Pseudomonas putida ATCC 12633 was mutagenized by the suicide mini-Tn5::lacZ1 transcriptional fusion vector. About sixty colonies defective in starvation promoter-driven genes were isolated from over 3,000 kanamycin resistant transconjugants. Several of these mutants, which showed strong -galactosidase activities, were selected for further analysis. These mutants expressed one to four times more -galactosidase activities upon starvation, while showing constant low level of activity during growth phase. Carbon starvation caused more rapid loss of viability in the mutants: 50% loss of viability in the mutants and the wild type occurred in about one and four weeks, respectively. The activities of -galactosidase in these mutants was constant during the starvation period.

Tn5 flanking DNA from the mutants MK107 and MK114 was cloned into the vector pMMB67EH. The wild type P. putida MK1 containing recombinant plasmid pMK101 or pMK201 derived from MK107 and MK114, respectively, expressed high -galactosidase activity upon starvation. In subsequent studies, P. putida carrying the starvation promoter sequences will be evaluated for their efficacy in in situ bioremediation and for ecological studies.

Environmental contamination by industrial, agricultural, military and municipal pollutants is a serious problem worldwide. In the U.S. alone, more than 32,000 hazardous waste sites are currently considered Superfund candidates (Lindow et al., 1989), and the actual number could be as high as 75,000 (Abelson, 1992).

Indigenous bacteria found in contaminated environments frequently possess the biochemical potential to degrade many of these toxic compounds. However, their biological activities are often too slow, due mainly to low amounts of nutrients in natural environments. The feasibility of overcoming the low nutrient problem by feeding nutrients into the contaminated sites to stimulate bacterial growth, i.e., biostimulation of indigenous strains, has been demonstrated by McCarty and co-workers (Semprini et al., 1990). Although it may be effective in some situations, the biostimulation approach faces technical problems with regard to (a) introducing large quantities of nutrients, and (b) inhibition of nutrient penetration (and degradation of the xenobiotics) due to the plugging resulting from large biomass increase. Additionallly, the cost of the nutrients may be prohibitive. Application of the starvation promoters technology (Matin, 1994) might address this problem by reducing both the amounts of nutrients and biomass to manageable levels for in situ bioremediation. These promoters permit the decoupling of bacterial growth from expression of a desired biochemical activity. Pioneer work has been done for the degradation of trichloroethylene (TCE) by using one of the E. coli starvation promoters (Little et al., 1991). Such recombinant E. coli strains degraded TCE repidly in nongrowth state; in contrast, strains in which gene expression was under a growth dependent promoter, no TCE degradation was observed during nongrowth. This paper describes the characterization of starvation promoters in Pseudomonas putida ATCC 12633 by using a transcriptional fusion vector system. Pseudomonas strains are indigenous to contaminated sites, while E. coli is not. This study also revealed that P. putida possesses greater survivability than E. coli under carbon starvation conditions, making this strain a suitable candidate for in situ bioremediation.

Bacterial strains and plasmids. Pseudomoans putida ATCC 12633 was used in this study. P putida MK1, used as the recipient strain for Tn5 mutagenesis, is a rifampicin-resistant derivative of P. putida ATCC12633. P putida MK104, MK107, MK114, and MK201 are Tn5-induced mutants of P putida MK1. E. coli S17-1 lpir (pUT mini-Tn5:: lacZ1) was used as the donor strain for Tn5 mutagenesis. This strain was obtained by triparental mating between E.coli CC118 lpir (pUT mini-Tn5 lacZ1) (Herrero et al., 1990) and E. coli S17-1 lpir (Lorenzo and Timmis, 1993) in the presence of the helper plasmid pRK600 (Lorenzo and Timmis, 1993) in this study. For cloning vectors, a plasmid pMMB67EH (Furste et al., 1986) and a cosmid pLAFR1 (Friedman et al., 1982) were used.

Transposon mutagenesis. The mating experiment between donor and recipient strains was performed according to Lorenzo et al. (1993).

Screening of the starvation response gene mutants. Each exconjugant resistant to kanamycin and rifampicin was transferred by a toothpick onto M9 agar plus 5'-bromo-4'-chloro-3'-indolyl--D-galactosidase (X-gal) containing either 0.025 or 0.3% glucose and grown at 30^oC. The mutants whose starvation response genes were impaired could be distinguished by their ability to form blue color on the low glucose plates, but not on the high plates during the incubation period (ca. 10 to 15 hrs) (Groat et al., 1986).

Liquid assays for b-galactosidase. Cells were grown up to the late exponential phase in M9 medium containing 0.1% glucose, then subcultured into the same medium. An exponential-phase culture of this medium was again subcultured 1:10 into prewarmed experimental M9 medium containing glucose (0.1%, 0.05%, or 0.025%) as sole carbon source and incubated at 30^oC with shaking at 150 rpm. Each sample removed at different time intervals was analyzed for its optical density at 660 nm and for -galactosidase activity. The assay for -galactosidase activity was performed as described by Clark and Switzer (1977).

