Sensors of iron availability and metabolic activity were constructed. A sensor responsive to iron was constructed by fusing a promoterless ice nucleation activity gene (inaZ) to an iron-regulated promoter of a genomic region involved in pyoverdine (fluorescent siderophore) (pvd) production in Pseudomonas syringae. Cells of Pseudomonas fluorescens and P. syringae that contained the pvd-inaZ fusion expressed iron-responsive ice nucleation activity in culture. P. fluorescens containing the pvd-inaZ fusion was responsive to iron in the bean rhizosphere, as shown by reduced ice nucleation activity upon addition of Fe(III) to soil. A metabolic activity sensor was generated by fusion of a ribosomal RNA (rRNA) promoter from P. syringae to a chloramphenicol acetyltransferase (CAT) reporter gene. A rRNA promoter region was identified in P. syringae by hybridization to a probe generated from conserved regions of other bacterial rRNA genes. Nucleotide sequence analysis revealed a portion of the P. syringae 16S gene to be similar to Escherichia coli; however, the promoter regions differed between species. CAT activity measurements of fusions with both P. syringae and E. coli rRNA promoters indicate that both have similar transcriptional activities in rich media conditions.

Concentrations of environmental pollutants and other chemical compounds that influence biological systems are primarily assessed by chemical procedures. Although such methods are sensitive indicators of chemical concentrations in bulk environmental samples, they do not assess whether compounds are present in forms that are biologically available to organisms occupying the natural habitat. Numerous physicochemical factors influence the form of chemical compounds or elements in solution or soil (Hickey and Kittrick, 1984; Blumhorst et al., 1990); because the chemical form has a profound effect on the toxicity or biological activity of a chemical, methods that evaluate only the concentration of the chemical do not provide an accurate assessment of the biological relevance of the chemical to the ecosystem. Thus, it is commonly observed that assessed chemical concentrations do not have the expected effects on biological communities or indicator organisms (Walton et al., 1989). There is need for new methods that can assess biologically meaningful levels of chemical compounds that are present in disturbed or natural ecosystems. Biological sensors for iron and copper, which are based on promoters of genes that are specifically regulated by these elements, have been developed (Loper and Lindow, 1994; Lindow and Rogers, 1991). The transcriptional activities of these promoters are monitored in situ by the ice nucleation reporter system (Lindgren et al., 1989). We wish to evaluate these chemically responsive promoters and develop a sensor of in situ metabolic activity of bacteria inhabiting different microhabitats. A metabolic activity sensor is being developed by the fusion of a ribosomal RNA gene regulatory region, which is transcribed at a rate proportional to metabolic activity (Jinks-Robertson, 1987), to a reporter gene.

Strains, plasmids, and growth conditions. P. syringae 31 is ice nucleation-active (Ina⁺) and was originally isolated from a corn leaf surface (Arny, 1976). Strain 31R1 is a spontaneous rifampicin resistant mutant. 31R1-P6, an Ina^- strain, was derived by deletion of a region of the inaZ gene and was obtained from J. Lindemann (DNA Plant Technologies, Oakland, CA). Pseudomonas fluorescens Pf-5 (Howell and Stipanovic, 1979), which was isolated from cotton rhizosphere, was obtained from C. R. Howell (USDA, ARS, College Station, TX). pVSP61 was obtained from W. Tucker (DNA Plant Technologies). Pseudomonas spp. were grown on nutrient agar (Difco Laboratories, Detroit, MI) supplemented with 1% glycerol or a minimal salts medium (SM) (Loper and Lindow, 1987). E. coli was grown on Luria-Bertani medium (Sambrook et al., 1989). pKK520-3 was obtained from J. Brosius (Brosius, 1987).

Construction of iron sensor. A pvd-inaZ fusion was constructed by fusing the 8 kb EcoRI fragment of pSFL12, containing the iron responsive promoter (Loper et al., 1984), to a promoterless inaZ gene in pVSP61 (Fig. 1). Constructs were introduced by conjugation into Pseudomonas fluorescens Pf-5 and P. syringae 31R1-P6, neither of which produced detectable ice nuclei in the absence of an introduced ice nucleation gene.

