between the bacteriophage promoter P_L and a promoterless reporter gene; (iii) constitutive expression of any transcriptional activators that are needed for expression of the promoter of interest; and (iv) a cI repressor gene that is flanked by P22 att sites, which are the DNA sequences at which the enzymes excisionase and integrase function. In the absence of expression from the promoter of interest, the cI repressor tightly represses the expression of the P_L-reporter gene fusion. After induction of the promoter of interest, the enzymes excisionase and integrase are produced; they recognize the att sites and catalyze a recombination event that excises the cI repressor gene. Loss of the cI repressor protein results in irreversible expression of the P_L-reporter gene fusion, which is relatively highly expressed in the organisms tested. This entire genetic system was placed on a stable, multicopy, broad-host-range vector (26), which allows it to be readily introduced into other bacteria.

We demonstrated the effectiveness of this genetic system by tailoring it to indicate the availability of arabinose to the biocontrol agent E. cloacae JL1157 in culture and in the barley rhizosphere (Fig. 2 to 4). Arabinose is one of many different carbohydrates available to bacteria in the spermosphere and/or rhizosphere, and it is not in great abundance relative to other carbohydrates (6, 31). Specifically, we examined the expression of the E. coli L-arabinose-inducible araBAD promoter (P_BAD), which is one of the best-characterized promoter regulatory systems. The binding of arabinose to the AraC regulatory protein induces the expression of the ara loci, and the resulting production of arabinose transporters results in higher internal arabinose concentrations and hence further induction. This autocatalytic mechanism of induction results in all-or-none expression of the P_BAD promoter (35). That is, cells exposed to arabinose concentrations below a threshold exhibit no P_BAD activity, whereas those exposed to concentrations above a threshold exhibit P_BAD activity at a level that is not influenced by arabinose concentration. The threshold arabinose concentration may not be identical for all of the cells in a population, however. Thus, a population exposed to an intermediate arabinose concentration is likely to contain both P_BAD-expressing and P_BAD-nonexpressing cells. Previous studies demonstrated that the proportion of a population that exhibits P_BAD activity is influenced by arabinose concentration within the range of 1.33 to 133 µM (14, 19, 35). Similarly, we demonstrated that the proportion of a population that exhibited P_BAD-mediated site-specific recombination was influenced by arabinose concentration. Specifically, only 52% of the cells had experienced recombination after exposure to 6.5 µM arabinose (Fig. 2), whereas almost 100% of the population had undergone recombination after exposure to 65 µM arabinose. This suggests that in metabolically active cells, the dominant factor affecting the removal of the cI cassette is the kinetics of P_BAD-xis,int expression rather than the kinetics of Xis and Int protein production or Xis- and Int-mediated recombination.

The absence of an increase in the percentage of fluorescent biosensor cells after their introduction into soil could indicate that arabinose was not readily available in the soil but more likely reflects the entry of the JL1157(pAraLHB) cells into a physiological state in which the cells did not exhibit a visible response to arabinose. The latter possibility is supported by the fact that soils are generally viewed as nutritionally limited, and stationary-phase cells, which are nutritionally limited, also exhibited a small visible response to arabinose. In fact, it is possible that in both the stationary-phase culture and the soil populations examined, the cells that exhibited fluorescence after exposure to arabinose did so only after the arabinose was utilized as a carbon source, thus changing their physiological status to one with greater energy reserves. This would explain why the stationary-phase cultures and soil populations exhibited a much longer period between arabinose exposure and the appearance of fluorescent cells than did exponential-phase cultures. Among the steps required for the cells to exhibit a visible response to arabinose, the arabinose transport step and the Xis- and Int-mediated recombination step are prime targets for inhibition in a nutrient-limited cell because of the high energy requirements of both permease synthesis and recombination (38). Since the in vivo half-life of excisionase (ca. 4 to 7 min) is considerably shorter than that for integrase (ca. 60 min) in wild-type E. coli cells, recombination will be particularly dependent on excisionase levels (22, 39). In the genetic system described here, the excisionase levels depend on the strength of the signal-responsive promoter that controls xis/int expression, as well as on physiological factors that influence target promoter expression or excisionase stability. Previous studies have demonstrated that the Lon and FtsH proteases target degradation of excisionase (22); thus, any factors that influence the activity of these proteases, such as starvation and heat shock, will also influence recombination.

