Cooperative dissolved organic carbon assimilation by a linuron-degrading bacterial consortium

Authors

  • Benjamin Horemans,

    Corresponding author
    • Division of Soil and Water Management, Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering, KU Leuven, Leuven, Belgium
    Search for more papers by this author
  • Johanna Vandermaesen,

    1. Division of Soil and Water Management, Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering, KU Leuven, Leuven, Belgium
    Search for more papers by this author
  • Erik Smolders,

    1. Division of Soil and Water Management, Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering, KU Leuven, Leuven, Belgium
    Search for more papers by this author
  • Dirk Springael

    1. Division of Soil and Water Management, Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering, KU Leuven, Leuven, Belgium
    Search for more papers by this author

Correspondence: Benjamin Horemans, Division of Soil and Water Management, Department of Earth and Environmental Sciences, Faculty of Bioscience Engineering, KU Leuven, Kasteelpark Arenberg 20 – bus 2459, B-3001 Leuven, Belgium. Tel.: ++3216329675; fax: ++3216321997; e-mail: benjamin.horemans@ees.kuleuven.be

Abstract

Dissolved organic matter (DOM) is the primary environmental carbon source for heterotrophic bacteria and its quality and quantity have been shown to affect microbial community structure and functioning. In that context, it was examined whether a bacterial consortium synergistically degrading the herbicide linuron extends this synergism toward natural DOM degradation. Biodegradable dissolved organic carbon (BDOC) of DOM of various origins and concomitant growth was determined for the consortium members in isolation and in combination. BDOC decreased with increasing DOM aromaticity, which is a recalcitrance indicator. BDOC in DOM of low aromaticity was 40–50% for all inocula. For DOM with high aromaticity, BDOC decreased with increasing aromaticity and was inoculum dependent, that is, BDOC was > 23% for consortium members in combination compared with BDOC < 16% for isolated strains. The observed BDOC and growth indicated that synergism existed within the consortium for degradation of DOM of both low and high aromaticity. All members benefited from this synergism as growth of all strains increased when incubated in combination, but their relative growth benefit depended on DOM composition. These data suggest that consortia benefit more than individual pesticide degraders from environmental DOM due to cooperation. This is important to understand the effects of DOM on stability and activity of pollutant-degrading consortia.

Introduction

Synergistic cooperation within bacterial consortia is important for the execution of several ecosystem functions such as the degradation of toxic and recalcitrant xenobiotic organic pollutants including pesticides (Feigel & Knackmuss, 1993; Bradley, 2003; Katsuyama et al., 2009). For instance, Dejonghe et al. (2003) described a triple-species bacterial consortium whose members synergistically degrade the phenyl urea herbicide linuron. Within this consortium, Variovorax sp. WDL1 transforms linuron into 3,4-dichloroaniline (3,4-DCA) and N,O-dimethylhydroxylamine (N,O-DMHA). 3,4-DCA is further mineralized by WDL1 but inefficiently and the excreted 3,4-DCA is used by Comamonas testosteroni WDL7, while N,O-DMHA is used by Hyphomicrobium sulfonivorans WDL6.

In the environment, pesticides form an unreliable C-source as the concentrations are dynamic and often very low (Lapworth et al., 2006; Schwarzenbach et al., 2006). Instead, a variety of dissolved organic compounds each of them present at low concentrations in natural dissolved organic matter (DOM) are the most important carbon and energy sources for heterotrophic bacteria (Metting, 1993). The quality of DOM, that is, the extent to which DOM is biodegraded, depends on intrinsic properties and the microbial community involved (Marschner & Kalbitz, 2003). Variations in DOM biodegradation between 4% and 93% were reported (Kalbitz et al., 2003a, b). DOM shows additionally temporal and spatial variations in quality and quantity, inherently linked with the DOM source (Volk et al., 1997). The quantity and quality of DOM have been shown to impact abundance, diversity and activity of heterotrophic microbial communities in various ecosystems (Kaplan & Bott, 1983; Methé et al., 1998; Pinhassi et al., 1999; Stepanauskas et al., 1999; Crump et al., 2003; Eiler et al., 2003; Findlay et al., 2003; Jardillier et al., 2004; Docherty et al., 2006; Kritzberg et al., 2006; Langenheder et al., 2006) and it can be hypothesized that DOM also affects the interactions and combined functionality of specific microbial consortia involved in the degradation of xenobiotic compounds such as pesticides. DOM has been shown to sustain organic pollutant-degrading bacteria in carbon-limited environments (Langwaldt et al., 2005). In single-strain systems, the availability of other organic compounds besides the pollutant enhanced pollutant degradation by providing carbon and energy for supplementary growth (Topp et al., 1988; Wick et al., 2003) or by improving pollutant bioavailability (Smith et al., 2011). Supplementary organic carbon has however also been shown to inhibit pollutant-degrading activity (McFall et al., 1997; Rentz et al., 2004).

