- Email: [email protected]
Influence of fungal-bacterial interactions on bacterial conjugation in the residuesphere
FEMS Microbiology Ecology 31 (2000) 39^45
www.fems-microbiology.org
In£uence of fungal-bacterial interactions on bacterial conjugation in the residuesphere Gitte SengelÖv a , George A. Kowalchuk b , SÖren J. SÖrensen b
a;
*
a Department of General Microbiology, University of Copenhagen, SÖlvgade 83 H, DK-1307 Copenhagen, Denmark Department of Plant-Microorganism Interactions, Centre for Terrestrial Ecology, Netherlands Institute of Ecology, 6666 ZG Heteren, The Netherlands
Received 25 June 1999; received in revised form 4 October 1999; accepted 7 October 1999
Abstract Conjugal gene transfer among bacteria in the residuesphere (area between decaying plant material and soil) of leaves of barley straw was studied. The residuesphere was shown to be a hot-spot for conjugal gene transfer compared to conjugation in sterile sand and non-sterile bulk soil. Impact of fungal colonisation of the residuesphere on bacterial colonisation and conjugation was also investigated. The inhibition of fungal colonisation, due to the application of an eukaryotic inhibitor, increased bacterial colonisation of the residuesphere in soil microcosms compared to non-treated leaves. This treatment also had a transient, positive effect on conjugation. Bacterial conjugation in the residuesphere of leaves subjected to 17 days of fungal colonisation was significantly lower than in the residuesphere of non-colonised leaves. Fungal biomass, as measured by chitinase activity, was inversely related to the conjugation efficiency. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Soil microcosm; Gene transfer; Microbial colonisation ; Litter; Pseudomonas putida
1. Introduction Gene transfer between bacteria has been shown to be of potential importance for adaptation to changes in the environment. An example is the acquisition of new degradation pathways in aquatic bacteria [1]. Several studies have shown that gene transfer by plasmid conjugation occurs in terrestrial environments [2,3], strongly suggesting that conjugation can contribute to interbacterial gene £ow in natural ecosystems. Conjugal gene transfer, however, does not occur with the same e¤ciency in all microhabitats. Conjugation seems especially stimulated at interfaces in the environment, where high bacterial densities are often found [4]. Several hot-spots for conjugal gene transfer in the terrestrial environment have been identi¢ed, including the rhizosphere [5,6], the phylloplane [7] and the spermosphere [8]. The residuesphere is de¢ned as the interface between decaying plant material and soil and conjugal gene transfer has yet to be studied in this important micro-
* Corresponding author. Tel. : +45 (35) 32 20 53; Fax: +45 (35) 32 20 40; E-mail : [email protected]
habitat. Enhanced gene transfer in the residuesphere might lead to signi¢cant increases in the spread of antibiotic resistance genes from sewage and animal husbandry manure used as fertiliser on crops. In this study, we have investigated conjugal gene transfer in the residuesphere by using a series of microcosm experiments containing leaves of barley straw in sand or soil. We have also investigated the impact of fungal colonisation on conjugation in the residuesphere. Bacterial degradation of lignin is far less e¤cient than that performed by fungi, but many bacteria readily utilise catabolic intermediates generated in the degradation of lignin and might therefore be stimulated by fungal activity [9]. Furthermore, conjugation of the TOL plasmid between Pseudomonas strains has been shown to be depending on the donors' speci¢c growth rate [10]. On the other hand, fungi can also be antagonistic towards bacteria, either by production of antibiotics or via competition for space on the colonised material. Thus, fungal-bacterial interactions would be hypothesised to either (1) increase the number and the activity of the bacteria and thereby possibly the e¤ciency of conjugal gene transfer or (2) decrease conjugal gene transfer due to antagonistic e¡ects of the fungi on the bacteria. The aim of the present study was to determine if the
0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 6 4 9 6 ( 9 9 ) 0 0 0 7 9 - 3
FEMSEC 1085 27-12-99
40
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
residuesphere is a hot-spot for conjugal gene transfer and if fungal colonisation of the leaves has an impact on the conjugation e¤ciency. To the best of our knowledge, this is the ¢rst study investigating gene transfer in the residuesphere and the ¢rst report addressing fungal-bacterial interactions in relation to gene transfer. 2. Materials and methods
2.3. Natamycin-treated leaves The fungal inhibition treatment consisted of incubating barley leaves in a fungicide (natamycin 1 mg ml31 in methanol) for 24 h at room temperature prior to set-up of the microcosms. Control leaves, which were to be colonised by fungi in the microcosms, were incubated in phosphatebu¡ered saline (PBS) bu¡er. 2.4. Fungal pre-colonisation of barley leaves
2.1. Bacterial strains Transfer experiments were carried out between derivatives of Pseudomonas putida KT2440 [11]. The donor strain KT2442LacIq [7] harbours a derivative of the TOL plasmid (TOL,gfp,Km), which confers resistance to kanamycin (Km) and carries the gfpmut3 gene (encoding the green £uorescent protein) regulated by the synthetic lac promoter PA1ÿ04=03 [7]. The recipient strain is a tetracycline resistant (Tc) derivative of KT2440 (KT2440Tc ). Liquid cultures were grown in Luria Bertani (LB) broth and colony forming unit (CFU) counts were counted on LB agar plates [12]. For speci¢c growth and selection of KT2442LacIq/TOL,gfp,Km (donor) and KT2440Tc (recipient), the medium was amended with 50 Wg ml31 Km and 15 Wg ml31 Tc, respectively. Transconjugants were selected on LB agar amended with both 50 Wg ml31 Km and 15 Wg ml31 Tc. 2.2. Microcosms Microcosms consisted of 16 g soil or 15 g baked sand in small plastic cups (height 41 mm, upper diameter 37 mm). The sand was quartz sand, 0.3^0.6 mm. Sand was washed ¢ve times in sterile water and baked (500³C, 8 h) to ensure sterility and remove carbon. The soil was a sandy loam obtained from Ta®strup near Copenhagen, Denmark [13]. Soil was dried at room temperature and sieved (2-mm grid). For all experiments, donors and recipients were depleted of intracellular energy reserves by incubation in sterile 0.9% NaCl for 48 h at 30³C prior to set-up of the microcosms. Aliquots of 1.5 ml donor cells and 1.5 ml recipient cells in 0.9% NaCl were added to the microcosms, creating moisture conditions corresponding to 78% of the water holding capacity of the soil. The total inoculum was 1^6U106 donors and 1^6U106 recipients per g sand or soil. Some microcosms were inoculated with 1.5 ml donors plus 1.5 ml 0.9% NaCl or 1.5 ml recipients plus 1.5 ml 0.9% NaCl. These microcosms were used for plate mating control (see Section 2.5). Microcosms were incubated in closed plastic containers at 25³C and adjusted for water loss during the experiment. Three or four pieces of leaves of barley straw, each 1.5 cm long, were inserted vertically, such that they were separated from each other, for each of the residuesphere microcosms.