The results of the soil study illustrate the importance of using this recombination-based biosensor system for the study of metabolically active cells. Although the poorer response in stationary-phase cells versus exponential-phase cells may be unique to this arabinose biosensor, it is likely that the effect of a cell's metabolic status on the ability to respond to a signal is an important factor to consider with any recombination-based reporter system. Alternatively, the poorer response in stationary-phase cells may be a consequence of the need for cell division to dilute the concentration of cI repressor sufficiently so that the P_L-gfp fusion would be derepressed. Previous studies that employed recombination-based genetic tools have not documented this phenomenon.

The proportion of JL1157(pAraLHB) cells that was fluorescent was significantly larger for populations recovered from the rhizosphere of barley seedlings than for cells in the original inoculum, providing strong evidence that arabinose was available to the rhizosphere microflora under these conditions (Fig. 3). Use of the JL1157(pAraLHB) biosensor in culture with seed exudates provided clear evidence that barley seed exudates contained arabinose and that the sugar was available mostly in regions near the seed-root junction (Fig. 4B and C). Although the arabinose on the roots may have originated in the seed and arrived at the roots during leakage of the seed exudate, the arabinose on the roots may also have originated in the roots. Arabinose is a component of various cell wall polymers of the grasses, including glucuronoarabinoxylans, arabinans, and arabinogalactans (6). In dividing and elongating cells, such as in young roots, glucuronoarabinoxylans are highly branched with arabinosyl units; these arabinosyl units are not cleaved, and thus arabinose is probably not released. In contrast, after elongation and differentiation are complete, as in older roots, the cell wall assembly process cleaves the arabinosyl units (6, 13), thus releasing arabinose. This release of arabinose has been observed in isolated coleoptile cell walls (6, 10, 15). Based on this information, we would predict that in a barley seedling that has both seminal roots and adventitious roots, the seminal roots, which are older, would more likely release arabinose than the adventitious roots. Unfortunately, the seminal roots are also closer in proximity to the seedling. Thus, the observation that arabinose-induced JL1157(pAraLHB) cells were frequently on the seminal roots and were rarely observed on the adventitious roots does not support, nor does it argue against, a root origin for the arabinose.

Although the arabinose-responsive biosensor system described here was subject to catabolite repression, it nonetheless responded to the presence of arabinose in both isolated seed exudates and the rhizosphere. Thus, while glucose and other catabolites that can repress P_BAD may have been present in the seed, and possibly root, exudates, they evidently were not present in sufficient concentrations to repress the arabinose-induced response. To eliminate this consideration in future studies, the P_BAD promoter could be modified to reduce its sensitivity to catabolite repression by removing the region on the promoter where the catabolite activator protein binds (17).

Although the recombination-based biosensor system does not quantify arabinose availability to the biosensor cells, it does indicate the presence or absence of available arabinose at any time during the life history of a cell. This is in contrast to biosensor systems that are designed to quantify availability by relying on transcriptional fusions between a signal-responsive promoter and a reporter gene (2, 16, 18, 20). For such quantitative biosensors, the strength of the reporter signal is a function of the concentration of the inducer; therefore, when the inducer concentration is low, the reporter signal may be weak. In the recombination-based biosensor system described here, the strength of the reporter signal (green fluorescence) was a function exclusively of the expression level of the P_L promoter and was independent of the concentration of inducer present. Thus, a major advantage of this biosensor system over quantitative promoter-reporter fusion-based biosensors is its ability to amplify the reporter signal that results from a low concentration of inducer. In some cases, this amplification function may allow for the detection of a low inducer concentration or a transiently available inducer that would otherwise be missed by a quantitative biosensor.