In a previous study (Horemans et al., 2012), the above mentioned linuron-degrading bacterial consortium displayed an increased metabolic performance and metabolic range in oxidizing regular C-substrates such as carbohydrates, carboxylic acids and amino acids compared with its individual members. The consortium members apparently also display synergism in the metabolism of C-substrates other than linuron. We hypothesize that this cooperative feature extends to organic carbon compounds present in natural DOM. To examine this hypothesis, biodegradable dissolved organic carbon (BDOC) of DOM and concomitant growth were determined for individual consortium members and the consortium as a whole. The DOM tested was of various origins and included both DOM of low and high recalcitrance. For instance, in surface water, allochthonous DOM originating from the terrestrial environment and autochthonous DOM originating from primary production by algal biomass have different quality (Kaplan & Bott, 1982). Exudates released by algal biomass are rapidly utilized (Chen & Wangersky, 1996) and lead to substantial heterotrophic bacterial secondary production (Cole et al., 1988; Kaplan & Bott, 1989), while allochthonous DOM is more recalcitrant, although sustaining the heterotrophic community to a great extent (Cole et al., 2000; Kritzberg et al., 2005). We hypothesize that DOM is more efficiently metabolized by a consortium than by the members in isolation and that this metabolic efficiency will depend on DOM recalcitrance. Easily degradable DOM will entice competition between consortium members for the use of readily available C-compounds, while complex organic compounds present in recalcitrant DOM will demand the cooperative behavior of a consortium for metabolism. The linuron-degrading strain Variovorax sp. SRS16 (Sørensen et al., 2003) that degrades linuron efficiently on its own was included in the study. This paper represents the first study on DOM degradation by a defined microbial consortium.

Materials and methods

Bacterial strains and culture conditions

Strains Variovorax sp. WDL1, C. testosteroni WDL7, H. sulfonivorans WDL6 (Dejonghe et al., 2003) and Variovorax sp. SRS16 (Sørensen et al., 2003) were previously described. Red (Rfp) and yellow fluorescent protein (Yfp)–tagged derivatives of C. testosteroni WDL7 (designated as WDL7-Rfp) and H. sulfonivorans WDL6 (designated as WDL6-Yfp) were used (Breugelmans et al., 2008). Strains WDL1 and SRS16 were cultured in R2A (Reasoner & Geldreich, 1985). Strain WDL7-Rfp was grown in Luria–Bertani (LB) medium (Sambrook & Russell, 2001) supplemented with 50 mg L−1 of kanamycin. Strain WDL6-Yfp was grown in MMO (Dejonghe et al., 2003) supplemented with 1% (v/v) methanol and 50 mg L−1 of kanamycine. Cultures were harvested in late exponential phase and washed three times with 10 mM MgSO4 solution.