Viability experiments. Cells were grown in 50 ml of M9 medium plus 0.05% glucose in 125 ml Erlenmeyer flasks at 150 rpm in 30^oC water bath. Zero time samples were taken when the value of optical density reached the maximum level. The viability of starving cultures was determined by spreading serial dilution of cells on LB plates.

Cloning of mutant DNA. For cosmid cloning, total cellular DNA was partially digested with EcoRI and 20 to 30 kb DNA fragments were isolated from a 0.5% agarose gel. The DNA was purified by Gene Clean II (Bio 101, Inc.), then ligated into pLAFR1, which was EcoRI-digested and dephosphorylated, overnight at 12^oC. The ligated DNA was packaged into Lambda Packagene (Promega, CO) and transduced into E. coli LE392. Transductants were selected on LB medium containing kanamycin and ampicillin. For plasmid cloning, the genomic DNA was partially digested with PstI and 5 to 15 kb DNA fragments were isolated after gel electrophoresis in a 0.7% agarose. 0.5 µg of DNA was ligated with 0.5 µg of PstI-digested and dephosphorylated pMMB67EH overnight at 15^oC. The ligated mixture was electroporated into E. coli MC1061 competent cells.

Genetic techniques. Transformation protocol and preparation of competent cells, DNA extraction, restriction enzyme analysis, ligation of DNA with T4 DNA ligase, calf intestinal alkaline phosphatase treatment, purification of DNA fragments from agarose gel, and other genetic techniques were performed by standard procedures (Ausubel et al., 1989; Maniatis et al., 1982) or as recommended by the suppliers. Transfer of plasmids from E.coli to P. putida was performed by triparental mating method (Andersen and Douglas, 1984) in the presence of helper plasmid pRK600.

Isolation and characterization of Tn5-induced mutants. The mini Tn5::lacZ1 transcriptional fusion plasmid was used to generate P. putida MK1 mutants defective in starvation-induced genes. This plasmid is a suicide plasmid which is able to replicate only in donor host strains that produce the R6K-specified protein (Herrero et al., 1990). The plasmid was first transferred from E.coli CC118 lpir into E. coli S17-1 lpir, because E.coli CC118 is resistant to rifampicin, but not E. coli S17-1. The recipient P. putida MK1 was isolated after growth of the parental strain, P. putida ATCC 12633, on LB medium containing rifampicin (150 g/ml). Thus, LB plates containing kanamycin and rifampicin effectively selected P. putida::Tn5 exconjugants and did not permit growth of the donor cells. Over 3,000 Km^r Rif^r exconjugants were obtained by conjugation. Preliminary screening of mutants performed on M9 agar plate containing either 0.3% or 0.025% glucose as sole carbon source enabled us to isolate sixty starvation-related mutants. This screening strategy is based on the fact that lower amount of carbon source permits bacteria to reach the stationary phase earlier while growth is still progressing on the the higher amount plates, differentiating starvation-dependent gene activity from growth-related one. Among these sixty mutants, some, which showed strong differences in blue color formation on low and high glucose plates, were further analyzed in liquid medium for -galactosidase activity.

All mutants expressed one to four times more -galactosidase activity upon starvation, while showing constant low level of activity during growth phase (Fig. 1). The mutant MK201 was used as a control because it expressed deep blue color on high (0.3%) glucose plate, while showing very faint color on low (0.025%) even after glucose was consumed. As expected, this mutant did not show any starvation-induced promoter activity. Two mutants, MK107 and MK114, expressed strong b-galactosidase activity upon starvation. These mutants were further selected for cloning of their promoter regions and analysis of the structural genes.

Survivability of the wild type and mutants on glucose starvation. The viability of the wild type MK1 decreased to about 34% of the initial within one month of glucose starvation. In contrast, the mutants MK107 and MK114 lost up to 98% viability within the same period (Fig. 2). The mutation did not affect the doubling time which is approximately 100 min for both the wild type and the mutants in M9 medium. Thus, these genes appear to be concerned only with starvation survival.

Cloning of Km^r mutant chromosomal DNA. Chromosomal DNA isolated from mutants MK107 and MK114 was ligated both into the cosmid vector pLAFR1 and into pMMB67EH plasmid vector. These vectors were chosen because they are broad-host-range which are able to replicate in P. putida as well as in E. coli. Furthermore, they can easily be transferred from E. coli into P. putida by conjugation in the presence of a helper plasmid. Plasmid cloning was found to be as efficient as cosmid cloning when the vector was introduced into E. coli MC1061 by electroporation. About fifteen recombinant DNA cloned into pMMB67EH from each mutant were subjected to restriction analysis. The size of insert DNA ranged from 6 kb to more than 15 kb. Further clonal analysis showed that all clones from a given mutant shared a common fragment, carrying the lacZ1 and kanamycin resistance genes.