To obtain leaves colonised with soil fungi and accompanying non-colonised control leaves, sterilised leaves were moistened in sterile PBS for 1 h and hereafter incubated on top of the surface of 20 g non-sterile soil or sterile sand in a small plastic cup covered with para¢lm. The leaves were incubated 5 or 17 days at room temperature before three leaves were transferred to microcosms consisting of 15 g sterile sand plus donor and recipient cells. The microcosms were incubated for 48 h as described above. A microcosm with leaves inoculated for 1 h in sterile PBS was also prepared. At sampling, the leaves were cut into halves, with one half used for CFU enumeration and the other half for chitinase activity measurement. 2.5. Sampling of microcosms At each sampling day, three replicate microcosms were sampled. For extraction of bacteria, all the soil or sand from the microcosms was transferred to 20 ml of 0.9% NaCl in 50-ml bottles and shaken either for 2 min on a multi wrist-shaker at speed 4 (Lab-Line Instruments) or for 10 min at 300 rpm on a horizontal shaker (KS 250 basic, IKA Labortechnik). Bacteria present in the residuesphere were extracted from leaves in 1.8 ml 0.9% NaCl in 2-ml microcentrifuge tubes by vortexing vigorously for 1 min. Samples were then either serial diluted and plated on selective plates for CFU counts of donors, recipients and transconjugants or used in enzyme assays for the detection of fungal activity (see below). Plate mating was estimated by mixing and spread plating aliquots of separately incubated donors and recipients on transconjugant selective plates. 2.6. Measurement of metabolic activity Incorporation of [3 H]leucine was used to measure bacterial metabolic activity [23]. Sand microcosms were amended with 1 ml of a mixture of unlabelled leucine and tritiated leucine (Amersham, 158 Ci mmol31 ) plus 2 ml sterile distilled H2 O, giving a ¢nal concentration of 2000 nM leucine and 1 WCi [3 H]leucine per sand microcosm. Measurements of bacterial activity in the residuesphere were performed by carefully removing the leaves from the microcosms, mixing 1 ml of the leucine solution
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
plus 2 ml sterile distilled H2 O in the sand and placing the leaves on top of the water-saturated microcosm. Controls, inactivated with formaldehyde (¢nal concentration 2%) prior to addition of the leucine mixture, were included in all experiments to determine unspeci¢c binding of [3 H]leucine. After 15 min, the leaves were transferred to 1.8 ml 0.9% NaCl and the sand microcosms to 20 ml 0.9% NaCl. Bacteria were extracted by shaking on a multi wristshaker as described above. After removal of 100 Wl for CFU determination, samples were amended with 37% formaldehyde (¢nal concentration 2%) to prevent further incorporation of tritiated leucine. A 7.5-ml extract from sand microcosms and 1.8 ml from barley leaf extracts, respectively, were ¢ltered through Whatman GF/F ¢lters which were pre-rinsed in 1% K2 HPO4 to reduce unspeci¢c binding of leucine [23]. After application of the sample, 10 ml of sterile Milli-Q water was ¢ltered. The ¢lters were incubated overnight in the dark in 5 ml scintillation £uid (Ready Safe) and then counted in a scintillation counter. Previous experiments have shown a linear bacterial assimilation of leucine with time, using these conditions [15]. 2.7. Enzyme assays Fungal biomass was estimated by determination of fungal enzyme activities for endo- and exo-cellulase (cellulase) or L-N-acetylglucosaminidase (chitinase) using £uorogenic substrates [16]. Assays were performed for 2 h (cellulase) or 3 h (chitinase).
41
measured (Table 1). The metabolic activity of bacteria declined to day 1, especially on the leaves, and remained then on a steady level throughout the experiment. The metabolic activity of bacteria extracted from bulk sand was in general higher than bacteria extracted from the residuesphere. 3.2. Conjugation in quartz sand and in the residuesphere The experiment performed to assess conjugation in the residuesphere versus in bulk sand involved ¢ve sampling dates 0, 1, 2, 3 and 7 days after set-up of the microcosms. Conjugation e¤ciency is expressed as numbers of transconjugants per geometric mean of donors and recipients ( = transconjugants/(donorsUrecipients)1=2 ) to achieve transfer e¤ciencies that include both donor and recipient densities [17]. At sampling days 1, 2, 3 and 7, the conjugation e¤ciency was signi¢cantly higher (P 6 0.02) in the residuesphere than in bulk sand (Fig. 1), indicating that the residuesphere can be a hot-spot for conjugal gene transfer in soil environments. No transfer was detected on day 0 and no plate mating was found. 3.3. E¡ect of soil fungi on the colonisation of leaves by donors and recipients
3.1. Metabolic activity of donors and recipients in quartz sand and in the residuesphere
To test the in£uence of fungal colonisation on bacterial colonisation of the residuesphere, fungicide-treated and -untreated leaves from soil microcosms were compared with respect to bacterial colonisation (Fig. 2). The number of donors and recipients stabilised after 1^2 days in both treatments, although a slight dip was observed at day 9 in some cases. Comparing numbers of donor versus recipient cells after day 1 of the experiment, recipient cell numbers were greater within a given treatment. In comparison of the fungicide-treated and untreated leaves after cell number stabilisation, both donor and recipient cell colonisation were greater for the fungicide-treated leaves, suggesting an antagonistic e¡ect of soil fungi on bacterial colonisation of leaves (Fig. 2).
Metabolic activity of donor and recipient bacteria extracted from quartz sand and from the residuesphere was
3.4. Conjugal gene transfer in soil and in the residuesphere and impact of fungi on conjugation
2.8. Statistical treatment of data All experiments were performed in triplicate. Student's t-test was performed using log10 -transformed data. 3. Results
Table 1 Metabolic activity of donor and recipient bacteria extracted from bulk sand and from the residuesphere of barley leaves Day
Activity of bacteria in bulk sanda (fmol leucine CFU31 h31 )
Activity of cells in the residuespherea (fmol leucine CFU31 h31 )
0 1 3 7
4.79U1031 þ 1.74U1031 1.30U1031 þ 1.26U1032 2.66U1031 þ 3.04U1032 2.58U1031 þ 3.72U1032
7.03U1031 þ 3.47U1031 6.75U1032 þ 2.80U1032 4.02U1032 þ 3.20U1032 2.94U1032 þ 7.32U1033
Transfer e¤ciencies are shown in Fig. 1. a Average þ S.D.s of three replicates.
Using the experimental set-up described above, conjugation in bulk soil was compared to conjugation in the residuesphere of leaves. In addition, conjugation on leaves with inhibited fungal colonisation was compared to conjugation on colonised leaves. Transconjugants appeared on day 0 in soil microcosms. This was most probably due to formation of transconjugants in the microcosms in the period between inoculation and sampling. The following sampling dates, and for both treatments, conjugal gene transfer was signi¢cantly higher (P 6 0.02) in the residuesphere than in bulk soil (Fig. 3). No plate mating was found.
FEMSEC 1085 27-12-99
42
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
Fig. 2. In£uence of fungal colonisation on bacterial colonisation of the residuesphere. Error bars shows S.D.s of three replicates.
