Natural Genetic Transformation in Monoculture Acinetobacter sp. Strain BD413 Biofilms

Horizontal gene transfer by natural genetic transformation inAcinetobacter sp. strain BD413 was investigated by using gfpcarried by the autonomously replicating plasmid pGAR1 in a modelmonoculture biofilm. Biofilm age, DNA concentration, and biofilmmode of growth were evaluated to determine their effects onnatural genetic transformation. The highest transfer frequencieswere obtained in young and actively growing biofilms when highDNA concentrations were used and when the biofilm developedduring continuous exposure to fresh medium without the presenceof a significant amount of cells in the suspended fraction.Biofilms were highly amenable to natural transformation. Theydid not need to advance to an optimal growth phase which ensuredthe presence of optimally competent biofilm cells. An exposuretime of only 15 min was adequate for transformation, and theaddition of minute amounts of DNA (2.4 fg of pGAR1 per h) wasenough to obtain detectable transfer frequencies. The transformabilityof biofilms lacking competent cells due to growth in the presenceof cells in the bulk phase could be reestablished by starvingthe noncompetent biofilm prior to DNA exposure. Overall, theevidence suggests that biofilms offer no barrier against effectivenatural genetic transformation of Acinetobacter sp. strain BD413.

INTRODUCTION

Top

Introduction

A great variety of synthetic chemicals have been released intothe environment, thereby causing serious pollution. Biologicaltreatment processes offer a way to reduce the amounts of xenobioticcompounds in the environment. Bioremediation (treatment of wastewith microorganisms) (49) and phytoremediation (treatment ofpolluted soil with plants) (17) are regarded as cost-efficienttechnologies for the clean-up of polluted waters or soils. Manywastewater treatment plants rely on biofilms. One of the advantagesof biofilm reactors is their compactness. The small footprintareas of biofilm reactors are due to two factors, high volumetricbiomass concentrations (29) and high retention times (6). Theextended retention time of bacteria growing near the substratumis the reason why a greater variety of microorganisms can developcompared to the variety in planktonic cultures (8). Anotherreason is the formation of microniches due to concentrationgradients of nutrients and electron acceptors. Chemical pollutantscan adsorb or be degraded further, contributing to the zonationon a microscale. Hence, extended retention times of solids (6),species diversity, higher local nutrient concentrations (14),and high retention times for recalcitrant compounds due to adsorptionin the extracellular polymeric matrix (48) in biofilm reactorsfacilitate remediation of synthetic pollutants. Furthermore,biofilms allow interspecies interactions by signaling and nutrientcycling (20) and offer resistance to toxicant (30), physical(13), or environmental stresses (44).

However, for a system to be able to degrade xenobiotic compounds,the biocatalysts that are needed for degradation of a specificcompound need to be present in the system (22). If reactorslack the desired biocatalysts, they need to be bioaugmentedto obtain efficient degradation. By addition of exogenous optimallyconstructed bacteria (11, 43) or functionally adapted bacterialconsortia (18) and communities (47), enrichment of the totalgene pool can be obtained. Likewise, a reactor can be augmentedby in situ gene transfer (42). It is surprising that in spiteof several successes, few studies have evaluated bioaugmentationby horizontal gene transfer in activated sludge wastewater treatmentreactors (33, 34, 50) or activated sludge microcosms (16, 36,41). When plasmids carrying catabolic genes were integratedinto indigenous bacterial organisms, increased and more rapiddegradation of the target compound was observed (33, 34, 36).Likewise, there is little information available regarding bioaugmentationof biofilm reactors by in situ gene transfer by bacterial conjugation(1, 4, 12, 15, 53). Conjugation inside a biofilm matrix offersa great advantage in terms of both gene transfer frequency (1,15, 19) and subsequent transconjugant stability (15). Still,the effect of in situ natural genetic transformation in orderto obtain bioenhancement in activated sludge- or biofilm-basedbiological process engineering systems has not been studied.

To investigate the feasibility of bioaugmentation by genetictransformation, in situ natural genetic transformation was investigatedwith a model system consisting of biofilm-cultured Acinetobactersp. strain BD413 (24) with the autonomously replicating gfp-carryingplasmid pGAR1 (19) as the transforming DNA. Acinetobacter speciesare ubiquitous, strictly aerobic, nonmotile organisms that canbe isolated from soil, water, or wastewater (23). Acinetobacterstrains can degrade recalcitrant aromatic and alicyclic compounds,as well as some aromatic amino acids, mineral oils, and syntheticpolymers (3, 5, 7, 40). In addition, Acinetobacter strains producebiosurfactants, like emulsan (25, 45) and alasan (2), that enhancethe bioavailability of poorly soluble compounds. Furthermore,Acinetobacter sp. strain BD413 (24) is amenable to gene manipulationby conjugation, transformation, and transduction (23), and thisproperty makes the strain particularly interesting as a toolfor biologically enhancing the catabolic properties of hazardouswaste treatment facilities.