Enzyme assays. Ice nucleation was measured by the droplet-freezing assay (Lindow, 1990). Chloramphenicol acetyl transferase activity was measured by the spectrophotometric method of Shaw (1975).

Assessment of transcriptional activity of the iron sensor in the rhizosphere. Bacterial inoculum was grown in SM supplemented with varying concentration of FeCl₃. Cells were harvested by centrifugation and resuspended in sterile deionized water, and diluted to a final concentration of 10⁶ CFU/ml for root inoculation. Bean seeds were surface sterilized in 1% hypochlorite for 10 min, rinsed in deionized water, and placed on moist paper towels for 3 days. The resulting roots were dipped in suspensions of P. fluorescens Pf-5 and planted in pots with soil (-0.3 bar, pH 7.2, 14 mg of Fe per kg). In some cases soil was amended with FeEDTA, which is thought to increase the biological availability of iron in the rhizosphere (Kloepper et al., 1980; Scher and Baker, 1982).

Construction of metabolic activity sensor. Oligonucleotides, corresponding to conserved regions of the 16S ribosomal RNA gene (5'AGAGTTTGATCCTGGCTC

AG3' and 5'GTATTACCGCGGCTGCTG3'), were primers in PCR reactions using P. fluorescens genomic DNA as a template to generate a probe for the rRNA operon. A cosmid library of P. syringae in pLAFR3 was probed with the PCR product. DNA sequence determination was performed using Sequenase (US Biochemical, Clevland, OH).

Iron sensor. The ice nucleating activities of P. syringae and P. fluorescens strains conferred by pvd-inaZ were dramatically regulated by iron (Fig. 2 and 3). Transcription of the iron regulated pvd gene(s) was detected as high levels of ice nucleating activity in cells grown in media containing low levels of iron (10^-6 M added FeCl₃) (Fig. 2 and 3). At such low concentrations, the maximum level of ice nucleating activity that can be measured by the droplet freezing assay was observed. Cells containing pvd-inaZ expressed 10⁴ to 10⁵ greater ice nucleating activity at low concentrations of FeCl₃ (10^-6 M) than at high concentrations (10^-4 M). In contrast, the ice nucleating activity of cells containing iceC, an ice nucleation gene transcribed from its native promoter, was not affected substantially by FeCl₃, indicating that inaZ protein production and function were not greatly affected by iron availability. Thus, a comparison of the ice nucleation activities of cells containing pvd-inaZ and cells containing iceC provided a way to assess iron availability to Pseudomonas spp.

Transcriptional activity in rhizosphere. The ice nucleation activity of P. fluorescens was not influenced by the rhizosphere environment or by the methods used to obtain bacterial samples from the rhizosphere soil; the cultured cells used as inoculum expressed levels of ice nucleation activity equivalent to the levels expressed by similar cells retrieved from the rhizosphere shortly after inoculation. P. fluorescens Pf-5 cells that were grown in media amended with various concentrations of FeCl₃ and then applied to roots initially expressed levels of ice nucleating activity equivalent to the levels expressed in culture (Fig. 4). By 12 hrs. after inoculation of plant roots, however, cells retrieved from the rhizosphere expressed equivalent levels of ice nucleation activity regardless of the FeCl₃ concentration in the initial inoculum (Fig. 4).

In the rhizosphere of bean, cells that contained pvd-inaZ produced ice nuclei, indicating that the concentration of available iron was sufficiently low to allow transcription (Table 1). Nevertheless, cells that contained pvd-inaZ expressed lower levels of ice nucleation activity than cells containing iceC (Table 1). Because cells containing pvd-inaZ expressed higher levels of ice nucleation activity than cells containing iceC expressed in low iron culture medium (SM containing <10^-5 M added FeCl₃) (Fig. 2), we concluded that available Fe(III) was present in microsites colonized by P. fluorescens in the rhizosphere. Amendment of soil with FeEDTA had no effect on the ice nucleation activity of cells containing iceC, indicating that this treatment had no direct effect on the production or activity of the ice nucleation protein (Table 1). In contrast, cells containing pvd-inaZ expressed much less ice nucleation activity on roots of plants in soil amended with FeEDTA than in non-amended soil, suggesting that the levels of soluble iron in the soil influenced the transcriptional activity of the pvd promoter.