Preparation of DOM formulations

Nine different DOM formulations were prepared representing different DOM sources. Trisodiumcitrate (CIT) was selected as a reference DOM, representing an easily degradable model carbon substrate. Citrate was dissolved in Milli-Q water at 500 mg L−1 and filter-sterilized (0.22 μm). Two formulations consisted of either the humic acid or fulvic acid fraction of DOM. Humic acids and fulvic acids are two recalcitrant fractions of DOM. The humic acids were isolated from peat soil (Zegveld, the Netherlands) as described by Buekers (2007) and freeze-dried. The fulvic acids were reference fulvic acids (Suwannee River, Georgia) purchased from the International Humic Substance Society. Humic acids and fulvic acids were dissolved in Milli-Q water at 500 mg dry material L−1 and sequentially filtered over 0.45- and 0.22-μm filters to obtain sterile solutions. Leachate from maize leaves represented terrestrial DOM and was prepared as follows. Maize leaves collected from standing crops before harvest were oven-dried at 60 °C for 48 h. Leaves were cut and sieved to obtain the fraction < 2 mm. Ten grams of leaf material was added to 1 L of aqueous 10 mM CaCl2 solution and incubated (25 °C, 1 h, shaken at 150 r.p.m.). The suspension was filtered over 0.45-μm filter to remove particulate matter and filter-sterilized (0.22 μm). To derive DOM from maize leaf material incubated in soil, dried leaf material (< 2 mm) was mixed (1% w/w) with topsoil from an agricultural field (Ter Munck, Leuven) and incubated (25 °C, 21 days, soil with a moisture content corresponding to its water-holding capacity). Soil mixtures were watered every 2–3 days and freely drained. DOM was extracted with 10 mM CaCl2 solution (1 : 1) by head-over-end shaking for 1 h. The extract was centrifuged at 7505 g for 15 min at 4 °C. The supernatant was filtered sequentially over 5-μm and 0.45-μm filters. DOM in the filtrate was concentrated by a reverse osmosis (RO) system (Aramis NV, Belgium) equipped with a high-pressure pump and RO membrane (Filmtec TW30-2521 RO membrane, Dow) applying pressure of 10 bar over the RO membrane as described (Van Moorleghem et al., 2011). Before RO, the filtrate was passed over a cation exchange resin (lewatit resin, Lanxess) exchanging divalent cations for Na+ preventing DOM precipitation during the concentration process. Finally, the solution was filter-sterilized (0.22 μm). DOM based on algae exudates representing DOM from primary production was obtained as follows. Algal material (20% wet weight), collected from a freshwater pond (Kasteelpark Arenberg, Leuven, Belgium) during algal bloom of the green alga Oedogonium sp., was added to 10 mM CaCl2 and incubated while shaking at 100 r.p.m. at 25 °C for 24 h to allow exudation. Afterward, the algal suspension was filtered over 0.45-μm and 0.22-μm filters. River DOM was prepared as follows. Surface water (50 L) was collected in autumn 2009 from both the Urftall Sperre river (Germany) and the Rourbron spring (Belgium). Urftall sperre is a wide, open river with a dissolved organic carbon (DOC) concentration of 5 mg L−1 at sampling. The Rourbron is a freshwater spring located in a peat moor and contained 20 mg L−1 DOC. River DOM was concentrated by RO as described previously for the DOM derived from the soil-incubated leaf extract and filtered-sterilized (0.22 μm). All DOM was stored in an autoclaved glass container (SCHOTT®) at 4 °C. The DOM was regularly checked for stability based on aromaticity and DOC concentration and on sterility by plate counting on R2A agar plates.

Characterization of DOM

DOC concentration, as measure for DOM quantity, was measured using a TOC analyzer (Analytik Jena multi N/C 2100) as described (Van Moorleghem et al., 2011). To determine concentrations of free carbohydrates (FCH), free amino acids and proteins in DOM, DOM stock solutions were diluted to 20 mg L−1 DOC with purified water (Milli-Q®). Milli-Q water was used as blank. FCH concentrations were assayed by reaction with p-hydroxybenzoic acid hydrazide using glucose (Merck) as internal standard at 0, 25, 50 and 75 μM as described by Lever (1972). Absorbance was measured at 405 nm. α-NH2 end groups present in free amino acids and primary and secondary amines were assayed with ninhydrin (Moore & Stein, 1948) using leucine (Sigma-Aldrich) as internal standard at 0, 25, 50 and 75 μM. Absorbance was measured at 570 nm. Proteins were determined by the modified Lowry protein assay (Lowry et al., 1951; Peterson, 1977) using bovine serum albumin (Biosciences) as internal standard at 0, 0.37, 0.75, 1.12 μM. Absorbance was measured at 750 nm. UV absorbance at 254 nm (UVA254) was measured after diluting DOM to 20 mg L−1 DOC with Milli-Q. Specific UV absorbance (SUVA254), as a measure for aromaticity (Kalbitz et al., 2003a, b), was calculated as UVA254 normalized for the DOC concentration (L g−1 TOC cm−1). Absorbance was measured with a PerkinElmer Lambda 20 spectrophotometer. Measurements of FCH, proteins and free amino acids were taken in triplicate using three different samples of DOM. DOC analysis was carried out once on triplicate samples of each DOM solution.