In this study the following conditions were investigated todetermine their effects on natural genetic transformation ina model biofilm: biofilm age, free DNA concentration, and growthof the biofilm in the presence or absence of cells in the bulkfluid.

MATERIALS AND METHODS

Top

Materials and Methods

Bacteria, plasmids, and media.
Acinetobacter sp. strain BD413 (24) was used as the model strainfor evaluation of natural genetic transformation in monoculturebiofilms. pGAR1 was isolated from Escherichia coli strain GM16(19). pGAR1 is pRK415 (26), a Mob⁺ Tra^- Tet^r Inc P1 plasmid,carrying the wild-type gfp (green fluorescent protein [GFP])gene (Clontech, Palo Alto, Calif.) under regulation of a Placpromoter. Induction of the promoter by addition of isopropyl-ß-D-thiogalactopyranosidewas not necessary. The fluorescence intensity of single BD413(pGAR1)cells could be clearly distinguished from the fluorescence intensityof unlabeled single BD413 cells by epifluorescence and confocallaser scanning microscopy (CLSM). Rich Luria-Bertani medium(38) and minimal medium M9 (0.2% gluconate) (46) were used duringtransformation experiments. Standard microbiological techniques(spectrophotometric DNA concentration measurement, bacterialcultivation techniques, DNA extraction and purification) wereperformed as described by Sambrook et al. (46).

Transformation in biofilms.
Transformation of biofilms was performed by the methods describedby Wuertz et al. (54). Biofilms of Acinetobacter sp. strainBD413 were grown in a flow cell containing four separate flowchannels (4 by 4 by 40 mm) for 3 days in rich Luria-Bertanimedium by using a flow rate of 2.4 ml/h after DNA was addedwith minimal medium M9 (0.2% gluconate). Transformation involvedincubation of DNA during 1 h of continuous flow with DNA-containingmedium, unless indicated otherwise. After overnight incubationin minimal medium M9 (0.2% gluconate) without DNA at a flowrate of 2.4 ml/h, which allowed expression of the received gene,biofilms were prepared for microscopic monitoring.

Microscopic monitoring, image acquisition, and data processing.
Biofilm cells in Acinetobacter sp. strain BD413 monoculturebiofilms undergoing transformation were visualized with thegeneral nucleic acid stain Syto 60 (Molecular Probes, Eugene,Oreg.) and were detected with an LSM 410 CLSM (Zeiss, Jena,Germany). The 633-nm laser line and a 665-nm long-pass emissionfilter were used to detect cells stained with Syto 60. The 488-nmlaser line and a 515-nm long-pass emission filter were usedfor detection of cells expressing gfp (Clontech). Automatedimage acquisition and data processing were performed by themethods described by Wuertz et al. (54).

Colocalization experiments.
Pure-culture BD413(pGAR1) biofilms grown for 46 h in selectiveminimal medium (minimal medium M9 containing 0.2% gluconateand 20 µg of tetracycline per ml) were stained with Syto60 to test colocalization of the two signals. Approximately18.5% of the signals in the BD413(pGAR1) biofilm were colocalizedGFP and Syto 60 signals, 22.6% were single Syto 60 signals,and 58.9% were single GFP signals. Therefore, in transformationexperiments the total cell volume was obtained by adding theSyto 60 signals and the GFP signals and subtracting the overlappingsignals. GFP signals were considered transformants. Potentialunderestimation of the 22.6% GFP signals was not consideredduring calculation of the results because it was not possibleto check whether cells still contained the introduced plasmid.When no DNA was added to the medium, no signals were detectablewith CLSM settings for detection of GFP signals in two separatetests. In an additional test, signals were obtained by usingthe 515-nm long-pass emission filter, but no overlap was detectablewith the Syto 60 settings. These false-positive signals mighthave been due to autofluorescent impurities in the inlet mediumand could be clearly distinguished from fluorescence-expressingcells on the basis of form and size. Hence, to avoid overestimationof GFP due to autofluorescent impurities, images had to be checkedmanually to discard possible false-positive signals on the basisof form and size.

