Vol. 174, No. 9

JOURNAL OF BACTERIOLOGY, May 1992, P. 2978-2985

0021-9193/92/092978-08$02.00/0 Copyright X 1992, American Society for Microbiology

Tracking Genetically Engineered Bacteria: Monoclonal Antibodies against Surface Determinants of the Soil Bacterium Pseudomonas putida 2440 MARIA-ISABEL RAMOS-GONZALEZ,' FRANCISCO RUIZ-CABELLO,2 INGRID BRETTrAR,3 FEDERICO GARRIDO,2 AND JUAN L. RAMOS1' Consejo Superior de Investigaciones Cientificas, Estacion Experimental del Zaidin, Departamento de Bioquimica Vegetal, Apto 419, 18080 Granada, 1 and Laboratorio de Anadlisis Clinicos e Inmunologia, Hospital Virgen de las Nieves y Universidad de Granada, Granada,2 Spain, and Max-Planck-Institut fur Limnologie, Plon, Federal Republic of Germany3 Received 22 October 1991/Accepted 24 February 1992 Assessment of potential risks involved in the release of genetically engineered microorganisms is facilitated by the availability of monoclonal antibodies (MAbs), a tool potentially able to monitor specific organisms. We raised a bank of MAbs against the soil bacterium Pseudomonas putida 2440, which is a host for modified TOL plasmids and other recombinant plasmids. Three MAbs, 7.3B, 7.4D, and 7.5D, were highly specific and recognized only P. putida bacteria. Furthermore, we developed a semiquantitative dot blot assay that allowed us to detect as few as 100 cells per spot. A 40-kDa cell surface protein was the target for MAbs 7.4D and 7.5D. Detection of the cell antigen depended on the bacterial growth phase and culture medium. The 0 antigen of lipopolysaccharide seems to be the target for MAb 7.3B, and its in vivo detection was independent of the bacterial growth phase and culture medium. MAb 7.3B was used successfully to track P. putida(pWWO) released in unsterile lake mesocosms. Recent achievements in molecular biology have provided efficient ways to manipulate prokaryotic and eukaryotic organisms, and a number of possible applications for these "new" organisms are envisaged in the food, health, and agricultural industries. Bioremediation is an area of increasing interest for decontamination of polluted areas; such treatments involve, among other strategies, release of large numbers of wild-type or modified microorganisms to decontaminate polluted soils, water, and sediments. Among soil bacteria, the pseudomonad group exhibits an enormous range of metabolic activities against biogenic and xenobiotic chemicals (31). Furthermore, the catabolic pathways of pseudomonads have been used as model systems to expand the range of recalcitrant chemicals that microorganisms can mineralize (34). We have centered our attention on Pseudomonas putida 2440 (12), a restriction-negative strain which was derived from P. putida mt-2, the natural host for archetypal TOL plasmid pWWO (49). P. putida 2440 has been used as the host for recombinant TOL plasmids able to eliminate aromatic hydrocarbons that cannot be degraded by non-pWWO-modified pathways (1, 35). Furthermore, P. putida 2440 has been shown to be an excellent recipient for recombinant plasmids through conjugation and transformation (4), and it has been used in several biotransformation processes, e.g., production of benzyl alcohols (9). Concerns about the potential risks associated with deliberate or accidental release of large numbers of genetically engineered microorganisms have been raised because of the unpredictability of their immediate and long-term behavior in the environment. The assessment of potential risks will be facilitated by the availability of monoclonal antibodies

*

Corresponding author.

(MAbs), which make it possible for an organism to be specifically monitored and efficiently quantitated. We raised highly specific MAbs against surface determinants of P. putida 2440 (lipopolysaccharide [LPS] and proteins) and developed a solid semiquantitative assay that allowed us to detect, within hours, as few as 100 cells. MATERIALS AND METHODS Bacterial strains and culture conditions. The bacterial strains used in this work were P. putida 2440 (12), 2440(pWWO) (12), 2440(pWWO-EB62) (1), PaW340 and PaW94 (S. Harayama), and 29 and 43 (D. Springael); Pseudomonas sp. strain B13 (10); P. aeruginosa (this laboratory); P. fluorescens (17); P. syringae (17); P. oleovorans (M. Wubbolts); P. stutzeri (45); Rhizobium meliloti GRB13 (8); R leguminosarum 3855 (19); Erwinia chrysanthemi (42); Acinetobacter calcoaceticus (29); Azotobacter vinelandii (20); Agrobacterium tumefaciens 281 (14); Alcaligenes eutrophus (32); Escherichia coli JM101 (40); Klebsiella pneumoniae (J. Casadesu's); K aerogenes (30); Salmonella typhimurium (J. Casadesu's); a Corynebacterium sp. (this laboratory); Bacillus subtilis (22); Flavobacterium marinotypicum (Spanish Culture Collection 578) (44); and Synechocystis sp. strain PCC 6803 (A. Muro). Bacterial strains were grown at 30°C in LB medium (40), except for R. meliloti GRB13, R. leguminosarum 3855, and A. tumefaciens 281, which were grown at 28°C on Ty medium (5). Cyanobacteria were grown on BG11 medium (37), and F. marinotypicum was grown on nutritive medium II (40). P. putida strains were also grown on M9 minimal medium (1) with glucose (0.5% [wt/vol]), 10 mM succinate, and 5 mM m-methyl- and p-ethylbenzoate as the sole C source. Reagents and solutions. Phosphate buffer solution (PBS) was 20 mM sodium phosphate and 15 mM sodium chloride, 2978