An important requirement for a recombination-based biosensor is the selection of a signal-responsive promoter that exhibits a low basal level of transcription. As in other recombinase-based genetic systems (4, 5, 21), the production of even low levels of recombinase can lead to a high background level. In this study, leaky transcription from P_BAD-xis,int led to premature excision of the cI cassette, which resulted in a high background level of green fluorescent cells. This high background could be reduced in several ways. First, we have observed that when the cI cassette is integrated into the chromosome instead of the plasmid, there is little to no background green fluorescent cells (data not shown). Second, the att sequence could be altered to reduce recognition by Xis and Int (4), thus decreasing the sensitivity of the recombinational switch to low concentrations of Xis and Int. However, this could possibly increase the minimal arabinose concentration that the biosensor detects. Third, the translational efficiency of the recombinase genes could be altered (21). Fourth, the high background can be minimized in the cells that are initially deployed into a habitat by using fluorescence-activated cell sorting. In recent studies, we have used this technique to separate the green fluorescent cells, which have resulted from premature excision of the cI cassette, from the desired nonfluorescent biosensor cells in an inoculum preparation and have obtained a background of <1.0% (data not shown).

The design and recombination elements of the site-specific recombination system described here could be applied to the development of a promoter trap to identify genes that are induced transiently, or only to low levels, in vivo (4, 25, 28, 36). A similar system could be used to express a target gene in an environment only under conditions in which that gene product is needed, thus minimizing the metabolic wastefulness that results from expression under nontarget conditions. Such target genes could include genes that kill the cell, thus mediating environmental containment, and genes that confer survival or enhanced competitiveness in a particular microhabitat.

 
This research was supported in part by the Iowa State University Agronomy Department, by an Iowa State University Biotechnology/Carver Trust Grant, by the Iowa Agriculture and Home Economics Experiment Station, and by the Hatch Act and the State of Iowa.

We thank Catherine Axtell for providing pVLacRed.

 
* Corresponding author. Mailing address: 2537 Agronomy Hall, Iowa State University, Ames, IA 50011-1010. Phone: (515) 294-0495. Fax: (515) 294-3163. E-mail: larryh{at}iastate.edu.

Journal paper no. J-19608 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa (project no. IOW04144).