Assessment of the clones in P. putida MK1. All recombinant clones were first transferred from E. coli MC1061 into the wild type P. putida MK1 by triparental mating method. P. putida MK1 containing each of these clones was analyzed for starvation-induced promoter activity. Two carrying the largest and the smallest insert from each mutant were tested for liquid culture assays of -galactosidase. P. putida MK1 containing either pMK101 or pMK103, which are derived from MK107, expressed very similar level of -galactosidase activity (Fig. 3). In the case of other clones, pMK201 and pMK203, derived from MK114, only pMK201 conferred starvation promoter activity, indicating that the region responsible for this activity is deleted in pMK203 (Fig. 3). The pMK1011 was made by removing 5' flanking DNA of transposon from pMK101 as a negative control. As expected, starvation-driven promoter activity was not detected in pMK1011, but, higher basal level of activity was observed (Fig. 3). Subcloning and primer extension experiments are in progress to precisely localize the starvation promoters.

Many naturally occurring bacteria degrade environmental pollutants, but desired biological activity in situ is limited due to the low amount of nutrients and growth rates. Supplementation with necessary nutrients to amplify bacterial growth and subsequent biochemical activity have been shown to be effective in in situ bioremediation (Semprini et al., 1990). However, nutrient supplementation can cause potential technical problems, for example, the clogging of the subsurface pores by the increase in aquifer biomass. The use of starvation promoters to drive the expression of bioremediation genes is an attractive alternative. This approach is to utilize natural stresses, such as starvation itself, as the trigger for the expression of biodegradative activity by splicing the genes of interest behind starvation promoters. In this study, we mutagenized P. putida MK1 by mini-Tn5::lacZ1 transcriptional fusion vector to clone starvation promoters. Among sixty mutants which were impaired in starvation-related genes, two mutants, MK107 and MK114, were selected for further analysis because they expressed strong -galactosidase activity upon starvation. Viability of MK107 and MK114 during carbon starvation was markedly lower than that of the wild type, but the mutation did not affect growth itself. Further studies will be performed to elucidate the function and regulatory circuits of these genes in P. putida. These studies along with the use of starvation promoters are expected to facilitate the construction of more effective strains for in situ bioremediation. The engineered strains will also be utilized for environmental effects and risk assessment studies.

Abelson PH (1992) Remediation of hazardous waste sites, Science. 255:901

Andersen K, Douglas M (1984) Construction and use of a gene bank of Alcaligenes eutrophus in the analysis of ribulose bisphosphate carboxylase genes. J. Bacteriol. 159:973-978.

Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (1989) Current protoclos in molecular biology. New York / John Wiley & Sons.

Clark JM, Switzer RL (1977). Experimental biochemistry, 2nd ed., p. 97-103. W. H. Freeman and Co., San Francisco.

Friedman A, Long S, Brown S, Buikema W, Ausubel F (1982) Gene 18:289-296

Fürste JP, Pansegrau W, Frank R, Blöocker H, Scholz P, Bagdasarian M, Lanka E (1986) Molecular cloning of the RP4 DNA primase region in a multirange tacP expression vector. Gene. 48:119-131.

Groat RG, Schultz JE, Zychlinsky E, Bockman A, Matin A (1986) Starvation proteins in Escherichia coli: Kinetics of synthesis and role in starvation survival. J. Bacteriol. 168:486-493.

Herrero M, Lorenzo V, Timmis K (1990) Transposon vectors containing non-antibiotic resistance selection markers for cloning and stable chromosomal insertion of foreign genes in Gram-negative bacteria. J. Bacteriol. 172:6557-6567.

Lindow SE, Panopoulos NJ, McFarland BL (1989) Genetic engineering of bacteria from managed and natural habitats. Science. 244:1300-1305.

Little C, Fraley C, McCann M, Matin A (1991) Use of bacterial stress promoters to induce biodegradation under conditions of environmental stress. Proceedings of In-situ and On-site Bioreclamation. Battelle Symposium, San Diego, Califormia 19-21 March. ed. Hichee, R. E. pp. 493-498. Stoneham, MA: Butterworths.

Lorenz V, Timmis K (1993) Analysis and construction of stable phenotypes in Gram-negative bacteria with Tn5 and Tn10-derived mini-transposons (personal communication)

Maniatis T, Fritsch EF, Sambrook J (1982) Molecular Cloning; a laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

Matin A (1994) Starvation promoters of Escherichia coli: Their function, regulation, and use in bioprocessing and bioremediation. Annals of N.Y. Acad. of Sciences. In press.

Semprini L, Roberts PV, Hopkins GD, McCarty PL (1990) A field evaluation of in situ biodegradation of chlorinated ethenes: Part 2, Results of biostimulation and biotransformation experiments. Ground Water, Vol. 28, No. 5:715-727.

Simon R, Priefer U, Puhler A (1983) A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in gram negative bacteris. Biotechnology 1:748-791.

Matin fig 1

Matin fig 2

Matin fig 3

1. Corresponding author