VOL. 174, 1992

pH 7.4. This solution was supplemented with 0.05% (vol/vol) Tween 80 (PBST), albumin at the concentrations indicated (PBSA), or 0.05% (vol/vol) Tween 80 and albumin (PBSTA) at the concentrations indicated. PBS-azide was PBS supplemented with 0.1% (wt/vol) sodium azide. Poly-L-lysine solution was 1 mg of poly-L-lysine in 100 ml of PBS. Tris buffer solution (TB) was 50 mM Tris-HCl, pH 7.5. Peroxidase-conjugated and fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse immunoglobulins were from Dakopatts (Denmark). Nitrocellulose membranes were type HA (0.45-,m pore size; Millipore Corp.). MAb production. P. putida 2440 cells were grown overnight on LB medium. Cells were harvested by centrifugation and washed three times in PBS. About 107 cells in 0.25 ml of PBS were intraperitoneally injected into BALB/c mice. The injection was repeated three times at 1-week intervals. In the fourth injection, 108 cells were subepidermally injected in incomplete Freund adjuvant. Finally, 3 days prior to fusion, the mice received 108 cells in PBS intravenously. About 1.2 x 108 spleen lymphocytes from a previously immunized mouse were fused with 3 x 107 cells of mouse myeloma cell line NS-1 using polyethylene glycol as described by Kohler and Milstein (21). Hybridomas were selected in hypoxanthine-aminopterin-thymidine medium and then amplified in hypoxanthine-thymidine medium. Production of antibodies against P. putida 2440 was assayed by indirect immunofluorescence. Hybridomas producing highly specific antibodies against P. putida 2440 were cloned by limiting dilution to less than 1 cell per well. Selected clones were then used to produce ascites tumors in pristane-primed mice. Cells were removed from the ascites fluids by centrifugation at 2,800 x g for 10 min, and aliquots of cleared ascites fluids were stored at -70°C. Isotype analysis. Antibody class was determined by Ouchterlony (double-diffusion) analysis using isotype-specific antisera. ELISAs. Enzyme-linked immunosorbent assays (ELISAs) using whole cells were performed basically as described by Hermoso et al. (15). To improve the binding of whole cells to the polyvinyl chloride microtiter plates, 50 pl of poly-Llysine solution was added per well and removed after 30 min at room temperature. About 108 CFU in 50 ,u of fixation buffer (100 mM sodium bicarbonate, pH 9.4) were then added and incubated overnight at 4°C. Cells were washed three times (5 min each time) with 200 pl of PBST. Subsequently, a first blocking step was carried out with 100 ,u of PBSTA containing 0.1% albumin for 30 min. Fifty microliters of 103-fold-diluted ascites fluid in PBSTA containing 0.1% albumin was immediately added and incubated for 2 h. Plates were washed as described above with PBST, and finally 75 pl of a 103 dilution of peroxidase-conjugated rabbit anti-mouse immunoglobulins was added. Incubation conditions and washes were as described for the first antibody. As the substrate for peroxidase, 200 RI of 100 mM citrate-200 mM phosphate buffer (pH 5.0) containing 40 mg of o-phenylenediamine hydrochloride per 100 ml and 40 RI of hydrogen peroxide per 100 ml was added. The reaction was developed for 15 min and stopped by adding 100 ,l of 4 M sulfuric acid. A492 was determined with an ELISA plate reader. Immunofluorescence. About 5 x 108 P. putida 2440 cells were harvested by centrifugation and washed twice with 3 ml of PBS-azide. Washed cells were then incubated for 30 min with hybridoma supernatants or ascites fluids at appropriate dilutions. All steps were carried out at 4°C. The cells were then washed twice and incubated with 25 ,ul of FITC-