Mathematical parameters.
The mathematical equations describing volumes and transformationfrequency are given below.

Calculation of the volume of transformants and the volume ofrecipients was based on the following equation:

where V_X,e is total volume of transformants or recipients (incubic micrometers), X_i is the area covered by cells of interestat position i (in square micrometers), z_i is the distance fromthe substratum at position i (in micrometers), i is the scanningposition in the z direction starting at the biofilm substratum,and e is the last scanned position in a biofilm in the z direction.

The equation was adapted as follows with the trapezoidal ruleto obtain a more correct approximation of the numerical integralas described by Kuehn et al. (28):

The transformation frequency (TF) was the fraction of transformantvolume per total cell volume: TF = V_T,e/V_R,e, where V_T,e isthe volume of transformants obtained with equation 2 (in cubicmicrometers) and V_R,e is the volume of recipients obtained withequation 2 (in cubic micrometers).

Below, transfer frequencies are expressed as the volume of transgeniccells per volume of recipient cells, unless indicated otherwise.

To obtain a reproducible estimate of the normalized distancefrom the substratum, we limited the total cell volume and biofilmthickness to contain 98% of the scanned biomass (V_R,e) startingfrom the substratum towards the biofilm-bulk fluid interfacewithin the biofilm volume investigated. The normalized distancefrom the substratum was calculated by dividing the distancefrom the biofilm attachment surface by the biofilm thickness:d_i = z_i/z_k for V_R,k $>=$ V_R,e x 0.98, where V_R,k is the volume ofrecipients calculated with equation 2 limited to 98% of thetotal scanned biomass (in cubic micrometers), d_i is the normalizeddistance from the substratum at position i (in micrometers),z_i is the distance from the substratum at position i (in micrometers),z_k is the biofilm thickness or distance from the substratumat position k (in micrometers), and k is the position in a biofilmwhere V_R,k reached 98% of V_R,e.

Reproducibility of in situ natural genetic transformation in monoculture Acinetobacter sp. strain BD413 biofilms.
Biofilms should be observed noninvasively and with confidencethat reliable and reproducible results will be obtained. Directin situ detection of large areas is, therefore, desirable. Astudy to determine statistically representative areas of Pseudomonasfluorescens biofilms showed that a minimal area of 1 x 10⁵ µm²should be scanned to obtain reproducible results in biofilminvestigations (27). In the present study a minimal area of2.4 x 10⁵ µm² was scanned to monitor a biofilm volumeof at least 1.2 x 10⁷ µm³. The standard transformationfrequencies with 0.2 µg of pGAR1/ml, derived from fourseparate experiments monitoring a minimum volume of 1.2 x 10⁷µm³, ranged from logarithmic values of -3.5 to -3.1, witha mean value of -3.3 and a standard deviation of 0.15. In contrastto the standard method in which planktonic cells are used (38),the in situ method in which CLSM was used provided reliableand reproducible results.

RESULTS

Top

Results

Effect of biofilm age on natural genetic transformation.
To test the transformability of biofilm cells at different growthstages, experiments were conducted with 0.2 µg of pGAR1DNA/ml (Table 1). The responses of cells to DNA exposure weremeasured with two different exposure times. Flow cell experimentswere performed with 1- and 3-day-old biofilms. The thicknessof Acinetobacter sp. strain BD413 biofilms grown in rich mediumreached a maximum after 2 days for four separately investigatedbiofilms, and after this the biofilm thickness remained at asteady state due to lysis, detachment, and subsequent growth,indicating a mature biofilm stage (results not shown). A 1-day-oldbiofilm could therefore be considered an actively growing biofilm,and a 3-day-old biofilm could be considered a mature biofilm.Although cells in young and growing biofilms are more readilytransformed, the biofilm heterogeneity in 3-day-old biofilmsstill ensured the presence of a significant fraction of competentcells that were able to take up DNA (Table 1; Fig. 1). Furthermore,cells responded within 15 min to the addition of exogenous DNA.

View this table:
[in this window]
[in a new window]
TABLE 1. Transformation frequencies obtained for biofilms of different ages exposed to 0.2 µg of plasmid pGAR1 DNA/ml for different periods of time^a

View larger version (95K):
[in this window]
[in a new window]
FIG. 1. Microscopic images showing pGAR1 transformants in 3-day-old (A) and 1-day-old (B) Acinetobacter sp. strain BD413 biofilms. The black and white images show GFP signals (panels I) or Syto 60-stained cells (panels II). Superimposed single optical images (panels III) show gfp transformants (yellow, green) against a background of recipient Acinetobacter sp. strain BD413 biofilm cells (red). The edges of each image correspond to a length of 90 µm.