Determination of BDOC

BDOC of DOM was determined according to a method modified from McDowell et al. (2006) with an incubation time of 14 days instead of 7 days to capture sufficient DOC removal by defined cultures and to allow separation of the biodegradable pool into a readily and slowly degradable fraction (Kalbitz et al., 2003a, b). As incubation of 14 days captured a large part of the BDOC and sufficed to differentiate among DOMs and biodegradation potential, incubation longer than 14 days was not considered (McDowell et al., 2006). All used glassware was baked at 550 °C for 24 h to oxidize residual organic carbon. DOM was adjusted to 20 mg DOC L−1 in Milli-Q water and ion compositions were measured as described (Van Moorleghem et al., 2011). Consequently, taking into account the ionic composition of each DOM, macronutrients and trace elements were added equally to those applied in the minimal medium MMO (Dejonghe et al., 2003) and furthermore adjusted to 1.59 mM NaHCO3, 1.11 mM NaCl, 0.54 mM Na2SO4, and 0.32 mM NaNO3 to obtain equal ionic composition among all DOM. Solutions were filter-sterilized (0.22 μm), and 10 mL of each solution transferred to triplicate reaction tubes. Tubes were inoculated at an initial cell density of 103 cells mL−1 of strains SRS16, WDL1, WDL6-Yfp or WDL7-Rfp separately and a combination with equal proportions of WDL1, WDL6-Yfp and WDL7-Rfp. As suggested by McDowell et al. (2006), an undefined microbial community with high biodiversity was used to capture the maximal attainable BDOC for comparison with these displayed by defined inocula. This undefined soil microbial community was extracted from topsoil of an agricultural field (Ter Munck, Leuven, Belgium) by shaking 5 g of sieved (< 2 mm) soil head-over-end shaking for 1 h in 45 mL of 10 mM MgSO4. After sedimentation of soil particles for 30 min, 100 μL of a dilution of the liquid phase (109 cells mL−1) by factor 104 was used as inoculum. Tubes were incubated at 25 °C on a rotary shaker (100 r.p.m.). For TOC analysis, samples were taken at days 0, 3, 7, and 14 and filter-sterilized (0.22 μm). Samples were put in 2-mL SNAP CAP™ (Chromacol®) vials, acidified (pH 3) with 5 M HCl solution to inhibit bacterial growth, while avoiding precipitation of humic acids (pH < 2) and stored at −20 °C prior to analysis. BDOC was calculated as

display math

where DOCTi,inoc/ctrl is the DOC concentration at day i (0, 3, 7, or 14) in either inoculated (inoc) or not-inoculated (ctrl) tubes.

Determination of growth and assimilable organic carbon (AOC)

Samples of DOM inoculated with WDL1, WDL6-Yfp, WDL7-Rfp, SRS16 and the linuron-degrading consortium were taken during the BDOC experiment at days 0, 7, and 14 and serially 10-fold diluted up to dilution factor 107. Dilutions containing single strains were plated on R2A agar and colony-forming units (CFU) counted after 4 days' incubation (25 °C). Cell numbers in DOM inoculated with the consortium were counted with a Helber Counting Chamber (Hawksley) with epifluorescence microscope (400×) (Olympus). WDL7-Rfp and WDL6-Yfp were counted using their specific fluorescence. Total cell numbers were counted in bright-field mode. Plate counting and microscopic counting of WDL1, WDL6-Yfp, and WDL7-Rfp cells yielded equal results when compared beforehand. Microbial growth was determined as difference in cell numbers between day 0 and day 7 or 14. AOC (van der Kooij et al., 1982), defined as microbial-C assimilated during net growth, was calculated using the conversion factor 26.24 ± 1.08 fg C cell−1 (Troussellier et al., 1997). Assimilation efficiency (%) of the consortium in each DOM was calculated as the ratio of corresponding AOC (mg C L−1) to BDOC (mg C L−1). Assimilation efficiency was calculated from either measured AOC and BDOC in each DOM or using predicted values from regression analysis of ln(AOC) and BDOC in function of SUVA254 on data from measurements taken after 14 days of incubation.

Statistical analysis

Statistical significant differences (< 0.05) were determined using Student's t-test (paired, two-tailed distribution) between final BDOC, SUVA254 or UVA254 (n = 3) of the different inocula for each DOM.

Results

DOM formulations and characteristics

DOM characteristics indicative of the microbial use of DOM as source of carbon and energy were determined (Table 1). The highest content of αNH2 groups, as a measure of free amino acids, primary and secondary amines, and proteins, was found in DOM derived from algae exudates and maize leaf leachate, followed by soil-incubated leaf extract and river DOM from the Urftall Sperre. Other DOM formulations contained much less αNH2 groups. The highest content of carbohydrates was measured in DOM of maize leaf leachate. Other DOM formulations contained three- to fivefold less. Protein contents were similar for all tested DOM.

Table 1. Initial characteristics of DOM and inoculum-specific final BDOC (% of initial DOC) as measured at day 14
DOM sampleAbbrevation SUVA254 L g−1 TOC cm−1 αNH2 Carbohydrates μM mM C−1 PeptidesBDOC (%)
WDL1WDL7SRS16CONSSMC
  1. The DOM samples are ordered according to increasing aromaticity (SUVA254).