MAbs AGAINST P. PUTIDA CELL SURFACE

2979

conjugated rabbit anti-mouse immunoglobulin diluted to 1/20 in PBS-azide for 30 min. The cells were washed twice and resuspended in PBS-azide containing 10% (vol/vol) glycerol and 100 ,ug of o-phenylendiamine per ml to prevent fading (25). Immunofluorescence was evaluated with a Zeiss epifluorescence microscope (magnification, x 1,000). Polypeptide analysis. Cell lysates were prepared by mixing suspensions of cells with sample buffer (24). After the viscosity of the samples was reduced by sonication (35 W, 2 x 30 s), samples were boiled for 10 min and 100 ,ug of protein was electrophoresed. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed by the Laemmli method (24). Running gels of 17.5% (wt/vol) acrylamide and stacking gels of 5% (wt/vol) acrylamide were poured. Proteins in Western blots (immunoblots) were stained with amido black (46). Protein concentrations were determined as described by Lowry et al. (26), with bovine serum albumin as the reference. Analysis of LPS patterns. For analysis of LPS, an overnight culture of P. putida 2440 was harvested by centrifugation, washed twice in PBS, and resuspended in this buffer to an optical density at 525 nm (OD525) of 0.5 to 0.6. A 1.5-ml volume of the cells in PBS was centrifuged and suspended in 50 pl of SDS-PAGE lysing buffer (24). Pronase digestion was then carried out at 60°C for 60 min (16). LPS were separated electrophoretically under the conditions described above for SDS-PAGE, and LPS were stained by the Ag LPS- and protein-staining method of Hitchcock and Brown (16). Dot blotting method. Dot blotting was performed with a Bio-Dot Microfiltration Apparatus (Bio-Rad). A nitrocellulose membrane was soaked in PBS for 30 min. One hundred microliters of PBSTA with 1% albumin was added per well to block the filter. Then, 50 ,ul of serial dilutions (from 105 to 10-1) of P. putida 2440 in PBSA with 1% albumin was added per well; cells were then blocked by adding 100 ,ul of PBSA with 1% albumin. The cells were washed three times with PBST, and then 50 ,ul of ascitic liquid appropriately diluted in PBSTA with 1% albumin was added. All subsequent steps were as for the ELISA, except that PBS and PBST were always supplemented with 1% (wt/vol) albumin. The reaction was developed outside the device with 40 mg of 3,3'diaminobenzidine and 40 lI of hydrogen peroxide per 100 ml of PBS and stopped with distilled water. Immunodetection of proteins and LPS. Electrophoretically separated polypeptides and pronase-resistant cellular components were transferred to nitrocellulose paper by electroblotting (250 mA for 5 h for proteins and 3.5 h for LPS) (43). Blots were dried at room temperature, and only those containing LPS were baked at 80°C for 2 h. Subsequently, blots were blocked with PBSA with 2% albumin for 1 h and incubated with 103-fold-diluted ascites fluids in PBSTA with 1% albumin for 2 h (or overnight at 4°C). After five washes in PBST for 6 min each, blots were incubated with 103-folddiluted peroxidase-conjugated rabbit anti-mouse immunoglobulins in PBSTA with 1% albumin for 2 h. The blots were then washed three times with PBST, once with PBS, and then once with TB. Western blots were developed by using 100 mg of 4-chloro-1-naphthol in 10 ml of 1-propanol and 250 pl of hydrogen peroxide per 100 ml of TB. LPS blots were developed by using either 4-chloro-1-naphthol or 50 mg of 3,3'-diaminobenzidine and 75 ,ul of hydrogen peroxide in 100 ml of TB. Reactions were stopped in TB. In situ analysis ofP. putida by 4,6-diamidino-2-phenylindole (DAPI) staining and immunofluorescence. The experimental design for release of monocultures of P. putida(pWWO) in

2980

RAMOS-GONZALEZ ET AL.

mesocosms in a eutrophic lake (Plussee, northern Germany) is described elsewhere (6). Bacterial cells were fixed after sampling with formaldehyde (final concentration, 2%). Cells were first stained with DAPI (Sigma, St. Louis, Mo.). DAPI staining was performed in the water samples with a final DAPI concentration of 6 ng/ml. The DAPI-stained cells were filtered on a black Nuclepore filter (25-mm diameter, 0.2-,um pore size; Nuclepore Corp.). For immunofluorescence staining on the filter, 0.5 ml of the primary antibody (MAb 7.3B) was applied for 1 h at a dilution of 1:100 in PBSA with 0.5% albumin. The FITC-conjugated anti-mouse immunoglobulin M antibody (Sigma Chemical Co.) was applied twice (0.5 ml each time) for 15 min at a 1:120 dilution in PBS. For microscopic observations, membrane filters were covered with a solution of 20% glycerol in 200 mM Tris-HCl, pH 8.3. To retard fading, 1% 1,4-diazabicyclo(2,2,2)octane (DABCO; Sigma Chemical Co.) was added to the mounting medium. The cells were observed with a Zeiss epifluorescence microscope equipped with an HBO 50-W light source. Photographs of DAPI-stained cells were made with the Zeiss filter set at 01 (bandpass 365-11 exciter filter, FT 395 chromatic beam splitter, LP 397 barrier filter). FITC immunofluorescence photographs were made with the filter set at 09 (bandpass 450-490 exciter filter, FT 510 chromatic beam splitter, bandpass 515-565 barrier filter). Exposures with subsequent use of both filter sets enabled direct comparison of DAPI-stained and fluorescent-antibody-stained cells. In double-exposure pictures, DAPI- and FITC-stained cells are slightly displaced; therefore, the different staining versions of the same cells can be seen beside each other. RESULTS Isolation of MAbs against P. putida 2440 surface determinants and specificity profile. Mice were immunized with P. putida 2440 as described in Material and Methods. Supernatants from 366 hybridomas were screened by indirect immunofluorescence for production of immunoglobulins against whole cells of P. putida 2440. Supernatants from 11 wells gave a positive reaction. The 11 hybridomas that produced immunoglobulins against P. putida 2440 were amplified in mice by inducing ascites tumors. Serial dilutions (500 to 5 x 104) of the 11 ascites fluids were used in ELISAs to test the reactivities of the immunoglobulins against (i) several strains of P. putida, (ii) different bacteria belonging to the pseudomonad group, and (iii) different bacterial phyla, including cyanobacteria, flavobacteria, gram-positive eubacteria, and purple bacteria. Among the former, several bacteria belonging to the a (e.g., Rhizobium sp.), i (e.g., Alcaligenes eutrophus), and -y (enteric bacteria) subdivisions were considered (48). In all, about 30 species of different eubacteria were considered (see below for further details). Three ascites fluids were highly specific against P. putida, while the other eight ascites fluids produced cross-reactions with several bacteria and were discarded. The three hybridomas that produced highly specific antibodies were cloned by dilution, and the clones secreting MAbs were used for production of ascites fluids. These MAbs belonged to the immunoglobulin M class and were called 7.3B, 7.4D, and 7.5D. In ELISAs, it was found that these three MAbs recognized P. putida 2440 with and without the TOL plasmid or recombinant TOL plasmids. P. putida 43 was recognized only by MAb 7.3B, and none of the MAbs showed cross-