Average densities of donors plus recipients per g of barley leaves were between 9 (day 1) and 198 (day 13) times higher than the average number of donors plus recipients per g soil (calculated from data to Fig. 3). On sampling days 1, 2 and 5, conjugal gene transfer was signi¢cantly higher (P 6 0.01) in the residuesphere of leaves with inhibited fungal populations compared to conjugal gene transfer in the residuesphere of leaves with intact fungal populations. The e¡ect of treatment of the leaves with a fungicide was transient and the conjugation e¤ciencies were equalised by day 9. No signi¢cant inverse correlation between cellulase activity and treatment of leaves with the fungicide was observed (data not shown), but this was most probably due to a limited number of
fungal species expressing cellulase activity (see Section 4). Bacterial colonisation was not a¡ected by the fungicide treatment of the leaves in controls without fungi. This was tested by comparing the numbers of cells extracted from natamycin-treated leaves and PBS-treated leaves incubated in sterile sand microcosms with donor and recipient cells (data not shown). 3.5. E¡ect of pre-colonisation by soil fungi on the appearance of transconjugants in the residuesphere In another approach to compare conjugation in the residuesphere of fungi-colonised leaves and non-colonised leaves, some leaves were pre-colonised with soil fungi for
Fig. 1. Transfer e¤ciency in bulk sand and in the residuesphere of barley leaves. Points represents replicate microcosms. Symbols on baseline show transfer e¤ciencies of 1U1037 or less. Metabolic activities are shown in Table 1.
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
43
Fig. 3. Transfer e¤ciency in bulk soil, in the residuesphere and in the residuesphere of fungicide-treated leaves. Points represent replicate microcosms. Symbols on baseline show transfer e¤ciencies of 1U1036 or less.
5 or 17 days before transfer to sterile sand for the subsequent conjugation experiment. Non-colonised leaves were also tested in sterile sand to serve as a control. In this way, conjugation on colonised leaves versus completely noncolonised leaves could be evaluated. Also, conjugation on leaves treated brie£y in PBS was determined in order to be able to evaluate the in£uence of senescence of the leaves on conjugation. In theory, the di¡erent incubation
periods could alter the leaves and thereby possibly the bacterial colonisation and/or conjugation e¤ciency. There were no signi¢cant di¡erences in the numbers of transconjugants in the residuesphere of non-colonised leaves as compared to the residuesphere of leaves exposed to 5 days of fungal pre-colonisation. Numbers of transconjugants were however signi¢cantly lower for leaves exposed to 17 days of fungal pre-colonisation as compared
Fig. 4. Number of transconjugants in the residuesphere of barley straw leaves. The leaves were pre-colonised by soil fungi or incubated on sterile sand and then incubated for 48 h in sand microcosms with addition of donor and recipients before enumeration of transconjugants. Relative chitinase activities of pre-colonised and non-colonised leaves are also shown.
FEMSEC 1085 27-12-99
44
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
to non-colonised leaves (P 6 0.05) (Fig. 4). No plate mating was found. Conjugation e¤ciencies (transconjugants/(donorsUrecipients)1=2 ) from triplicate samples were approximately the same from 0 and 5 days: 2.10U1033 þ 8.85U1034 for 0 day leaves, 1.31U1033 þ 5.52U1034 and 2.53U 1033 þ 1.73U1033 for non-colonised and colonised 5 days leaves, respectively. On leaves exposed to 17 days of fungal pre-colonisation, however, transconjugants/ (donorsUrecipients)1=2 was signi¢cantly lower in the residuesphere of colonised leaves than in the residuesphere of non-colonised leaves (P 6 0.05). Numbers of transconjugants/(donorsUrecipients)1=2 were 1.46U1033 þ 33 35 1.38U10 and 5.75U10 þ 5.10U1035 on non-colonised and fungi-colonised leaves, respectively, on leaves exposed to 17 days of fungal pre-colonisation. Comparing numbers of transconjugants on non-colonised leaves, a minor e¡ect of senescence of the leaves was observed (Fig. 4). However, conjugation e¤ciencies were not a¡ected. In this experiment, chitinase activity was used as a measurement of fungal biomass. Chitinase activity increased with increasing numbers of incubation days, thus the appearance of transconjugants in the residuesphere was inversely related to the fungal biomass. 4. Discussion Increased conjugation ratios for a TOL plasmid between introduced donor and recipient strains were demonstrated in both soil microcosm and control sand experiments. These observations are in accordance with increased gene transfer reported in other microhabitats. Van Elsas et al. [5] found transfer of the plasmid RP4 between introduced donor and recipient strains in wheat rhizosphere soil but not in non-rhizosphere soil, even when organic matter was added to the bulk soil. Also the phylloplane of bean has been suggested as a hot-spot for conjugation compared to conjugation on polycarbonate ¢lters [7]. Conjugation has previously also been studied in the spermosphere and rhizosphere of barley and higher conjugation rates in both microhabitats were detected in comparison with that found in unamended bulk soils [8]. The observed transfer rates in the residuesphere of barley leaves colonised by fungi in the soil microcosms ranged from 8.96U1035 to 7.94U1034 transconjugants/(donors recipients)1=2 and were comparable to transfer rates previously observed in the rhizosphere. An underestimation of the numbers of transconjugants in the present study cannot be ruled out, since impeded expression of a Tc resistance determinant in conjugation experiments has been reported [18]. SÖrensen and Jensen [8] reported 1.59U1034 ^ 2.51U1034 transconjugants/(donorsUrecipients)1=2 in barley rhizosphere and van Elsas et al. [5] found between 6.6U1037 and 3.8U1034 transconjugants per donor in
wheat rhizosphere. However, the transfer e¤ciencies in the residuesphere were lower than reported from the phylloplane, where up to 2.6U1032 transconjugants per donor and 3.4U1031 transconjugants per recipient were found [7]. Conjugation requires cell to cell contact, which may o¡er an explanation for the increased gene transfer rates observed in the residuesphere and other microbial hotspots. Average densities of donors plus recipients per g barley leaf in the present study were higher than the average number of donors plus recipients per g soil. The activity of bacteria extracted from the residuesphere of barley leaves incubated in sand microcosms was not higher than found for bacteria extracted from bulk sand. This is most probably due to the inability of the Pseudomonas strains to degrade the barley leaves, since the residuesphere, being a microhabitat with a high carbon turnover, normally would be an obvious environment for highly active bacteria. However, since conjugation was increased in the residuesphere of barley leaves compared to bulk sand, this enhanced gene transfer was not due to a higher metabolic activity of the bacteria, rather due to the higher donor and recipient densities on the surface of the leaves. Similarly, Normander et al. [7] found that metabolic activity was not rate limiting for conjugation, with no correlation between these two parameters being detected. Increased bacterial colonisation was observed on leaves pre-treated with a fungicide in this study, suggesting an antagonistic interaction between the indigenous soil fungal populations and the Pseudomonas donor and recipient strains. Antagonistic interactions between bacteria and fungi have been reported in other environments similar to the residuesphere. MÖller et al. [19] found a reduction in C mineralisation rate when both fungi and bacteria were present as compared to only fungal colonisation on decomposing beech leaves. Another example of bacterial antagonism has been demonstrated in dune soils by the inhibition of fungal growth due to bacterial activity [20]. Conversely, antagonistic behaviour of fungi against bacteria on decaying wood, an environment comparable to the residuesphere, has been detected by scanning electron microscopy [21]. The nature of the bacterial inhibition observed in this study is not yet known, but some possible mechanisms include competition for space, production of antibiotics or a combination of these factors. Conjugation was also increased on fungicide-treated leaves. This could be due to the higher bacterial numbers and therefore higher cell density for donor and recipient cells in the residuesphere of leaves not colonised by fungi. However, it should be taken into consideration that the fungicide also reduces the numbers of protozoa, which have been shown to have an e¡ect on the bacterial survival [22]. The fungal biomass was measured as cellulase or chitinase activity [16]. Our approach to use cellulase activity as an indicator of fungal presence did not give a clear result.