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Brandl, M. T., B. Quinones, and S. E. Lindow. 2001. Heterogeneous transcription of an indoleacetic acid biosynthetic gene in Erwinia herbicola on plant surfaces. Proc. Natl. Acad. Sci. USA 98:3454-3459.[Abstract/Free Full Text]
2
2. Bringhurst, R. M., Z. G. Cardon, and D. J. Gage. 2001. Galactosides in the rhizosphere: utilization by Sinorhizobium meliloti and development of a biosensor. Proc. Natl. Acad. Sci. USA 98:4540-4545.[Abstract/Free Full Text]
3
3. Broughton, W. J., and M. J. Dilworth. 1971. Control of leghaemoglobin synthesis in snake beans (Vigna sinensis, sesquipedalis). Biochem. J. 125:1075-1080.[Medline]
4
4. Camilli, A., D. T. Beattie, and J. J. Mekalanos. 1994. Use of genetic recombination as a reporter of gene expression. Proc. Natl. Acad. Sci. USA 91:2634-2638.[Abstract/Free Full Text]
5
5. Camilli, A., and J. J. Mekalanos. 1995. Use of recombinase gene fusions to identify Vibrio cholerae genes induced during infection. Mol. Microbiol. 18:671-683.[CrossRef][Medline]
6
6. Carpita, N. C. 1996. Structure and biogenesis of the cell walls of grasses. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47:445-476.[CrossRef]
7
7. Chalfie, M., Y. Tu, G. Euskirchen, W. W. Ward, and D. C. Prasher. 1994. Green fluorescent protein as a marker for gene expression. Science 263:802-805.[Abstract/Free Full Text]
8
8. Costa, J. M., and J. E. Loper. 1994. Characterization of siderophore production by the biological control agent Enterobacter cloacae. Mol. Plant-Microbe Interact. 7:440-448.
9
9. Curl, E. A., and B. Truelove. 1986. The rhizosphere. Springer-Verlag, New York, N.Y.
10
10. Darvill, A. G., C. J. Smith, and M. A. Hall. 1978. Cell wall structure and elongation growth in Zea mays (maize) coleoptile tissue. New Phytol. 80:503-516.[CrossRef]
11
11. Figurski, D., and D. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76:1648-1652.[Abstract/Free Full Text]
12
12. Frischauf, A. M., H. Lehrach, A. Poustka, and N. Murray. 1983. Lambda replacement vectors carrying polylinker sequences. J. Mol. Biol. 170:827-842.[Medline]
13
13. Gibeaut, D. M., and N. C. Carpita. 1991. Tracing cell wall biogenesis in intact cells and plants: selective turnover and alteration of soluble and cell wall polysaccharides in grasses. Plant Physiol. 97:551-561.[Abstract/Free Full Text]
14
14. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose P_BAD promoter. J. Bacteriol. 177:4121-4130.[Abstract/Free Full Text]
15
15. Heyn, A. N. J. 1986. A gas chromatographic analysis of the release of arabinose from coleoptile cell walls under the influence of auxin and dextranase preparations. Plant Sci. 45:77-82.
16
16. Jaeger, C. H., III, S. E. Lindow, W. Miller, E. Clark, and M. K. Firestone. 1999. Mapping of sugar and amino acid availability in soil around roots with bacterial sensors of sucrose and tryptophan. Appl. Environ. Microbiol. 65:2685-2690.[Abstract/Free Full Text]
17
17. Johnson, C. M., and R. F. Schleif. 1995. In vivo induction kinetics of the arabinose promoters in Escherichia coli. J. Bacteriol. 177:3438-3442.[Abstract/Free Full Text]
18
18. Joyner, D. C., and S. E. Lindow. 2000. Heterogeneity of iron bioavailability on plants assessed with a whole-cell GFP-based bacterial biosensor. Microbiology 146:2435-2445.[Abstract/Free Full Text]
19
19. Khlebnikov, A., O. Risa, T. Skaug, T. A. Carrier, and J. D. Keasling. 2000. Regulatable arabinose-inducible gene expression system with consistent control in all cells of a culture. J. Bacteriol. 182:7029-7034.[Abstract/Free Full Text]
20
20. King, J. M. H., P. M. DiGrazia, B. Applegate, R. Burlage, J. Sanseverino, P. Dunbar, F. Larimer, and G. S. Sayler. 1990. Rapid, sensitive bioluminescence reporter technology for naphthalene exposure and biodegradation. Science 249:778-781.[Abstract/Free Full Text]
21
21. Lee, S. H., S. M. Butler, and A. Camilli. 2001. Selection for in vivo regulators of bacterial virulence. Proc. Natl. Acad. Sci. USA 98:6889-6894.[Abstract/Free Full Text]
22
22. Leffers, G. G., Jr., and S. Gottesman. 1998. Lambda Xis degradation in vivo by Lon and FtsH. J. Bacteriol. 180:1573-1577.[Abstract/Free Full Text]