J. BACTERIOL. 1

2

3

4

5

6

7

8

9

10

11

12

10I

104 3

10

2

10

10 0

10

-1

10 FIG. 1. Dot blot semiquantitative detection of P. putida 2440. A bacterial culture of P. putida 2440 grown on LB medium was serially diluted and filtered through Millipore HA filter paper by using a dot blot microfiltration apparatus so that the numbers of CFU per spot were approximately those indicated on the left. Blocking of cells and incubation with different MAbs diluted 103 (lanes 1 to 4), 104 (lanes 5 to 8), and 5 x 104 (lanes 9 to 12) were as described in Material and Methods. Lanes: 1, 5, and 9, control immunoglobulin M; 2, 6, and 10, MAb 7.3B; 3, 7, and 11, MAb 7.4D; 4, 8, and 12, MAb 7.5D. Other conditions were as described in Materials and Methods.

reactivity with P. putida 29. None of the MAbs recognized pseudomonads such as Pseudomonas sp. strain B13, P. aeruginosa, P. fluorescens, P. syringae, P. oleovorans, and P. stutzeri. These MAbs did not recognize other purple soil bacteria, such as R. meliloti, R. leguminosarum, E. chrysanthemi, A. vinelandii, A. tumefaciens, A. calcoaceticus, and A. eutrophus; members of the family Enterobactenaceae, such as E. coli, S. typhimurium, K pneumoniae, and K aerogenes; or gram-positive eubacteria, such as B. subtilis and Corynebacterium sp.; cyanobacteria, such as Synechocystis sp. strain PPC 6803; and flavobacteria, such as F. marinotypicum. In addition to the above-described ELISAs, immunofluorescence assays were also carried out to test the crossreactivity of these MAbs. None of the above-mentioned bacteria gave a positive response, except P. putida 2440, which confirmed the high specificity of the MAbs against P. putida 2440. The potency of the MAbs was tested against P. putida 2440 in ELISAs. It was found that the ascites fluids could be diluted up to 5 x 104-fold before reaching background levels. Ascites fluids were routinely used at 1,000-fold dilutions. Solid-phase semiquantitative assay. To establish the minimum number of P. putida 2440 bacteria that could be detected by the specific MAbs, a dot blot ELISA was developed. Serial dilutions of a culture of P. putida 2440 grown on LB medium were filtered through a nitrocellulose membrane so that the number of bacteria per well ranged from 105 CFU to less than 1 CFU per well. The immunoassay revealed that spots containing as few as 102 CFU were consistently positive (Fig. 1). It is worth noting that in ELISAs, the minimum number of cells corresponded to about 104 CFU added per well of microtiter plates for MAb 7.3B and about 105 for MAbs 7.4D and 7.5D with 103-fold diluted MAbs (Fig. 2). Presence of the antigen under different growth conditions, in soil, and in lake water samples. The potential use of these MAbs as tracking reagents is based on expression and stable maintenance of the antigen under different growth condi-

MAbs AGAINST P. PUTIDA CELL SURFACE

VOL. 174, 1992

C"

a)

5xlo8

1o6

107

105

104

103

n cell/well FIG. 2. Counting of P. putida 2440 cells in ELISAs. Serial dilution [s of a P. putida 2440(pWWO) culture grown on LB were set up. CelIs were fixed to polyvinyl chloride microtiter plates, and ELISAss were done as described in Materials and Methods. A492 was determi ned. Symbols: 0, MAb 7.3B; U, MAb 7.4D; A, MAb 7.5D; *, cont .rol immunoglobulin M.

E '

2981

tions. We first studied expression of the antigen in P. putida 2440(pWWO) in different growth phases in rich media and minimal medium with different C sources by using microtiter plate ELISAs. With P. putida 2440(pWWO) grown on LB medium, it was found that regardless of the growth phase, cells were recognized by all three MAbs (Fig. 3A). However, with cells grown on minimal medium with glucose, succinate, and 3-methylbenzoate as a C-source, different patterns were found, depending on the MAb and the growth phase. Cells of P. putida 2440(pWWO) grown on minimal media and in different growth stages (early and late exponential phase and stationary phase) reacted strongly with MAb 7.3B. In contrast, cells grown in 3-methylbenzoate were not recognized (or were weakly recognized) by MAbs 7.4D and 7.5D in the early exponential phase as well as in the stationary phase; in the late exponential phase, the OD492 was about one-fourth of that of cells grown in LB. Cells grown in succinate were not recognized by MAbs 7.4D and 7.5D in the stationary phase (Fig. 3C). We also tested whether expression of the antigen was maintained after introduction of P. putida 2440 bearing recombinant TOL plasmid pWWO-EB62 in soils. P. putida(pWWO-EB62) was grown on minimal medium withp-ethylbenzoate as the sole C source and introduced into a sandy loam-fluvisol soil for periods of about 1 month. In situ determinations were hampered by the soil background immunofluorescence and the smaller size of the cells kept in soils. Cells were then recovered from the soil on selective solid agar plates and grown in LB for ELISA and immuno-