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
Cellulase activity showed no signi¢cant correlation with the observed decrease in bacterial colonisation and conjugation on leaves with intact fungal populations, as compared to leaves with inhibited fungal populations. However, this could be ascribed to the limited number of fungal species expressing cellulase activity [16]. In subsequent experiments, we performed measurements on chitinase activity as an indicator for fungal presence. This enzyme has been shown correlated with two independent indicators of fungal biomass [16]. Using chitinase as an indicator of fungal biomass, a strong positive correlation between time of exposure to fungi of the leaves and chitinase activity was observed. A negative correlation between chitinase activity and numbers of transconjugants as well as conjugation e¤ciency in the residuesphere was also observed. In conclusion, we have shown that (i) the residuesphere of leaves of barley straw is a hot-spot for gene transfer, (ii) the increased conjugation in the residuesphere was not caused by an elevated metabolic activity of the bacteria, (iii) fungal colonisation can inhibit bacterial colonisation of the residuesphere and (iv) fungal colonisation can inhibit bacterial conjugation in the residuesphere. Acknowledgements
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[15]
We wish to thank Niels Kroer for helpful suggestions and discussions and Morten Miller for help in the determination of fungal activity. This work was partially supported by the Danish Environmental Protection Agency (Gitte SengelÖv and SÖren J. SÖrensen).
[16]
[17]
[18]
References [1] Barkay, T., Liebert, C. and Gillman, M. (1993) Conjugal gene transfer to aquatic bacteria detected by the generation of a new phenotype. Appl. Environ. Microbiol. 59, 807^816. [2] Cresswell, N. and Wellington, E.M.H. (1992) Detection of genetic exchange in the terrestrial environment. In: Genetic Interactions among Microorganisms in the Natural Environment (Wellington, E.M.H. and van Elsas, J.D., Eds.), pp. 59^82. Pergamon Press, Oxford. [3] Hill, K.E. and Top, E.M. (1998) Gene transfer in soil systems using microcosms. FEMS Microbiol. Ecol. 25, 219^329. [4] SÖrensen, S.J., Kroer, N., SÖrensen, E., SengelÖv, G. and Barkay, T. (1996) Conjugation in aquatic environments. In: Molecular Microbial Ecology Manual (Akkermans, A.D.L., van Elsas, J.D. and de Bruijn, F.L., Eds.), 5.2.1., pp. 1^29. Kluwer Academic Publishers, Dordrecht. [5] van Elsas, J.D., Trevors, J.T. and Starodub, M.E. (1988) Bacterial
[19]
[20]
[21]
[22]
[23]
45
conjugation between pseudomonads in the rhizosphere of wheat. FEMS Microbiol. Ecol. 53, 299^306. Lilley, A.K., Fry, J.C., Day, M.J. and Bailey, M.J. (1994) In situ transfer of an exogenously isolated plasmid between Pseudomonas spp. in sugar beet rhizosphere. Microbiology 140, 27^33. Normander, B., Christensen, B.B., Molin, S. and Kroer, N. (1998) E¡ect of bacterial distribution and activity on conjugal gene transfer on the phylloplane of the Bush Bean (Phaseolus vulgaris). Appl. Environ. Microbiol. 64, 1902^1909. SÖrensen, S.J. and Jensen, L.E. (1998) Transfer of plasmid RP4 in the spermosphere and rhizosphere of barley seedling. Antonie Leeuwenhoek 73, 69^77. Ruttigman, C., Vicun¬a, R., Mozuch, M.D. and Kirk, T.K. (1991) Limited bacterial mineralization of fungal degradation intermediates from synthetic lignin. Appl. Environ. Microbiol. 57, 3652^3655. Smets, B.F., Rittmann, B.E. and Stahl, D.A. (1993) The speci¢c growth rate of Pseudomonas putida PAW1 in£uences the conjugal transfer rate of the TOL plasmid. Appl. Environ. Microbiol. 59, 3430^3437. Franklin, F.C.H., Bagdasarian, M., Bagdasarian, M.M. and Timmis, K.N. (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. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning - a Laboratory Manual. Cold Spring Harbor Laboratory Press, New York. Jensen, L.E., Kragelund, L. and Nybroe, O. (1998) Expression of a nitrogen regulated lux gene fusion in Pseudomonas £uorescens DF57 studied in pure culture and in soil. FEMS Microbiol. Ecol. 25, 23^ 32. Kroer, N., Barkay, T., SÖrensen, S. and Weber, D. (1998) E¡ect of root exudates and bacterial metabolic activity on conjugal gene transfer in the rhizosphere of a marsh plant. FEMS Microbiol. Ecol. 25, 375^384. Miller, M., Paloja«rvi, A., Rangger, A., Reeslev, M. and KjÖller, A. (1998) The use of £uorogenic substrates to measure fungal presence and activity in soil. Appl. Environ. Microbiol. 64, 613^617. Smit, E. and van Elsas, J.D. (1990) Determination of plasmid transfer in soil: Consequences of bacterial mating on selective agar media. Curr. Microbiol. 21, 151^157. O'Morchoe, S.B., Ogunseitan, O., Sayler, G.S. and Miller, R.V. (1988) Conjugal transfer of R68.45 and FP5 between Pseudomonas aeruginosa strains in a freshwater environment. Appl. Environ. Microbiol. 54, 1923^1929. MÖller, J., Miller, M. and KjÖller, A. (1999) Fungal-bacterial interaction on beech leaves : In£uence on decomposition and dissolved organic carbon quality. Soil Biol. Biochem. 31, 367^374. De Boer, W., Klein Gunnewiek, P.J.A., Lafeber, P., Janse, J.D., Spit, B.E. and Woldendorp, J.W. (1998) Anti-fungal properties of chitinolytic dune soil bacteria. Soil Biol. Biochem. 30, 193^203. Tsuneda, A. and Thorn, R.G. (1995) Interactions of wood decay fungi with other microorganisms, with emphasis on the degradation of cell walls. Can. J. Bot. 73, 1325^1333. SÖrensen, S.J., Schyberg, T. and RÖnn, R. (1999) Predation by protozoa on Escherichia coli K12 in soil and transfer of resistance plasmid RP4 to indigenous bacteria in soil. Appl. Soil Ecol. 11, 79^90. Ba®a®th, E. (1994) Measurement of protein synthesis by soil bacterial assemblages with the leucine incorporation technique. Biol. Fertil. Soils 17, 147^153.