  2. a

    Significant difference between the BDOC of the consortium (CONS) and the individual strains WDL1, WDL7, SRS16, and SMC as indicated based on student's t-test (n = 3; significance level at 0.05).

  3. Measured values with 95% confidence intervals (n = 3) or (n = 12).

CitrateCIT5.1 ± 0.00.0 ± 1.01.2 ± 0.90 ± 16
Algae exudatesAEx9.0 ± 0.923.3 ± 5.22.6 ± 1.440 ± 1449 ± 1a50 ± 236 ± 544 ± 269 ± 1a
Maize leaf leachateMLL15.4 ± 0.324.3 ± 2.943.8 ± 2.568 ± 1050 ± 253 ± 149 ± 254 ± 473 ± 1a
Humic acidsHA25.5 ± 0.32.6 ± 1.39.1 ± 1.382 ± 2235 ± 620 ± 5a7 ± 2a39 ± 247 ± 2a
Soil-incubated leaf extractsSILE30.8 ± 0.610.0 ± 1.95.0 ± 1.132 ± 1721 ± 2a19 ± 111 ± 3a26 ± 332 ± 1a
Fulvic acidsFA37.0 ± 0.80.0 ± 2.68.6 ± 1.962 ± 1616 ± 415 ± 49 ± 1a18 ± 121 ± 2
River DOM Urftall SperreURF37.4 ± 0.18.4 ± 2.99.9 ± 0.758 ± 2813 ± 4a14 ± 5−3 ± 4a23 ± 429 ± 2
River DOM RourbronROUR46.5 ± 0.02.1 ± 1.014.3 ± 1.398 ± 32−2 ± 2a1 ± 7a−4 ± 3a23 ± 124 ± 5

Determination of BDOC

Organic C removal from DOM, expressed as BDOC, in function of time by strains WDL1, SRS16, WDL7-Rfp, the consortium and the undefined soil microbial community was measured (Fig. 1). Final BDOC (at day 14) and significant differences in BDOC between inocula are reported in Table 1. No significant DOC removal from DOM was observed in noninoculated controls or in cultures with WDL6-Yfp. All other strains and the consortium showed DOC removal for most DOM. The soil microbial community removed most DOC from all DOM with most DOC removed in the first 3 days. All inocula showed high DOC removal in the first days from DOM of algae exudates and maize leaf leachate that is characterized by a low SUVA254. Individual strains resulted in significantly lower DOC removal compared with the soil microbial community and a longer incubation (∼14 days) was necessary to achieve maximal BDOC especially for DOM with high SUVA254. For most DOM, the slowest and lowest DOC removal was recorded for SRS16. Interestingly, the time needed for DOC removal and final BDOC displayed by the consortium equaled those exhibited by its individual members for DOM with low SUVA254 (algae exudates and maize leaf leachate). For several DOM with a high SUVA254 (river DOM from Rourbron and from Urftall Sperre), final BDOC displayed by the consortium was higher compared with individual strains and even equaled that recorded for the soil microbial community. SUVA254, hence, the aromaticity of DOM is a determining factor for the degree of DOC removal for all inocula. Indeed, for each inoculum, the final amount of DOC removed was linearly correlated with initial SUVA254 (Fig. 2). The slope of regressions between BDOC and SUVA254 does not differ significantly between individual strains and the soil microbial community but was significantly lower for the consortium. The BDOC for the consortium converged with the BDOC displayed by the soil microbial community for DOM with high SUVA254.

Figure 1.

The BDOC as percentage of initial DOC in function of degradation time determined for the different inocula and different DOM.

Figure 2.

The BDOC (fraction of total DOC, day 14) for the seven different DOM and inocula in function of the initial SUVA254. Variovorax sp. WDL1 (●), Variovorax sp. SRS16 (▼), Comamonas testosteroni WDL7-Rfp (○), linuron-degrading consortium consisting of WDL1, WDL6-Yfp, and WDL7-Rfp (∆) and SMC (■) and a noninoculated negative control (□). Linear regression is shown for WDL1 (R2 = 0.94), WDL7-Rfp (R2 = 0.88), SRS16 (R2 = 0.77), CONS (R2 = 0.73), SMC (R2 = 0.87) and the control (R2 = 0.01).