Nj

c 0 I"

4T

0.5

E C4 .4

FIG. 3. Expression of P. putida 2440 cell surface antigen determinants under different growth conditions. P. putida(pWWO) cells were grown on LB (A) or minimal medium with glucose (B), succinate (C), or m-toluate (D). Cultures in the early exponential phase (E; OD660, about 0.3), late exponential phase (L; OD60, about 1.5), and stationary phase (S; OD66, >3) were harvested and washed in PBS, and about 5 x 10' cells were fixed to a microtiter plate. After the ELISA, A492 was determined. Symbols: 0, MAb 7.3B; U, MAb 7.4D; A, MAb 7.5D.

2982

RAMOS-GONZALEZ ET AL.

J. BACTERIOL.

FIG. 4. Double exposure of DAPI- and FITC-stained P. putida(pWW0) released in lake mesocosms. P. putida(pWWO) was grown on LB medium and released in a natural lake mesocosm. Lake water from the mesocosm was sampled immediately after introduction of P. putida(pWWO) cells. Cells were stained with DAPI and FITC. Double exposures with different sets of filters for DAPI- and FITC-stained cells were done as described in Materials and Methods. Both P. putida and the autochthonous bacteria were stained with DAPI (blue), while only P. putida was stained by FITC-conjugated MAbs (yellow-green rods). The average size of the P. putida cells shown was 0.9 by 3.4 W.m.

fluorescence assay, which showed that these cells were recognized by all three MAbs. The MAbs raised against P. putida 2440 were successfully used to track the bacterium in aquatic ecosystems. We used MAb 7.3B to track P. putida(pWWO) after release in mesocosms installed in a eutrophic lake (Plussee) in northern Germany. Figure 4 shows well-fed P. putida cells just after introduction into lake water. In the double exposure, all bacteria in the lake water are shown as blue (DAPI-stained) cells; the P. putida cells are easily distinguishable from autochthonous bacteria by their FITC immunofluorescence staining. Figure 5 shows FITC-stained P. putida cells 1 week after introduction into the lake water mesocosms without nutrient addition. A morphological change from rod to small coccoid cells was observed, often with large agglomerations like that shown in Fig. 5. Therefore, MAb 7.3B turned out to be a reliable tool for tracking of P. putida bacteria since it showed no unspecific binding to autochthonous bacteria. Molecular identification of surface antigens which react with MAbs. LPS prepared from P. putida 2440 were subjected to SDS-PAGE and silver staining, which revealed a series of bands ranging from about 45 to 55 kDa (Fig. 6). In LPS blots, MAb 7.3B revealed the same banding pattern as that observed with SDS-PAGE after silver staining (Fig. 6) while MAbs 7.4D and 7.5D and a negative control mouse immunoglobulin M did not reveal any band (data not shown). These results suggest that MAb 7.3B was directed against P. putida 2440 LPS. To determine which cell antigens were recognized by MAbs 7.4D and 7.5D, total cell protein was prepared from cultures of P. putida 2440. After separation of the proteins

by SDS-PAGE and Western blotting, the immunological assays revealed that MAbs 7.4D) and 7.5D recognized a single protein band with an apparent molecular mass of 40 kDa (Fig. 7). In some immunoblots, in addition to the 40-kDa band, a band of about 65 kDa was also observed, which may have resulted from association of the 40-kDa protein with another protein. DISCUSSION MAbs provide an excellent tool for tracking of genetically engineered microorganisms in the environment. However, successful use of antibodies in the environment can be affected by a number of factors, namely, specificity and cross-reactivity, sensitivity, and loss of bacterial cell antigens under environmental conditions. MAbs against surface determinants of P. putida 2440, the host bacterium for recombinant TOL plasmid and other recombinant plasmids (4, 9, 35), were sought with a screening strategy that involved an anti-antibody coupled to FITC and intact microorganisms as the antigen source. To select highly specific MAbs, a large battery of eubacteria was used in the screening. The screening included closely related bacteria, e.g., species belonging to three different Pseudomonadaceae family rRNA groups (31), soil bacteria found in the same habitats as P. putida, and a number of nonrelated bacteria belonging to four different phyla according to Woese's classification (48). Of the 11 MAbs, 3 were highly specific for P. putida 2440 as they did not recognize bacteria other than P. putida. However, different degrees of stringency were found in these MAbs: MAbs 7.4D and 7.5D

MAbs AGAINST P. PUTIDA CELL SURFACE

VOL. 174, 1992

1

2

3

4

2983

5

94m". 6743. 43mo I

14w FIG. 5. FITC-stained P. putida cells after release in a lake mesocosm without addition of nutrients. Samples were taken 6 days after introduction of P. putida(pWWO) into the mesocosm without nutrient addition. Other conditions were as described in the legend to Fig. 4 and in Materials and Methods. The average size of the P. putida cells was 0.6 by 0.8 p.m.

recognized only P. putida 2440; in contrast, MAb 7.3B reacted with other P. putida strains, e.g., strain 43, although it did not react with P. putida 29. Therefore, these MAbs may be useful in establishing serotypes within P. putida species (28). Our findings with Western and LPS blots elucidated the molecular nature of the P. putida antigens recognized by the

1

2

3

4

170w973, 55wI

FIG. 6. Silver staining pattern of LPS of P. putida 2440 and its immunological recognition by MAb 7.3B. Pronase-resistant material was electrophoretically separated by SDS-PAGE and silver stained (lane 2). The material was electroblotted onto nitrocellulose paper and developed with MAb 7.3B (lane 3) or an immunoglobulin M control (lane 4). Molecular weight standards (103) are shown in lane 1.