Metabolic activity sensor. Four cosmid clones were identified from a library of P. syringae B728a that was screened with a probe to the 16S gene. Direct nucleotide sequencing of the cosmids revealed a region with a high degree of similarity to the E.coli mature 16S rRNA gene (data not shown). The region of the cosmid which would correspond to the promoter was much more diverged from the E. coli sequence. A 3 kb HindIII fragment, containing a putative 16S promoter from P. syringae, was subcloned from one cosmid into the promoter probe vector pKK520-3 (Brosius and Lupski, 1987) creating a fusion to a promoterless chloramphenicol acetyltransferase (CAT) gene. A 2 kb deletion clone was created by digestion with SacII and religation, thereby removing the putative promoter region. A 1.6 kb BamHI-HindIII fragment from pKK3535 (Brosius et al., 1981), containing the E.coli rRNA promoter, was also inserted into pKK520-3.

CAT activity measurements were performed on all three fusion constructs in E. coli (Table 2). E. coli cells containing the different plasmid constructs were grown in Luria-Bertani broth with ampicillin (50 ug/ml) and chloramphenicol (20 ug/ml). Mid-log cultures were assayed according to Shaw (1975). The CAT activity of the P. syringae promoter fusion (P6) was similar to the activity of the E. coli promoter fusion (pE18) (Table 2). However, the construct with a 2 kb deletion, which removed the putative promoter region (pP6SacII), had activity similar to the promoterless vector control (pKK520-3).

Biological sensors of iron and metabolic activity are being developed. An iron-regulated promoter was transcribed by P. fluorescens in the bean rhizosphere, as detected by an ice nucleation activity reporter gene system. Amending soil with iron resulted in a reduction in the ice nucleation activity expressed by cells containing pvd-inaZ, indicating that Fe(III), rather than other constituents of the rhizosphere environment, was the primary factor affecting the transcriptional activity of the pvd promoter. On the basis of the relative ice nucleation activities expressed by cells containing the pvd-inaZ or iceC constructions, we estimated that the Pseudomonas spp. examined sensed a concentration of iron in the rhizosphere that was intermediate between the concentration in iron-deplete culture media and the concentration in low-iron culture media. Because only a fraction of the iron added in the form of FeCl₃ is actually present in a soluble form in a neutral aqueous solution (Raymond and Carrano, 1979), it is difficult to estimate the exact concentration of Fe(III) available in a culture or on plant surfaces. Nevertheless, bacteria inhabiting the rhizosphere may not experience the extreme iron deprivation that has been suggested by chemical models which estimate the availability of the ferric ion in soil (Lindsay and Schwab, 1982).

The rate of transcription of rRNA by bacteria is closely correlated with growth rate (Jinks-Roberstson, 1987). In response to demand for increased protein synthesis, bacteria increase the number of ribosomes per unit amount of cell mass. As such, we predict that the initiation of rRNA transcription will be an accurate measure of metabolic activity in bacterial cells. To measure the rate of rRNA transcription, and hence metabolic activity, we have fused a rRNA regulatory region to a reporter gene. Here we report a fusion with a P. syringae rRNA promoter and a CAT reporter gene which results in CAT activity in E. coli grown rich growth media that is similar to the activity of an E. coli promoter. Interestingly, the nucleotide sequence of the P. syringae rRNA promoter appears diverged from the E. coli sequence, but the activity of the two in E. coli appears similar.

Arny DC, Lindow SE, Upper CD (1976) Frost sensitivity of Zea mays increased by application of Pseudomonas syringae. Nature 262:282-284.

Blumhorst MR, Weber JB, Swain LR (1990) Efficacy of selected herbicides as influenced by soil properties. Weed Technology 4:279-283.