Changes in UV absorbance

Changes in the aromaticity of DOM formulations were determined by measuring SUVA254 and UVA254 values of residual DOM after 14 days of incubation in inoculated and noninoculated setups (Fig. 3). SUVA254 increased with increasing BDOC, suggesting degradation of mainly nonaromatic compounds. However, in some cases, a significant decrease in UVA254 occurred in inoculated setups compared with noninoculated setups and indicated degradation of aromatic compounds. A decrease in UVA254 was significant for all DOM inoculated with the soil microbial community and some DOM (soil-incubated leaf extract, humic acids, and river DOM from Rourbron and Urftall Sperre) inoculated with the consortium but did not occur for DOM inoculated with the single strains.

Figure 3.

SUVA254 (top) and UVA254 (bottom) (with standard deviations (n = 3)) of residual DOM at day 14 of incubation as determined for the different inocula and DOM (AEx, algae exudates; MLL, maize leaf leachate; SILE, soil-incubated leaf extract; HA, humic acids; FA, fulvic acids; URF, river DOM Urftall Sperre; ROUR, river DOM Rourbron). Values that differ significantly (significance level of 5%) from the noninoculated controls are indicated with an asterisk.

Determination of growth and AOC

To determine whether BDOC was assimilated into cell material, growth of the strains separately and as the consortium was assessed at days 7 and 14. Growth was observed with all DOM to final cell densities of approximately 106–108 cells mL−1 at day 14. No apparent difference in growth was observed between individually inoculated strains, except with citrate and DOM originating from maize leaf leachate. The consortium increased 10- to 100-fold in total cell number compared with individually inoculated strains for all DOM. Growth and AOC as a function of initial SUVA254 based on data from day 14 are shown in Fig. 4. Growth for all inocula clearly correlated with the initial SUVA254 and could be described by a logarithmic equation. A significantly greater growth was recorded for the consortium compared with individually inoculated strains for all DOM. Differences in growth among individual strains were mainly observed for DOM with lower SUVA254. Final cell numbers of WDL7-Rfp were higher than those of WDL1 and SRS16. The cell number of SRS16 was in turn higher than that of WDL1.

Figure 4.

DOM-associated microbial growth (primary y-axis) and its conversion into AOC (secondary y-axis), after 14 days of incubation in function of initial DOM SUVA254 as determined for Variovorax sp. WDL1 (●), Variovorax sp. SRS16 (▼), Comamonas testosteroni WDL7-Rfp (○) and the consortium consisting of WDL1, WDL6-Yfp and WDL7-Rfp (∆). Logarithmic fits of growth/AOC in function of SUVA254 are according to the equation [growth/AOC] = a·ln [SUVA254] + b, for WDL1 (R2 = 0.77), WDL7-Rfp (R2 = 0.79), SRS16 (R2 = 0.82) and the consortium (R2 = 0.77).

The cell numbers of strains WDL1, WDL6-Yfp and WDL7-Rfp when grown as the consortium on the different DOM depended on the initial SUVA254 of the DOM (Fig. 5). At day seven, growth of the consortium was mainly due to growth of WDL7-Rfp and decreased with increasing SUVA254. For instance, a threefold increase in growth was observed with citrate compared with river DOM from Rourbron. The consortium's total cell numbers at day 14 further increased with all DOMs but to a greater extent with DOM with higher SUVA254. This was especially due to an up to 10-fold increase in WDL1's cell number. Between day seven and day 14, WDL7-Rfp numbers increased only by 50% with easily degradable DOM (citrate, algae exudates and maize leaf leachate). For all three strains, growth in function of SUVA254 was fitted with a logarithmic equation. Growth of WDL7-Rfp logarithmically decreased from 80 ± 10% with citrate to 13 ± 3% with river DOM from Rourbron, while growth of WDL1 increased from 17 ± 4% to 85 ± 11%. The abundance of WDL6-Yfp remained low and slightly decreased with increasing SUVA254.

Figure 5.

DOM-associated microbial growth (primary y-axis) and its conversion into AOC (secondary y-axis), after 7 (left) and 14 days (right) of incubation in function of SUVA254 as determined for the linuron-degrading consortium that was composed of equal proportional fractions of the three members at the start. Total cell numbers (cells mL−1) of the consortium (■) derived from the cell numbers of the consortium members WDL1 (●), WDL6-Yfp (▲) and WDL7-Rfp (♦). Logarithmic fits of growth/AOC in function of SUVA254 are according to equation [growth/AOC] = a·ln [SUVA254] + b. R2 values of the regressions are 0.81, 0, 0.85 and 0.39 at 7 days and 0.76, 0.95, 0.92 and 0.44 at day 14 for, respectively, the consortium, WDL1, WDL6-Yfp and WDL7-Rfp.