FIG. 7. Total cell protein of P. putida 2440 and immunological detection of a 40-kDa band by MAbs 7.4D and 7.5D. Total cell protein was electrophoretically separated by SDS-PAGE and transferred to nitrocellulose paper. Lanes: 1, molecular weight markers (103); 2, material stained with amido black; 3 to 5, immunological reactions with MLAbs 7.4D and 7.5D and a control immunoglobulin M, respectively. The arrow on the right points out the 40-kDa band.

MAbs. LPS and a 40-kDa protein constitute the antigenic components of the cell surface recognized by MAb 7.3B (Fig. 6) and by 7.4D and 7.5D (Fig. 7), respectively. The typical ladder of pronase-resistant cellular material observed in Fig. 6 has been identified in other gram-negative bacteria as a population of LPS molecules whose heterogeneity is due to a variable number of repeating 0-antigen units (7, 11, 13, 18, 32, 38, 39) that probably make up the antigen recognized by MAb 7.3B. MAbs 7.4D and 7.5D both recognized a 40-kDa protein. Our studies did not allow us to discern whether these MAbs recognized different epitopes of the molecule, although the protein seems to be a major one in nondenaturing gels and is apparently highly antigenic, as 2 of 11 MAbs recognized the same target. We are currently cloning the gene that encodes this antigen. For risk assessment studies, the sensitivity as well as the specificity of MAbs is important. We have developed a semiquantitative dot blot immunoassay for bacteria in liquid media. This allowed us to detect as few as 100 cells per spot by using peroxidase-conjugated antibody against the antibody that recognized P. putida 2440 (Fig. 1). An intrinsic limitation of this technique is the turbidity of the samples, which may limit maximum assay volume. Prefiltration to eliminate particles in suspension may overcome this limitation. This assay is also of limited use for bacteria introduced into soils or sediments because of intrinsic fluorescence backgrounds. However, selective enrichments similar to the immunocapture methods described by Morgan et al. (27) and Wipat et al. (47) may be useful for MAb-based quantification of genetically engineered microorganisms in soils. In addition, the combination of MAb technology with flow cytometry may improve the sensitivity of genetically engineered microorganism detection (2, 3). The usefulness of MAbs as tracking reagents depends on expression of the antigen under environmental conditions. In

2984

RAMOS-GONZALEZ ET AL.

the laboratory, we studied the expression of the antigens recognized by MAbs 7.3B, 7.4D, and 7.5D under different growth conditions. The titer of the 40-kDa protein antigen recognized by MAbs 7.4D and 7.5D varied depending on culture conditions and growth phase. The 40-kDa antigen was easily detectable when P. putida was grown on LB medium or minimal medium with glucose, but it was less readily detected in cells grown on minimal medium with succinate and methylbenzoate as a C source and became negligible on these media in the stationary phase. This could be due to either the lack of expression of the antigen under these conditions or inaccessibility of the antigen in vivo owing to alterations in the cell envelope (41). Many cases of cell protein expression are affected by growth conditions (33), and this could indeed be the case for this protein. We isolated TnS::phoA mutants of P. putida 2440 which exhibit the insertion in a gene that encodes a protein involved in LPS synthesis. The 40-kDa protein was synthesized in these P. putida TnS::phoA mutants, as the protein was detected in Western blots, although it is undetectable in vivo (36). In contrast to the antigen recognized by MAbs 7.4D and 7.5D, the target of MAb 7.3B was always expressed at high levels: (i) its detection was independent of the bacterial growth phase and the nutrient used for culture (Fig. 3), (ii) dot blot assays consistently allowed us to detect as few as 100 cells per spot (Fig. 1), and (iii) immunofluorescence assays gave a strong signal; hence, these results were appropriate for flow cytometry studies. Therefore, we conclude that MAb 7.3B is a useful tool for tracking P. putida in natural settings. We used MAb 7.3B to track P. putida(pWWO) released in lake mesocosms. The double staining procedure based on DAPI staining and indirect immunofluorescence allowed us to distinguish autochthonous bacteria from P. putida (pWW0) (Fig. 4). The indigenous bacteria appeared after DAPI staining as coccoid organisms, while the well-fed P. putida(pWWO) bacteria appeared as rods. After 48 h in a nutrient-poor environment, the P. putida(pWWO) bacteria became coccoid and tended to form clumps like those shown in Fig. 5. At this stage, our CFU counts revealed that the number of P. putida(pWWO) bacteria was similar to the number after introduction (6). The shrinkage of P. putida(pWWO) cells under starvation and their relatively high surface-to-volume ratio may offer survival advantages at very low concentrations of limiting substrates (23). The clump structure tended to be maintained after prolonged incubation in the nutrient-poor environment. This confirms the usefulness of MAb 7.3B not only as a tracking reagent but also as a tool for studies of population structure in a natural setting. ACKNOWLEDGMENTS This work was supported by a Biotechnology Action Programme contract from the European Communities (BAP-0411-E). M.I.R.-G. was supported by a fellowship from the Spanish Ministry of Education. We thank R. Hermoso and J. Fonolla, Jr., for help in early stages of this work and continuous support. We thank J. Casadesus, A. Muro, and F. Uruburu for strains. We also thank M. D. Fandila and M. M. Fandila for typing the manuscript, M. Martinez for art work, and M. Burgos for photographic work. REFERENCES 1. Abril, M.-A., C. Michan, K. N. Timmis, and J. L. Ramos. 1989. Regulator and enzyme specificities of the TOL plasmid-encoded upper pathway for degradation of aromatic hydrocarbons and expansion of the substrate range of the pathway. J. Bacteriol. 171:6782-6790.