Brosuis J, Lupski J (1987) Plasmids for the selection and analysis of prokaryotic promoters. Methods in Enzymology 153:54-68.

Brosuis J, Ullrich A, Raker MA, Gray A, Dull TJ, Gutell RR, Noller HF (1981) Construction and fine mapping of recombinant plasmids containing the rrnB ribosomal RNA operon of E. coli. Plasmid 6:112-118.

Hickey MG, Kittrick MG (1984) Chemical partitioning of cadmium, copper, nickel and zinc in soils and sediments containing high levels of heavy metals. J. Environ. Qual. 13:372-376.

Howell CR, Stipanovic RD (1979) Control of Rhizoctonia solani on cotton seedlings with Pseudomonas fluorescens and with an antibiotic produced by the bacterium. Phytopathology 69:480-482.

Jinks-Robertson S, Masayasu N (1987) Ribosomes and tRNA. In: Neidhardt FC, Ingraham JL, Low KB, Magasanik B, Schaechter M, Umbarger HE (eds) Escherichia coli and Salmonella typhimurium cellular and molecular biology. American Society for Microbiology, Washington, D.C.

Kloepper JW, Leong J, Teintze M, Schroth MN (1980) Enhanced plant growth by siderophores produced by plant growth-promoting rhizobacteria. Nature 286:885-886.

Lindgren PB, Frederick R, Govindarajan AG, Panopoulos NJ, Staskawicz BJ, Lindow SE (1989) An ice nucleation reporter gene system: identification of inducible pathogenicity genes in Pseudomonas syringae pv. phaseolicola. EMBO J. 8:1291-1301.

Lindow SE, Rogers S (1991) Copper tolerance of Psuedomonas syringae strains isolated from fruit trees in California. In: Durbin R, Surico G, Mugnai L (eds) Pseudomonas syringae pathovars. Stamperia Granducale, Florence.

Lindow SE (1990) Bacterial ice nucleating activity. In: Klement S, Rudolf K, Sands DC (eds) Methods in phytobacteriology. Akademiai Kiado, Budapest.

Lindsay WL, Schwab AP (1982) The chemistry of iron in soils and its availability to plants. J. Plant Nutr. 5:821-840.

Loper JE, Lindow SE (1987) Lack of evidence for in situ fluorescent pigment production by Pseudomonas syringae pv. syringae on bean leaf surfaces. Phytopathology 77:1449-1454.

Loper JE, Orser CS, Panopoulas NJ, Schroth MN (1984) Genetic analysis of fluorescent pigment production in Pseudomonas syringae pv. syringae. J. Gen. Microbiol. 130:1507-1515.

Loper JE, Lindow SE (1994) A biological sensor for iron available to bacteria in their habitats on plant surfaces. Appl. Environ. Microbiol.60:1934-1941.

Orser C, Staskawicz BJ, Panopoulos NJ, Dahlbeck D, Lindow SE (1985) Cloning and expression of bacterial ice nucleation genes in Escherichia coli. J. Bacteriol. 164:359-366.

Raymond KN, Carrano CJ (1979) Coordination chemistry and microbial iron transport. Acc. Chem. Res. 12:183-190.

Sambrook JE, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

Scher FM, Baker R (1982) Effects of Pseudomonas putida and a synthetic iron chelator on induction of soil suppressiveness to Fusarium wilt pathogens. Phytopathology 72:1567-1573.

Shaw WV (1975) Chloramphenicol acetyltransferase from chloramphenicol-resistant bacteria. In: Colowick SP, Kaplan NO (eds) Methods in Enzymology. Academic Press, New York.

Walton BT, Anderson TA, Hendricks MS (1989) Physicochemical properties as predictors of organic chemical effects on soil microbial respiration. Environ. Tox. Chem. 8:53-63.

Hopkins Table 1

Hopkins Table 2

Hopkins fig 1

Hopkins fig 2

Hopkins fig 3

Hopkins fig 4

1. Offprint requests: S.E. Lindow, Department of Environmental Science, Policy, and Management, 108 Higard Hall, University of California, Berkeley, CA 94720-3110