FEMSEC 1085 27-12-99
www.fems-microbiology.org
In£uence of fungal-bacterial interactions on bacterial conjugation in the residuesphere Gitte SengelÖv a , George A. Kowalchuk b , SÖren J. SÖrensen b
a;
*
a Department of General Microbiology, University of Copenhagen, SÖlvgade 83 H, DK-1307 Copenhagen, Denmark Department of Plant-Microorganism Interactions, Centre for Terrestrial Ecology, Netherlands Institute of Ecology, 6666 ZG Heteren, The Netherlands
Received 25 June 1999; received in revised form 4 October 1999; accepted 7 October 1999
Abstract Conjugal gene transfer among bacteria in the residuesphere (area between decaying plant material and soil) of leaves of barley straw was studied. The residuesphere was shown to be a hot-spot for conjugal gene transfer compared to conjugation in sterile sand and non-sterile bulk soil. Impact of fungal colonisation of the residuesphere on bacterial colonisation and conjugation was also investigated. The inhibition of fungal colonisation, due to the application of an eukaryotic inhibitor, increased bacterial colonisation of the residuesphere in soil microcosms compared to non-treated leaves. This treatment also had a transient, positive effect on conjugation. Bacterial conjugation in the residuesphere of leaves subjected to 17 days of fungal colonisation was significantly lower than in the residuesphere of non-colonised leaves. Fungal biomass, as measured by chitinase activity, was inversely related to the conjugation efficiency. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Soil microcosm; Gene transfer; Microbial colonisation ; Litter; Pseudomonas putida
1. Introduction Gene transfer between bacteria has been shown to be of potential importance for adaptation to changes in the environment. An example is the acquisition of new degradation pathways in aquatic bacteria [1]. Several studies have shown that gene transfer by plasmid conjugation occurs in terrestrial environments [2,3], strongly suggesting that conjugation can contribute to interbacterial gene £ow in natural ecosystems. Conjugal gene transfer, however, does not occur with the same e¤ciency in all microhabitats. Conjugation seems especially stimulated at interfaces in the environment, where high bacterial densities are often found [4]. Several hot-spots for conjugal gene transfer in the terrestrial environment have been identi¢ed, including the rhizosphere [5,6], the phylloplane [7] and the spermosphere [8]. The residuesphere is de¢ned as the interface between decaying plant material and soil and conjugal gene transfer has yet to be studied in this important micro-
* Corresponding author. Tel. : +45 (35) 32 20 53; Fax: +45 (35) 32 20 40; E-mail : [email protected]
habitat. Enhanced gene transfer in the residuesphere might lead to signi¢cant increases in the spread of antibiotic resistance genes from sewage and animal husbandry manure used as fertiliser on crops. In this study, we have investigated conjugal gene transfer in the residuesphere by using a series of microcosm experiments containing leaves of barley straw in sand or soil. We have also investigated the impact of fungal colonisation on conjugation in the residuesphere. Bacterial degradation of lignin is far less e¤cient than that performed by fungi, but many bacteria readily utilise catabolic intermediates generated in the degradation of lignin and might therefore be stimulated by fungal activity [9]. Furthermore, conjugation of the TOL plasmid between Pseudomonas strains has been shown to be depending on the donors' speci¢c growth rate [10]. On the other hand, fungi can also be antagonistic towards bacteria, either by production of antibiotics or via competition for space on the colonised material. Thus, fungal-bacterial interactions would be hypothesised to either (1) increase the number and the activity of the bacteria and thereby possibly the e¤ciency of conjugal gene transfer or (2) decrease conjugal gene transfer due to antagonistic e¡ects of the fungi on the bacteria. The aim of the present study was to determine if the
0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 6 4 9 6 ( 9 9 ) 0 0 0 7 9 - 3
FEMSEC 1085 27-12-99
40
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
residuesphere is a hot-spot for conjugal gene transfer and if fungal colonisation of the leaves has an impact on the conjugation e¤ciency. To the best of our knowledge, this is the ¢rst study investigating gene transfer in the residuesphere and the ¢rst report addressing fungal-bacterial interactions in relation to gene transfer. 2. Materials and methods
2.3. Natamycin-treated leaves The fungal inhibition treatment consisted of incubating barley leaves in a fungicide (natamycin 1 mg ml31 in methanol) for 24 h at room temperature prior to set-up of the microcosms. Control leaves, which were to be colonised by fungi in the microcosms, were incubated in phosphatebu¡ered saline (PBS) bu¡er. 2.4. Fungal pre-colonisation of barley leaves
2.1. Bacterial strains Transfer experiments were carried out between derivatives of Pseudomonas putida KT2440 [11]. The donor strain KT2442LacIq [7] harbours a derivative of the TOL plasmid (TOL,gfp,Km), which confers resistance to kanamycin (Km) and carries the gfpmut3 gene (encoding the green £uorescent protein) regulated by the synthetic lac promoter PA1ÿ04=03 [7]. The recipient strain is a tetracycline resistant (Tc) derivative of KT2440 (KT2440Tc ). Liquid cultures were grown in Luria Bertani (LB) broth and colony forming unit (CFU) counts were counted on LB agar plates [12]. For speci¢c growth and selection of KT2442LacIq/TOL,gfp,Km (donor) and KT2440Tc (recipient), the medium was amended with 50 Wg ml31 Km and 15 Wg ml31 Tc, respectively. Transconjugants were selected on LB agar amended with both 50 Wg ml31 Km and 15 Wg ml31 Tc. 2.2. Microcosms Microcosms consisted of 16 g soil or 15 g baked sand in small plastic cups (height 41 mm, upper diameter 37 mm). The sand was quartz sand, 0.3^0.6 mm. Sand was washed ¢ve times in sterile water and baked (500³C, 8 h) to ensure sterility and remove carbon. The soil was a sandy loam obtained from Ta®strup near Copenhagen, Denmark [13]. Soil was dried at room temperature and sieved (2-mm grid). For all experiments, donors and recipients were depleted of intracellular energy reserves by incubation in sterile 0.9% NaCl for 48 h at 30³C prior to set-up of the microcosms. Aliquots of 1.5 ml donor cells and 1.5 ml recipient cells in 0.9% NaCl were added to the microcosms, creating moisture conditions corresponding to 78% of the water holding capacity of the soil. The total inoculum was 1^6U106 donors and 1^6U106 recipients per g sand or soil. Some microcosms were inoculated with 1.5 ml donors plus 1.5 ml 0.9% NaCl or 1.5 ml recipients plus 1.5 ml 0.9% NaCl. These microcosms were used for plate mating control (see Section 2.5). Microcosms were incubated in closed plastic containers at 25³C and adjusted for water loss during the experiment. Three or four pieces of leaves of barley straw, each 1.5 cm long, were inserted vertically, such that they were separated from each other, for each of the residuesphere microcosms.