J. BACTERIOL.

2. Amann, R. I., B. J. Binder, R. J. Olson, S. W. Chisholm, R. Devereux, and D. A. Stahl. 1990. Combination of 16S rRNAtargeted oligonucleotide probe with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56:1919-1925. 3. Amann, R. I., L. Krumholz, and D. A. Stahl. 1990. Fluorescentoligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J. Bacteriol. 177:762-770. 4. Bagdasarian, M., and K. N. Timmis. 1982. Host vector system for gene cloning in Pseudomonas. Curr. Top. Microbiol. Immunol. 96:47-67.

5. Beringer, J. E. 1974. R factor transfer in Rhizobium leguminosarum leguminosarum. J. Gen. Microbiol. 84:188-198. 6. Brettar, I., M. I. Ramos-Gonz6lez, J. L. Ramos, and M. Hofle. Unpublished data. 7. Carlson, R. W. 1984. Heterogeneity of Rhizobium lipopolysaccharides. J. Bacteriol. 158:1012-1017. 8. Casadesu's, J., and J. Olivares. 1979. Rough and fine linkage mapping of the Rhizobium meliloti chromosome. Mol. Gen. Genet. 174:203-209. 9. Delgado, A., M. G. Wubbolts, M.-A. Abril, and J. L. Ramos. 1992. Nitroaromatics are substrates for the TOL plasmid upperpathway enzymes. Appl. Environ. Microbiol. 58:415-417. 10. Dorn, E., M. Hellwig, W. Reineke, and H. J. Knackmuss. 1974. Isolation and characterization of a 3-chlorobenzoate degrading pseudomonad. Arch. Microbiol. 99:61-70. 11. Fink, J. M., and J. F. Zissler. 1989. Characterization of lipopolysaccharide from Myxococcusxanthus by use of monoclonal antibodies. J. Bacteriol. 171:2028-2032. 12. Franklin, F. C. H., M. Bagdasarian, M. M. Bagdasarian, and K. N. Timmis. 1981. Molecular and functional analysis of the TOL plasmid pWWO from Pseudomonas putida and cloning of genes for the entire regulated aromatic ring meta-cleavage pathway. Proc. Natl. Acad. Sci. USA 78:7458-7462. 13. Goldman, C., and L. Leive. 1980. Heterogeneity of antigenicside-chain length in lipopolysaccharide from Escherichia coli 0111 and Salmonella typhimurinum LT2. Eur. J. Biochem. 107:145-153. 14. Guyon, P., M.-D. Chilton, A. Petit, and J. Tempe. 1980. Agropine in "null-type" crown gall tumors: evidence for generality of the opine concept. Proc. Natl. Acad. Sci. USA 77:2693-2697. 15. Hermoso, R., A. Chueca, J. J. Lazaro, and J. Lopez-Gorge. 1987. An immunological method for quantitative determination of photosynthetic fructose-1,6-biphosphatase in leaf crude extracts. Photosynth. Res. 14:269-278. 16. Hitchcock, P. J., and T. M. Brown. 1983. Morphological heterogeneity among Salmonella lipopolysaccharide chemotypes in silver-stained polyacrylamide gels. J. Bacteriol. 154:269277. 17. Huang, H.-C., R. Schuurink, T. P. Denny, M. M. Atkinson, C. J. Baker, I. Yucel, S. W. Hutcheson, and A. Collmer. 1988. Molecular cloning of a Pseudomonas syringae pv. synngae gene cluster that enables Pseudomonas fluorescens to elicit the hypersensitive response in tobacco plants. J. Bacteriol. 170: 4748-4756. 18. Jann, B., K. Reske, and K. Jann. 1975. Heterogeneity of lipopolysaccharides. Analysis of polysaccharide chain lengths by sodium dodecylsulfate-polyacrylamide gel electrophoresis. Eur. J. Biochem. 60:239-246. 19. Kagan, S. A., and N. J. Brewin. 1985. Mutagenesis of a Rhizobium plasmid carrying hydrogenase determinants. J. Gen. Microbiol. 131:1141-1147. 20. Kennedy, C., R. Gamal, R. Humphrey, J. Ramos, H. Brigle, and D. Dean. 1986. The nifH, nifM and nifN genes of Azotobacter vinelandii: characterization by Tn5 mutagenesis and isolation from pLAFR1 gene banks. Mol. Gen. Genet. 205:318-325. 21. Kohler, G., and C. Milstein. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature (London) 256:495-497. 22. Kooistra, J., B. Vosman, and G. Venema. 1988. Cloning and characterization of a Bacillus subtilis transcription unit involved

VOL. 174, 1992

23.