23
23. Leveau, J. H., and S. E. Lindow. 2001. Quantitative monitoring of bacterial sugar consumption in the phyllosphere. Proc. Natl. Acad. Sci. USA 98:3446-3453.[Abstract/Free Full Text]
24
24. Loper, J. E., C. A. Ishimaru, S. R. Carnegie, and A. Vanavichit. 1993. Cloning and characterization of aerobactin biosynthesis genes of the biological control agent Enterobacter cloacae. Appl. Environ. Microbiol. 59:4189-4197.[Abstract/Free Full Text]
25
25. Mahan, M. J., J. W. Tobias, J. M. Slauch, P. C. Hanna, R. J. Collier, and J. J. Mekalanos. 1995. Antibiotic-based selection for bacterial genes that are specifically induced during infection of a host. Proc. Natl. Acad. Sci. USA 92:669-673.[Abstract/Free Full Text]
26
26. Miller, W. G., J. H. Leveau, and S. E. Lindow. 2000. Improved gfp and inaZ broad-host-range promoter-probe vectors. Mol. Plant-Microbe Interact. 13:1243-1250.[Medline]
27
27. Newman, J. R., and C. Fuqua. 1999. Broad-host-range expression vectors that carry the L-arabinose-inducible Escherichia coli araBAD promoter and the araC regulator. Gene 227:197-203.[CrossRef][Medline]
28
28. Oke, V., and S. R. Long. 1999. Bacterial genes induced within the nodule during the Rhizobium-legume symbiosis. Mol. Microbiol. 32:837-849.[CrossRef][Medline]
29
29. Platt, R. 2000. Ph.D. thesis. Iowa State University, Ames.
30
30. Platt, R., C. Drescher, S. K. Park, and G. J. Phillips. 2000. Genetic system for reversible integration of DNA constructs and lacZ gene fusions into the Escherichia coli chromosome. Plasmid 43:12-23.[CrossRef][Medline]
31
31. Roberts, D. P., P. D. Dery, I. Yucel, J. Buyer, M. A. Holtman, and D. Y. Kobayashi. 1999. Role of pfkA and general carbohydrate catabolism in seed colonization by Enterobacter cloacae. Appl. Environ. Microbiol. 65:2513-2519.[Abstract/Free Full Text]
32
32. Roberts, D. P., A. M. Marty, P. D. Dery, I. Yucel, and J. S. Hartung. 1996. Amino acids as reduced carbon sources for Enterobacter cloacae during colonization of the spermospheres of crop plant. Soil Biol. Biochem. 28:1015-1020.[CrossRef]
33
33. Roberts, D. P., C. J. Sheets, and J. S. Hartling. 1992. Evidence for proliferation of Enterobacter cloacae on carbohydrates in cucumber and pea spermosphere. Can. J. Microbiol. 38:1128-1134.
34
34. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed., vol. 1. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
35
35. Siegele, D. A., and J. C. Hu. 1997. Gene expression from plasmids containing the araBAD promoter at subsaturating inducer concentrations represents mixed populations. Proc. Natl. Acad. Sci. USA 94:8168-8172.[Abstract/Free Full Text]
36
36. Slauch, J. M., M. J. Mahan, and J. J. Mekalanos. 1994. In vivo expression technology for selection of bacterial genes specifically induced in host tissues. Methods Enzymol. 235:481-492.[Medline]
37
37. Smibert, R. M., and N. R. Krieg. 1994. Phenotypic characterization, p. 611-654. In P. Gerhardt, R. G. E. Murray, W. A. Wood, and N. R. Krieg (ed.), Methods for general and molecular bacteriology. American Society for Microbiology, Washington, D.C.
38
38. Weisberg, R. A., and J. A. Gallant. 1967. Dual function of the lambda prophage repressor. J. Mol. Biol. 25:537-544.[CrossRef][Medline]
39
39. Weisberg, R. A., and M. E. Gottesman. 1971. The stability of Int and Xis functions, p. 489-500. In A. D. Hershey (ed.), The bacteriophage lambda. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

---

Applied and Environmental Microbiology, July 2002, p. 3588-3596, Vol. 68, No. 7
0099-2240/02/$04.00+0     DOI: 10.1128/AEM.68.7.3588-3596.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Zhang, X.-X., George, A., Bailey, M. J., Rainey, P. B. (2006). The histidine utilization (hut) genes of Pseudomonas fluorescens SBW25 are active on plant surfaces, but are not required for competitive colonization of sugar beet seedlings. Microbiology 152: 1867-1875 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Casavant, N. C. Articles by Halverson, L. J. Search for Related Content PubMed PubMed Citation Articles by Casavant, N. C. Articles by Halverson, L. J. Agricola Articles by Casavant, N. C. Articles by Halverson, L. J.