To obtain leaves colonised with soil fungi and accompanying non-colonised control leaves, sterilised leaves were moistened in sterile PBS for 1 h and hereafter incubated on top of the surface of 20 g non-sterile soil or sterile sand in a small plastic cup covered with para¢lm. The leaves were incubated 5 or 17 days at room temperature before three leaves were transferred to microcosms consisting of 15 g sterile sand plus donor and recipient cells. The microcosms were incubated for 48 h as described above. A microcosm with leaves inoculated for 1 h in sterile PBS was also prepared. At sampling, the leaves were cut into halves, with one half used for CFU enumeration and the other half for chitinase activity measurement. 2.5. Sampling of microcosms At each sampling day, three replicate microcosms were sampled. For extraction of bacteria, all the soil or sand from the microcosms was transferred to 20 ml of 0.9% NaCl in 50-ml bottles and shaken either for 2 min on a multi wrist-shaker at speed 4 (Lab-Line Instruments) or for 10 min at 300 rpm on a horizontal shaker (KS 250 basic, IKA Labortechnik). Bacteria present in the residuesphere were extracted from leaves in 1.8 ml 0.9% NaCl in 2-ml microcentrifuge tubes by vortexing vigorously for 1 min. Samples were then either serial diluted and plated on selective plates for CFU counts of donors, recipients and transconjugants or used in enzyme assays for the detection of fungal activity (see below). Plate mating was estimated by mixing and spread plating aliquots of separately incubated donors and recipients on transconjugant selective plates. 2.6. Measurement of metabolic activity Incorporation of [3 H]leucine was used to measure bacterial metabolic activity [23]. Sand microcosms were amended with 1 ml of a mixture of unlabelled leucine and tritiated leucine (Amersham, 158 Ci mmol31 ) plus 2 ml sterile distilled H2 O, giving a ¢nal concentration of 2000 nM leucine and 1 WCi [3 H]leucine per sand microcosm. Measurements of bacterial activity in the residuesphere were performed by carefully removing the leaves from the microcosms, mixing 1 ml of the leucine solution
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
plus 2 ml sterile distilled H2 O in the sand and placing the leaves on top of the water-saturated microcosm. Controls, inactivated with formaldehyde (¢nal concentration 2%) prior to addition of the leucine mixture, were included in all experiments to determine unspeci¢c binding of [3 H]leucine. After 15 min, the leaves were transferred to 1.8 ml 0.9% NaCl and the sand microcosms to 20 ml 0.9% NaCl. Bacteria were extracted by shaking on a multi wristshaker as described above. After removal of 100 Wl for CFU determination, samples were amended with 37% formaldehyde (¢nal concentration 2%) to prevent further incorporation of tritiated leucine. A 7.5-ml extract from sand microcosms and 1.8 ml from barley leaf extracts, respectively, were ¢ltered through Whatman GF/F ¢lters which were pre-rinsed in 1% K2 HPO4 to reduce unspeci¢c binding of leucine [23]. After application of the sample, 10 ml of sterile Milli-Q water was ¢ltered. The ¢lters were incubated overnight in the dark in 5 ml scintillation £uid (Ready Safe) and then counted in a scintillation counter. Previous experiments have shown a linear bacterial assimilation of leucine with time, using these conditions [15]. 2.7. Enzyme assays Fungal biomass was estimated by determination of fungal enzyme activities for endo- and exo-cellulase (cellulase) or L-N-acetylglucosaminidase (chitinase) using £uorogenic substrates [16]. Assays were performed for 2 h (cellulase) or 3 h (chitinase).
41
measured (Table 1). The metabolic activity of bacteria declined to day 1, especially on the leaves, and remained then on a steady level throughout the experiment. The metabolic activity of bacteria extracted from bulk sand was in general higher than bacteria extracted from the residuesphere. 3.2. Conjugation in quartz sand and in the residuesphere The experiment performed to assess conjugation in the residuesphere versus in bulk sand involved ¢ve sampling dates 0, 1, 2, 3 and 7 days after set-up of the microcosms. Conjugation e¤ciency is expressed as numbers of transconjugants per geometric mean of donors and recipients ( = transconjugants/(donorsUrecipients)1=2 ) to achieve transfer e¤ciencies that include both donor and recipient densities [17]. At sampling days 1, 2, 3 and 7, the conjugation e¤ciency was signi¢cantly higher (P 6 0.02) in the residuesphere than in bulk sand (Fig. 1), indicating that the residuesphere can be a hot-spot for conjugal gene transfer in soil environments. No transfer was detected on day 0 and no plate mating was found. 3.3. E¡ect of soil fungi on the colonisation of leaves by donors and recipients
3.1. Metabolic activity of donors and recipients in quartz sand and in the residuesphere
To test the in£uence of fungal colonisation on bacterial colonisation of the residuesphere, fungicide-treated and -untreated leaves from soil microcosms were compared with respect to bacterial colonisation (Fig. 2). The number of donors and recipients stabilised after 1^2 days in both treatments, although a slight dip was observed at day 9 in some cases. Comparing numbers of donor versus recipient cells after day 1 of the experiment, recipient cell numbers were greater within a given treatment. In comparison of the fungicide-treated and untreated leaves after cell number stabilisation, both donor and recipient cell colonisation were greater for the fungicide-treated leaves, suggesting an antagonistic e¡ect of soil fungi on bacterial colonisation of leaves (Fig. 2).
Metabolic activity of donor and recipient bacteria extracted from quartz sand and from the residuesphere was
3.4. Conjugal gene transfer in soil and in the residuesphere and impact of fungi on conjugation
2.8. Statistical treatment of data All experiments were performed in triplicate. Student's t-test was performed using log10 -transformed data. 3. Results
Table 1 Metabolic activity of donor and recipient bacteria extracted from bulk sand and from the residuesphere of barley leaves Day
Activity of bacteria in bulk sanda (fmol leucine CFU31 h31 )
Activity of cells in the residuespherea (fmol leucine CFU31 h31 )
0 1 3 7
4.79U1031 þ 1.74U1031 1.30U1031 þ 1.26U1032 2.66U1031 þ 3.04U1032 2.58U1031 þ 3.72U1032
7.03U1031 þ 3.47U1031 6.75U1032 þ 2.80U1032 4.02U1032 þ 3.20U1032 2.94U1032 þ 7.32U1033
Transfer e¤ciencies are shown in Fig. 1. a Average þ S.D.s of three replicates.
Using the experimental set-up described above, conjugation in bulk soil was compared to conjugation in the residuesphere of leaves. In addition, conjugation on leaves with inhibited fungal colonisation was compared to conjugation on colonised leaves. Transconjugants appeared on day 0 in soil microcosms. This was most probably due to formation of transconjugants in the microcosms in the period between inoculation and sampling. The following sampling dates, and for both treatments, conjugal gene transfer was signi¢cantly higher (P 6 0.02) in the residuesphere than in bulk soil (Fig. 3). No plate mating was found.
FEMSEC 1085 27-12-99
42
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
Fig. 2. In£uence of fungal colonisation on bacterial colonisation of the residuesphere. Error bars shows S.D.s of three replicates.