24.

25. 26. 27.

28. 29.

30.

in ATP-dependent DNase synthesis. J. Bacteriol. 170:47914797. Kuenen, J. G., J. Boonstra, H. G. J. Schroder, and H. Veldkamp. 1977. Competition for inorganic substrates among chemoorganotrophic and chemolithotrophic bacteria. Microb. Ecol. 3:119-130. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Lanier, L. L., and N. L. Warner. 1981. Paraformaldehyde fixation of hematopoietic cells for quantitative flow cytometry (FACS) analysis. J. Immunol. Methods 47:24-30. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275. Morgan, J. A. W., C. Winstanley, R. W. Pickup, and J. R. Saunders. 1991. The rapid immunocapture of Pseudomonas putida cells from lake water using bacterial flagella. Appl. Environ. Microbiol. 57:503-509. Nandapalan, N., and B. J. Chang. 1989. Production and characterization of monoclonal antibodies to Aeromonas sobria surface antigens. FEMS Microbiol. Immunol. 47:515-524. Neidle, E. L., C. Hartnett, S. Bonitz, and L. N. Ornston. 1988. DNA sequence of the Acinetobacter calcoaceticus catechol 1,2-dioxygenase I structural gene catA: evidence for evolutionary divergence of intradiol dioxygenases by acquisition of DNA sequence repetitions. J. Bacteriol. 170:4874-4880. Nievwkoop, A. J., and R. A. Bender. 1988. RNA polymerase as a repressor of transcription in the hut(P) region of mutant Klebsiella aerogenes histidine utilization operons. J. Bacteriol.

170:4986-4990. 31. Palleroni, N. J. 1986. Taxonomy of pseudomonads, p. 3-25. In J. R. Sokatch and L. N. Ornston (ed.), The bacteria, vol. X. Academic Press, Inc., New York. 32. Pieper, D. H., K. H. Engesser, R. H. Don, K. N. Timmis, and H. J. Knackmuss. 1985. Modified ortho-cleavage pathway in Alcaligenes eutrophus JMP134 for the degradation of 4-methylcatechol. FEMS Microbiol. Lett. 29:63-67. 33. Raibaud, O., and M. Schwartz. 1984. Positive control of transcription initiation in bacteria. Annu. Rev. Genet. 18:173-206. 34. Ramos, J. L., and K. N. Timmis. 1987. Experimental evolution of catabolic pathways of bacteria. Microbiol. Sci. 4:228-237. 35. Ramos, J. L., A. Wasserfallen, K. Rose, and K. N. Timmis. 1987. Redesigning metabolic routes: manipulation of TOL plasmid pathway for catabolism of alkylbenzoates. Science 235:593-596.

MAbs AGAINST P. PUTIDA CELL SURFACE

2985

36. Ramos-Gonzalez, M. I. 1992. Unpublished data. 37. Rippka, R., J. Dervelles, J. B. Waterbury, M. Merdman, and R. Y. Stainer. 1979. Genetic assignments, strain histories and properties of pore cultures of cyanobacteria. J. Gen. Microbiol. 111:1-61. 38. Rivera, M., and E. J. McGroarty. 1989. Analysis of a commonantigen lipopolysaccharide from Pseudomonas aeruginosa. J. Bacteriol. 171:2244-2248. 39. Ryan, J. M., and H. E. Conrad. 1974. Structural heterogeneity in the lipopolysaccharide of Salmonella newington. Arch. Biochem. Biophys. 162:530-535. 40. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 41. Sen, K., and H. Nikaido. 1991. Lipopolysaccharide structure required for in vitro trimerization of Escherichia coli OmpF porin. J. Bacteriol. 173:926-928. 42. Tamaki, S. J., S. Gold, M. Robeson, S. Manulis, and N. T. Keen. 1988. Structure and organization of the pel genes from Erwinia chrysanthemi EC16. J. Bacteriol. 170:3468-3478. 43. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76:4350-4354. 44. Uruburu, F. 1990. Spanish Type Culture Collection catalogue of strains, 3rd ed. Burjasot, Valencia, Spain. 45. Viebrock, A., and W. G. Zumft. 1988. Molecular cloning, heterologous expression, and primary structure of the structural gene for the copper enzyme nitrous oxide reductase from denitrifying Pseudomonas stutzeri. J. Bacteriol. 170:4658-4668. 46. Wilson, C. M. 1979. Studies and critique of amido black 10B, Coomassie blue R, and fast green FCF stains for proteins after polyacrylamide gel electrophoresis. Anal. Biochem. 96:263278. 47. Wipat, A., E. M. Wellington, and V. A. Saunders. Detection systems for streptomyces. In E. M. Wellington and J. D. van Elsas (ed.), Genetic interactions between microorganisms in natural environments, in press. Manchester University Press, Manchester, England. 48. Woese, C. R. 1978. Bacterial evolution. Microbiol. Rev. 51:221271. 49. Worsey, M. J., and P. A. Williams. 1974. Metabolism of toluene and xylenes by Pseudomonasputida (arvilla) mt-2: evidence for a new function of the TOL plasmid. J. Bacteriol. 124:7-13.