After 14 days of incubation, a significantly higher cell number and growth were observed for the consortium compared with individually inoculated strains WDL1 and WDL7-Rfp. Growth as a function of initial SUVA254 for consortium members both in combination and in isolation (Fig. 6) showed that both strains reached higher cell numbers with all DOM in combination compared with that in isolation. Growth of WDL7-Rfp in the consortium increased up to factor 5 compared with that in isolation when incubated with easily degradable DOM and a factor 2 when incubated in the presence of more recalcitrant DOM. Growth of WDL1 in the consortium increased with a factor 3 with citrate till up to a factor 100 with less degradable DOM.

Figure 6.

DOM-associated microbial growth (primary y-axis) of Variovorax sp. WDL1 and Comamonas testosteroni WDL7-Rfp and its conversion into AOC (secondary y-axis) after 14 days of incubation in function of initial SUVA254 when strains were inoculated individually (●) or as a member of the linuron-degrading consortium (○). Logarithmic regressions are shown according to the equation y = a·ln (x) + b.

DOM-related AOC for the different inocula was calculated from microbial cell numbers (Figs 4–6). Final BDOC and final AOC only correlated for the consortium and was best described with a logarithmic equation (Fig. 7; left). DOM with higher BDOC showed also higher AOC. Interestingly, BDOC and AOC converged with increasing initial SUVA254 (Fig. 7; middle), indicating that low BDOC due to high DOM recalcitrance is correlated with high AOC. The DOM assimilation efficiency of the consortium increased from 40% to 80% when SUVA254 increased from 5 to 45 L g−1 cm−1 (Fig. 7; right). The assimilation efficiency displayed by individual inocula did not change in function of SUVA254 (data not shown). For WDL1, SRS16, and WDL7, a maximal assimilation efficiency of, respectively, 3.2 ± 0.6, 25.7 ± 3.0, and 14.7 ± 3.7% was calculated.

Figure 7.

Comparison of assimilation and biodegradation of DOC from DOM by the linuron-degrading consortium. Left: AOC, calculated as percentage of initial DOC, in function of BDOC was described by a logarithmic relation (AEx, algae exudates; MLL, maize leaf leachate; SILE, soil-incubated leaf extract; HA, humic acids; FA, fulvic acids; URF, river DOM Urftall Sperre; ROUR, river DOM Rourbron); middle: BDOC (●) and AOC (○) in function of initial SUVA254; right: C-utilization efficiency in function of initial SUVA254. C-utilization efficiency values are based on measured (●) and predicted (□) values.

Discussion

This study is the first assessing synergism within a defined microbial consortium in the utilization of DOM originating from different environmental sources. Variovorax sp. WDL1, Variovorax sp. SRS16, and C. testosteroni WDL7 showed significant degradation for various DOM, which is in accordance with the reported broad substrate range of those bacterial genera. Strains WDL1, SRS16 and WDL7 are members of the Comamonadaceae known as generalists with a low habitat specificity and wide metabolic range (Nemergut et al., 2010). Variovorax and Comamonas are also well known as degraders of recalcitrant compounds such as high molecular weight (HMW) and chlorinated aromatics (Ma et al., 2009; Han et al., 2010). For most DOM, BDOC recorded for WDL1 and WDL7 were similar except for humic acids, suggesting a different substrate range for WDL1 and WDL7 at least for compounds present in humic acids. On the other hand, for many DOM, strain SRS16 showed a lower final BDOC compared with WDL7 and WDL1, in particular for DOM with high SUVA254 and hence with higher recalcitrance. However, differences are not due to degradation capacities toward the more recalcitrant aromatic fraction because UVA254 values did not change during DOM degradation, suggesting that the differences are rather due to differences in metabolic capabilities toward the easily biodegradable DOM fraction. Also, in the GN2 MicroPlate assay, strain SRS16 showed a lower substrate range than strains WDL1 and WDL7 (Horemans et al., 2012). WDL6 showed no degradation for any DOM as expected as Hyphomicrobium grows best on low carbon content organic substrates (Moore, 1981; Bergey & Holt, 1994). As DOM contains a mixture of compounds each present at relatively low concentrations, our data indicate simultaneous utilization of different substrates. Kovárová-kovar & Egli (1998) suggested that such a strategy can be used by heterotrophic bacteria to survive and grow in carbon-limited environments.