Effect of concentration of added free DNA.
To investigate the influence of DNA concentration on transformationfrequency and transformant location in a mature biofilm of thehighly competent organism Acinetobacter sp. strain BD413, naturalgenetic transformation was investigated in biofilms that wereexposed for 1 h to a specific pGAR1 DNA load.

The DNA concentration in the inlet medium ranged from 1 x 10^-9to 1.5 µg of pGAR1/ml. The transformation frequency increasedas a function of DNA concentration within the range of DNA concentrationsinvestigated (Fig. 2).

View larger version (20K):
[in this window]
[in a new window]
FIG. 2. Transformation in monoculture strain BD413 biofilms with pGAR1 DNA at various concentrations in the inlet medium expressed as a semilog plot. The insert shows the same data points on a log-log scale.

The observed relationship between DNA concentration and transformationfrequency corresponds to the results of previous studies involvingbatch transformation experiments (38). However, a saturationpoint was not reached in this study. Further examination withDNA concentrations of 100 µg/ml or more would be neededto see if a saturation point is reached or if transformationkeeps increasing until 100% of the cells are transformed. Suchhigh concentrations were not investigated in this study as theyare unlikely to occur in nature or are not very practical forbioaugmentation of bioreactors.

With low concentrations of pGAR1 in the feed, transformantswere formed at the biofilm attachment surface. Exposure to increasingamounts of pGAR1 resulted in gradual accumulation of transformantsat the bottom of the biofilm (Fig. 3), where the biofilm densitywas the greatest, and not in the middle or upper part of thebiofilm. If it were true that the fraction of competent cellsis homogeneously distributed inside a biofilm, most transformantswould be found in the layers with the highest biofilm densityand the transformation frequency would be equally high throughoutthe biofilm. However, this was observed only 5 of 10 times.Hence, other factors may have contributed to the distributionof frequencies detected.

View larger version (24K):
[in this window]
[in a new window]
FIG. 3. Distribution profiles for the volume of gfp transformants as a function of the normalized distance from the substratum when organisms were exposed to 1 x 10^-9 µg of DNA/ml (•), 1 x 10^-7 µg of DNA/ml ( ${square}$ ), 1 x 10^-4 µg of DNA/ml ( ${blacklozenge}$ ), 1 x 10^-1 µg of DNA/ml ( ${triangleup}$ ), and 1.5 µg of DNA/ml (_*).

Effect of biofilm ontogenesis on natural genetic transformation.
Biofilms can grow either in a batch mode (on the surface ofa shaken container containing a bacterial suspension) or ina continuous mode (by constant feeding of a biofilm grown ona surface with fresh medium in a flow channel). A biofilm grownin a shaken container emerges with suspended cells present inthe bulk fluid, while a biofilm in a flow channel grows in anenvironment continuously rinsed with suspended cells in themedium. All previous experiments were performed with biofilmsgrown in continuously fed flow cells. How would a biofilm thathad developed in the presence of stationary-phase bacterialcells in the bulk phase respond to the addition of free DNA?To answer this question, flow cells were continuously fed withmedium containing stationary-phase Acinetobacter sp. strainBD413 cells to mimic biofilm ontogenesis in the presence ofa batch-grown suspended culture. After this, the biofilms wererinsed with a 0.01 M MgSO₄ to remove any suspended cells fromthe flow channel. Cell-free minimal medium containing 0.1 µgof pGAR1 DNA/ml was added to the flow cells for 1 h. The preparationswere incubated overnight in cell-free minimal medium, and afterthis the biofilms were stained, washed, microscopically investigated,and quantified. No transformation was detected in the biofilms,which had grown in the presence of cells in the bulk fluid.

Next, we investigated the effect of prestarvation on transformationof biofilm cells grown in the presence of cells in the bulkphase. Suspended cells were removed as described above by rinsingwith 0.01 M MgSO₄. The biofilm was then starved by feeding theflow cell either with a continuous supply of minimal mediumM9 without a carbon source or with a salt solution (0.01 M MgSO₄)for 24 h. Subsequent transformation was done with identicalexperimental transformation steps, as described above. Transformationwas observed at low but measurable frequencies (9.1 x 10^-5 and1.9 x 10^-5 for biofilms starved in the presence of minimal mediumM9 and in the presence of the salt solution, respectively).