Average densities of donors plus recipients per g of barley leaves were between 9 (day 1) and 198 (day 13) times higher than the average number of donors plus recipients per g soil (calculated from data to Fig. 3). On sampling days 1, 2 and 5, conjugal gene transfer was signi¢cantly higher (P 6 0.01) in the residuesphere of leaves with inhibited fungal populations compared to conjugal gene transfer in the residuesphere of leaves with intact fungal populations. The e¡ect of treatment of the leaves with a fungicide was transient and the conjugation e¤ciencies were equalised by day 9. No signi¢cant inverse correlation between cellulase activity and treatment of leaves with the fungicide was observed (data not shown), but this was most probably due to a limited number of
fungal species expressing cellulase activity (see Section 4). Bacterial colonisation was not a¡ected by the fungicide treatment of the leaves in controls without fungi. This was tested by comparing the numbers of cells extracted from natamycin-treated leaves and PBS-treated leaves incubated in sterile sand microcosms with donor and recipient cells (data not shown). 3.5. E¡ect of pre-colonisation by soil fungi on the appearance of transconjugants in the residuesphere In another approach to compare conjugation in the residuesphere of fungi-colonised leaves and non-colonised leaves, some leaves were pre-colonised with soil fungi for
Fig. 1. Transfer e¤ciency in bulk sand and in the residuesphere of barley leaves. Points represents replicate microcosms. Symbols on baseline show transfer e¤ciencies of 1U1037 or less. Metabolic activities are shown in Table 1.
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
43
Fig. 3. Transfer e¤ciency in bulk soil, in the residuesphere and in the residuesphere of fungicide-treated leaves. Points represent replicate microcosms. Symbols on baseline show transfer e¤ciencies of 1U1036 or less.
5 or 17 days before transfer to sterile sand for the subsequent conjugation experiment. Non-colonised leaves were also tested in sterile sand to serve as a control. In this way, conjugation on colonised leaves versus completely noncolonised leaves could be evaluated. Also, conjugation on leaves treated brie£y in PBS was determined in order to be able to evaluate the in£uence of senescence of the leaves on conjugation. In theory, the di¡erent incubation
periods could alter the leaves and thereby possibly the bacterial colonisation and/or conjugation e¤ciency. There were no signi¢cant di¡erences in the numbers of transconjugants in the residuesphere of non-colonised leaves as compared to the residuesphere of leaves exposed to 5 days of fungal pre-colonisation. Numbers of transconjugants were however signi¢cantly lower for leaves exposed to 17 days of fungal pre-colonisation as compared
Fig. 4. Number of transconjugants in the residuesphere of barley straw leaves. The leaves were pre-colonised by soil fungi or incubated on sterile sand and then incubated for 48 h in sand microcosms with addition of donor and recipients before enumeration of transconjugants. Relative chitinase activities of pre-colonised and non-colonised leaves are also shown.
FEMSEC 1085 27-12-99
44
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
to non-colonised leaves (P 6 0.05) (Fig. 4). No plate mating was found. Conjugation e¤ciencies (transconjugants/(donorsUrecipients)1=2 ) from triplicate samples were approximately the same from 0 and 5 days: 2.10U1033 þ 8.85U1034 for 0 day leaves, 1.31U1033 þ 5.52U1034 and 2.53U 1033 þ 1.73U1033 for non-colonised and colonised 5 days leaves, respectively. On leaves exposed to 17 days of fungal pre-colonisation, however, transconjugants/ (donorsUrecipients)1=2 was signi¢cantly lower in the residuesphere of colonised leaves than in the residuesphere of non-colonised leaves (P 6 0.05). Numbers of transconjugants/(donorsUrecipients)1=2 were 1.46U1033 þ 33 35 1.38U10 and 5.75U10 þ 5.10U1035 on non-colonised and fungi-colonised leaves, respectively, on leaves exposed to 17 days of fungal pre-colonisation. Comparing numbers of transconjugants on non-colonised leaves, a minor e¡ect of senescence of the leaves was observed (Fig. 4). However, conjugation e¤ciencies were not a¡ected. In this experiment, chitinase activity was used as a measurement of fungal biomass. Chitinase activity increased with increasing numbers of incubation days, thus the appearance of transconjugants in the residuesphere was inversely related to the fungal biomass. 4. Discussion Increased conjugation ratios for a TOL plasmid between introduced donor and recipient strains were demonstrated in both soil microcosm and control sand experiments. These observations are in accordance with increased gene transfer reported in other microhabitats. Van Elsas et al. [5] found transfer of the plasmid RP4 between introduced donor and recipient strains in wheat rhizosphere soil but not in non-rhizosphere soil, even when organic matter was added to the bulk soil. Also the phylloplane of bean has been suggested as a hot-spot for conjugation compared to conjugation on polycarbonate ¢lters [7]. Conjugation has previously also been studied in the spermosphere and rhizosphere of barley and higher conjugation rates in both microhabitats were detected in comparison with that found in unamended bulk soils [8]. The observed transfer rates in the residuesphere of barley leaves colonised by fungi in the soil microcosms ranged from 8.96U1035 to 7.94U1034 transconjugants/(donors recipients)1=2 and were comparable to transfer rates previously observed in the rhizosphere. An underestimation of the numbers of transconjugants in the present study cannot be ruled out, since impeded expression of a Tc resistance determinant in conjugation experiments has been reported [18]. SÖrensen and Jensen [8] reported 1.59U1034 ^ 2.51U1034 transconjugants/(donorsUrecipients)1=2 in barley rhizosphere and van Elsas et al. [5] found between 6.6U1037 and 3.8U1034 transconjugants per donor in
wheat rhizosphere. However, the transfer e¤ciencies in the residuesphere were lower than reported from the phylloplane, where up to 2.6U1032 transconjugants per donor and 3.4U1031 transconjugants per recipient were found [7]. Conjugation requires cell to cell contact, which may o¡er an explanation for the increased gene transfer rates observed in the residuesphere and other microbial hotspots. Average densities of donors plus recipients per g barley leaf in the present study were higher than the average number of donors plus recipients per g soil. The activity of bacteria extracted from the residuesphere of barley leaves incubated in sand microcosms was not higher than found for bacteria extracted from bulk sand. This is most probably due to the inability of the Pseudomonas strains to degrade the barley leaves, since the residuesphere, being a microhabitat with a high carbon turnover, normally would be an obvious environment for highly active bacteria. However, since conjugation was increased in the residuesphere of barley leaves compared to bulk sand, this enhanced gene transfer was not due to a higher metabolic activity of the bacteria, rather due to the higher donor and recipient densities on the surface of the leaves. Similarly, Normander et al. [7] found that metabolic activity was not rate limiting for conjugation, with no correlation between these two parameters being detected. Increased bacterial colonisation was observed on leaves pre-treated with a fungicide in this study, suggesting an antagonistic interaction between the indigenous soil fungal populations and the Pseudomonas donor and recipient strains. Antagonistic interactions between bacteria and fungi have been reported in other environments similar to the residuesphere. MÖller et al. [19] found a reduction in C mineralisation rate when both fungi and bacteria were present as compared to only fungal colonisation on decomposing beech leaves. Another example of bacterial antagonism has been demonstrated in dune soils by the inhibition of fungal growth due to bacterial activity [20]. Conversely, antagonistic behaviour of fungi against bacteria on decaying wood, an environment comparable to the residuesphere, has been detected by scanning electron microscopy [21]. The nature of the bacterial inhibition observed in this study is not yet known, but some possible mechanisms include competition for space, production of antibiotics or a combination of these factors. Conjugation was also increased on fungicide-treated leaves. This could be due to the higher bacterial numbers and therefore higher cell density for donor and recipient cells in the residuesphere of leaves not colonised by fungi. However, it should be taken into consideration that the fungicide also reduces the numbers of protozoa, which have been shown to have an e¡ect on the bacterial survival [22]. The fungal biomass was measured as cellulase or chitinase activity [16]. Our approach to use cellulase activity as an indicator of fungal presence did not give a clear result.