The consortium removed more DOC from the various DOM compared with individually inoculated strains, indicating that the three strains cooperate in degrading environmental DOM. Moreover, the observed decrease in aromaticity of DOM formula (like river DOM from Rourbron), containing a considerable recalcitrant fraction, after DOC removal by the consortium indicates that this metabolic synergism includes the degradation of aromatic compounds other than linuron (Dejonghe et al., 2003). On the other hand, as changes in UVA254 were minimal and SUVA254 increased, cooperation mainly existed in degradation of more easily biodegradable fraction. Cooperation in the metabolism of easily degradable C-sources within the studied consortium was previously suggested by comparing metabolic fingerprints of the consortium with those of the individual strains in GN2 microplate assays (Horemans et al., 2012). Such metabolic cooperation can be based on the existence of complementary pathways in different members (Nielsen et al., 2000) or exchange of cofactors (Miura et al., 1980; Graber & Breznak, 2005; Rezaïki et al., 2008). Interestingly, the synergistic interactions between consortium members for DOM carbon assimilation appear to be mutualistic as increased growth was observed for all members compared with their growth in isolation. The extent of this benefit for each consortium member changed however with DOM quality.

The assimilable amount of organic carbon, calculated from bacterial growth using a conversion factor of 24 fg C per cell (Troussellier et al., 1997; Batte et al., 2003), indicated that more carbon was assimilated when the consortium was grown on recalcitrant DOM than when it was grown on easily degradable DOM. A reason could be that cells were smaller when grown on DOM with high recalcitrance compared with cells grown on DOM with low recalcitrance. Although bacterial cell size indeed depends on nutritional conditions, Troussellier et al. (1997) reported that C-contents per cell remained equal for several nonmarine bacteria despite differences in cell size.

Fractions in DOM are classified according to degradation by heterotrophic microbial communities into easily degradable, slowly degradable, and recalcitrant fractions (Kalbitz et al., 2003a, b). SUVA254 is often used as measure for DOM biodegradability because of the negative linear correlation between SUVA254 and the observed extent of DOM mineralization by different microbial cultures (Kalbitz et al., 2003a, b). UVA254 of DOM is mainly determined by the concentration of aromatic compounds, indicating that the recalcitrant fraction in DOM is mainly composed of aromatics, including HMW compounds such as fulvic and humic acids (Moran & Hodson, 1990; Kalbitz et al., 2003a, b). Easily degradable fractions include more low molecular weight compounds such as carbohydrates, carboxylic acids and amino acids (Thurman, 1985). DOM biodegradability linearly correlated with DOM aromaticity for the undefined soil microbial community, the defined consortium and the single-strain cultures. This largely agrees with previous observations on the relationship between DOM source, DOM structure and reactivity. DOM origin allows to predict biodegradability of the isolated DOM (Volk et al., 1997). DOM produced from algae exudates and maize leaf leachate contained a high fraction of easily degradable compounds as their contact times with endogenous degraders were minimal before extraction. In contrast, DOM present in river water (river DOM from Rourbron and Urftall Sperre) has been susceptible to local degradation for a longer time and contained a greater fraction of less degradable substances.

In conclusion, DOM-quality effects on carbon removal and assimilation previously shown for undefined microbial communities also apply to defined pure strains and consortia. In addition, the consortium performed better than the individual strains showing that cooperative interactions between a limited number of strains can substantially improve environmental organic substrate utilization. It also shows that in addition to linuron, other synergistic metabolic interactions exist between consortium members. It would be interesting to determine whether synergism in DOM degradation contributes to maintaining composition and functionality of the consortium in the absence or poor availability of linuron. As the availability of supplementary C-sources affects organic pollutant degradation in several ways (Topp et al., 1988; McFall et al., 1997; Wick et al., 2003; Rentz et al., 2004; Langwaldt et al., 2005; Smith et al., 2011), information about DOM utilization within a pesticide-degrading consortium can be used to understand effects of environmental DOM on xenobiotic degradation. How such synergistic metabolic features affect linuron-degrading activity within the consortium remains to be elucidated. However, consortia should be included in examining the effects of supplementary C-sources on organic pollutant degradation, as consortia show a different DOM utilization profile compared with individual strains.

Acknowledgements

Research was funded by IWT (Innovation by Science and Technology) project SB/73381, OT10/03 project of KU Leuven, and the Inter-University Attraction Pole (IUAP) ‘μ-manager’ of the Belgian Science Policy (BELSPO, P7/25). We thank K. Simoens for the excellent technical support and C. Van Moorleghem for in-river water sampling and RO.

Ancillary