When biofilms that were grown with cells present in the inletmedium were rinsed with 0.01 M MgSO₄ and subsequently fed withcell-free rich medium for 24 h before transformation, againno transformation was observed in the biofilms.

DISCUSSION

Top

Discussion

Standard transformation experiments in which Acinetobacter sp.strain BD413 is used require cells to be transformed in a stateof competency (38). Acinetobacter sp. strain BD413 reaches competenceat the early log phase. The cells remain competent during thelog phase, and competence drops to almost zero when the stationaryphase is reached (31). In standard batch culture experiments(31) or soil microcosms (35), additional transformation eventswere not detected after a prolonged period that was more than12 h long. Similar to exponentially growing suspended cultures,young and growing biofilms allowed a high frequency of transformationevents. In addition, the biofilms still contained significantfractions of competent cells for at least 3 days after inoculation.The biofilm mode of growth could therefore be compared to thecontinuously exponentially growing steady-state batch or turbidostatcultures of Palmen et al. (37). These authors studied Acinetobactersp. strain BD413 cultured in a continuous mode in order to prolongthe period of competency. After 3 days they were still ableto observe a detectable transformation frequency, but it wasless than the initial transformation frequencies (37); thiswas impossible in transformation experiments performed withAcinetobacter sp. strain BD413 cultured in a batch mode (31,35).

The DNA concentration in the inlet was positively correlatedwith increasing transformation frequency. Natural genetic transformationoccurred readily at high frequencies in monoculture Acinetobactersp. strain BD413 biofilms. When 1.2 µg of pGAR1/ml wasused, transformation frequencies as high as 2.4 x 10^-3 wereobserved. Furthermore, it was possible to obtain detectabletransformation frequencies with minute amounts (as little as1 fg of pGAR1/ml) in biofilms. The lowest tested DNA concentrationused for transformation of Acinetobacter sp. strain BD413 reportedpreviously was approximately 1 ng of DNA/ml (38). Transfer frequencieseven higher than those reported here may be obtainable if afeed with a DNA concentration greater than 10 µg of pGAR1/mlis added to a biofilm. For bioaugmentation via gene transfer,however, it is not necessary to have maximum transformationfrequencies. Even if transfer frequencies are low, transformantsmay undergo cell division, and the transgenic strain could thereforestill establish itself in a reactor system. In such a case,low but significant transfer frequencies should be enough toobtain successful bioaugmentation.

It was interesting that an increased DNA concentration in theinfluent resulted in accumulation of transformants in the biofilmsubstratum. However, it seems odd that most transformation eventstake place at the biofilm base. This observation implies thatfree DNA first has to diffuse through the biofilm before competentcells take up and subsequently replicate the integrated plasmid.It is possible that immobilization of cells was responsiblefor the occurrence of most transformation events at the bottomof the biofilm. The location of transformation events seemedto be simply a matter of chance, and the probability increasedwith increasing cell density at a certain distance from thesubstratum. In the case of monoculture Acinetobacter sp. strainBD413 biofilms the cell density was greatest near the biofilmattachment surface. Therefore, the chance for transformationwas greater near the biofilm substratum. This suggests thatthe chances for transformation to occur are highest in a tightlypacked biofilm (a biofilm with a low porosity value). However,it is not correct that porous biofilms are ill suited for transformation.Young biofilms, for example, are very porous (Fig. 1), and theyallowed transformation at increased rates due to enhanced competencylevels during exponential growth. Porosity plays an importantrole only in mature biofilms. Therefore, it is necessary toinvestigate the true impact of biofilm density and porosityon natural genetic transformation. If density played an importantpart in natural genetic transformation, anthropogenic manipulationof the factors that decrease biofilm porosity could lead toenhanced occurrence of transformation events. van Loosdrechtet al. (51) included flow rate, nutrient loading rate, and growthrate of the biofilm cells as parameters that influence biofilmthickness and porosity. In addition, cell-to-cell signalingmolecules influence biofilm structure (10) and may indirectlyinfluence transformation events.