FEMSEC 1085 27-12-99
G. SengelÖv et al. / FEMS Microbiology Ecology 31 (2000) 39^45
Cellulase activity showed no signi¢cant correlation with the observed decrease in bacterial colonisation and conjugation on leaves with intact fungal populations, as compared to leaves with inhibited fungal populations. However, this could be ascribed to the limited number of fungal species expressing cellulase activity [16]. In subsequent experiments, we performed measurements on chitinase activity as an indicator for fungal presence. This enzyme has been shown correlated with two independent indicators of fungal biomass [16]. Using chitinase as an indicator of fungal biomass, a strong positive correlation between time of exposure to fungi of the leaves and chitinase activity was observed. A negative correlation between chitinase activity and numbers of transconjugants as well as conjugation e¤ciency in the residuesphere was also observed. In conclusion, we have shown that (i) the residuesphere of leaves of barley straw is a hot-spot for gene transfer, (ii) the increased conjugation in the residuesphere was not caused by an elevated metabolic activity of the bacteria, (iii) fungal colonisation can inhibit bacterial colonisation of the residuesphere and (iv) fungal colonisation can inhibit bacterial conjugation in the residuesphere. Acknowledgements
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[15]
We wish to thank Niels Kroer for helpful suggestions and discussions and Morten Miller for help in the determination of fungal activity. This work was partially supported by the Danish Environmental Protection Agency (Gitte SengelÖv and SÖren J. SÖrensen).
[16]
[17]
[18]
References [1] Barkay, T., Liebert, C. and Gillman, M. (1993) Conjugal gene transfer to aquatic bacteria detected by the generation of a new phenotype. Appl. Environ. Microbiol. 59, 807^816. [2] Cresswell, N. and Wellington, E.M.H. (1992) Detection of genetic exchange in the terrestrial environment. In: Genetic Interactions among Microorganisms in the Natural Environment (Wellington, E.M.H. and van Elsas, J.D., Eds.), pp. 59^82. Pergamon Press, Oxford. [3] Hill, K.E. and Top, E.M. (1998) Gene transfer in soil systems using microcosms. FEMS Microbiol. Ecol. 25, 219^329. [4] SÖrensen, S.J., Kroer, N., SÖrensen, E., SengelÖv, G. and Barkay, T. (1996) Conjugation in aquatic environments. In: Molecular Microbial Ecology Manual (Akkermans, A.D.L., van Elsas, J.D. and de Bruijn, F.L., Eds.), 5.2.1., pp. 1^29. Kluwer Academic Publishers, Dordrecht. [5] van Elsas, J.D., Trevors, J.T. and Starodub, M.E. (1988) Bacterial
[19]
[20]
[21]
[22]
[23]
45
conjugation between pseudomonads in the rhizosphere of wheat. FEMS Microbiol. Ecol. 53, 299^306. Lilley, A.K., Fry, J.C., Day, M.J. and Bailey, M.J. (1994) In situ transfer of an exogenously isolated plasmid between Pseudomonas spp. in sugar beet rhizosphere. Microbiology 140, 27^33. Normander, B., Christensen, B.B., Molin, S. and Kroer, N. (1998) E¡ect of bacterial distribution and activity on conjugal gene transfer on the phylloplane of the Bush Bean (Phaseolus vulgaris). Appl. Environ. Microbiol. 64, 1902^1909. SÖrensen, S.J. and Jensen, L.E. (1998) Transfer of plasmid RP4 in the spermosphere and rhizosphere of barley seedling. Antonie Leeuwenhoek 73, 69^77. Ruttigman, C., Vicun¬a, R., Mozuch, M.D. and Kirk, T.K. (1991) Limited bacterial mineralization of fungal degradation intermediates from synthetic lignin. Appl. Environ. Microbiol. 57, 3652^3655. Smets, B.F., Rittmann, B.E. and Stahl, D.A. (1993) The speci¢c growth rate of Pseudomonas putida PAW1 in£uences the conjugal transfer rate of the TOL plasmid. Appl. Environ. Microbiol. 59, 3430^3437. Franklin, F.C.H., Bagdasarian, M., Bagdasarian, M.M. and Timmis, K.N. (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. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning - a Laboratory Manual. Cold Spring Harbor Laboratory Press, New York. Jensen, L.E., Kragelund, L. and Nybroe, O. (1998) Expression of a nitrogen regulated lux gene fusion in Pseudomonas £uorescens DF57 studied in pure culture and in soil. FEMS Microbiol. Ecol. 25, 23^ 32. Kroer, N., Barkay, T., SÖrensen, S. and Weber, D. (1998) E¡ect of root exudates and bacterial metabolic activity on conjugal gene transfer in the rhizosphere of a marsh plant. FEMS Microbiol. Ecol. 25, 375^384. Miller, M., Paloja«rvi, A., Rangger, A., Reeslev, M. and KjÖller, A. (1998) The use of £uorogenic substrates to measure fungal presence and activity in soil. Appl. Environ. Microbiol. 64, 613^617. Smit, E. and van Elsas, J.D. (1990) Determination of plasmid transfer in soil: Consequences of bacterial mating on selective agar media. Curr. Microbiol. 21, 151^157. O'Morchoe, S.B., Ogunseitan, O., Sayler, G.S. and Miller, R.V. (1988) Conjugal transfer of R68.45 and FP5 between Pseudomonas aeruginosa strains in a freshwater environment. Appl. Environ. Microbiol. 54, 1923^1929. MÖller, J., Miller, M. and KjÖller, A. (1999) Fungal-bacterial interaction on beech leaves : In£uence on decomposition and dissolved organic carbon quality. Soil Biol. Biochem. 31, 367^374. De Boer, W., Klein Gunnewiek, P.J.A., Lafeber, P., Janse, J.D., Spit, B.E. and Woldendorp, J.W. (1998) Anti-fungal properties of chitinolytic dune soil bacteria. Soil Biol. Biochem. 30, 193^203. Tsuneda, A. and Thorn, R.G. (1995) Interactions of wood decay fungi with other microorganisms, with emphasis on the degradation of cell walls. Can. J. Bot. 73, 1325^1333. SÖrensen, S.J., Schyberg, T. and RÖnn, R. (1999) Predation by protozoa on Escherichia coli K12 in soil and transfer of resistance plasmid RP4 to indigenous bacteria in soil. Appl. Soil Ecol. 11, 79^90. Ba®a®th, E. (1994) Measurement of protein synthesis by soil bacterial assemblages with the leucine incorporation technique. Biol. Fertil. Soils 17, 147^153.
FEMSEC 1085 27-12-99