Another important parameter in this study was the mode of growthin which the biofilm itself emerged. In many biofilm investigationsscientists use batch-grown biofilms in 96-well microtiter plates(9, 39). In contrast to batch-grown biofilms, biofilms grownin flow chambers are in continuous contact with fresh mediumthat does not contain suspended cells. One needs to considerthat biofilms grown in different modes could possess differentqualities. In the present study, differences in the transformabilityof Acinetobacter sp. strain BD413 biofilm cells grown in thepresence and in the absence of cells in the bulk fluid wereobserved. The presence of high numbers of cells in the surroundingmedium inhibited transformation in biofilms (Hendrickx, resultsnot shown). Likewise, when biofilms developed in the presenceof high numbers of cells in the bulk fluid but were rinsed toeliminate cells present in the surrounding medium before theywere exposed to naked DNA, transformation was strongly inhibited.Also, differences in the morphotypes of the bacterial cellswere observed. Acinetobacter sp. displays two different morphotypes,the bacillar morphotype and the coccoid morphotype (21). Inbiofilms fed with cell-free medium, both the bacillar and coccoidforms were detected. In contrast, when biofilms were allowedto emerge in the presence of medium containing suspended bacterialcells, the cells exhibited almost exclusively the bacillar morphotype.

As determined by microscopy, batch-cultured stationary-phaseAcinetobacter sp. strain BD413 cells also displayed the bacillarmorphotype when the organism was grown in rich medium. Hence,it is possible that the lack of transformability in biofilmsgrown in the presence of cells in the bulk fluid is due to increasedattachment of bacterial cells that have entered the stationaryphase (and hence have become noncompetent) in the surroundingmedium.

Biofilms grown in biofilm reactors may encounter many passingfree-floating bacteria. This would discourage bioaugmentationby in situ transformation if the problem of inhibited gene transferdue to the presence of suspended cells during biofilm ontogenesiswere unsolvable. Rinsing a biofilm to remove most cells presentin the drifting fraction, followed by a starvation period, couldreinduce in situ transformation when the cells are exposed toa nutrient-containing substrate with plasmid DNA. James andcoworkers (21) discovered that upon starvation, bacillar cellsrevert to the coccoid form by reduction division, resultingin conservation of biomass but increased cell number. In thisstudy the CLSM investigation revealed that the biofilms treatedwith starvation-inducing medium contained many cells with thecoccoid morphotype, as observed when biofilms were grown inthe absence of cells in the bulk fluid. Thus, all signs suggestthat coccoid cells are the competent cells and bacillar cellsare not competent for DNA uptake. Still, when cell shape waschecked as a function of distance from the substratum, it wasobserved that cells at the biofilm substratum had the bacillarshape, while the prominent morphotype of cells at the biofilm-mediuminterface was coccoid (Hendrickx, data not shown). However,most transformants were formed at the biofilm substratum, wheremost bacillus-shaped cells resided. Further experiments aretherefore needed to establish if other parameters related tothe amount or location of bacillus-shaped or coccoid cells canbe correlated with transformation frequency.

Nevertheless, the observed differences in transformation frequencyin biofilms grown in the presence and in the absence of planktoniccells show that care should be taken in designing biofilm experiments.In addition to the effects of standard environmental parameters(temperature, pH, nutrient content, substratum, moisture content,biotic and abiotic stresses, flow rate, etc.), the results mightdiffer considerably depending on whether the biofilms are grownunder batch or continuous-flow conditions.

It should be noted that the experiments which we performed werenot designed to prove the general feasibility of in situ biofilmcell transformation with any highly competent soil bacteriumin naturally occurring biofilms. Our results do indicate thattransformation has potential as a tool for bioaugmentation ofbiofilm reactors. It is interesting that transformation withAcinetobacter sp. strain BD413 resulted in transformation frequenciesin nonsterile groundwater and wet soil microcosms that wereas high as those obtained under sterile conditions (32). Likewise,transformation of Acinetobacter sp. strain BD413 biofilm cellsembedded in river epilithon was not inhibited by the presenceof indigenous organisms (52). It can be speculated that thepresence of an ambient community in wastewater treatment systemsshould have few negative effects on transformation of Acinetobactersp. strain BD413 cells. Future research should elucidate theefficiency of transforming cells which reside in natural biofilmsinside bioreactors and other systems.

ACKNOWLEDGMENTS

This work was supported by the Research Center for FundamentalStudies of Aerobic Biological Wastewater Treatment, Munich,Germany (grant SFB411), and by the University of California,Davis.

FOOTNOTES

* Corresponding author. Mailing address: Department of Civil and Environmental Engineering, University of California, Davis, One Shields Avenue, Davis, CA 95616. Phone: (530) 754-6407. Fax: (530) 752-7872. E-mail: swuertz{at}ucdavis.edu.

Present adress: Laboratory for Microbiology, Radioactive Wasteand Cleanup, Belgian Nuclear Research Center, Mol, Belgium.

REFERENCES

Top