This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.
Agricola
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, January 2006, p. 212-220, Vol. 72, No. 1
0099-2240/06/$08.00+0     doi:10.1128/AEM.72.1.212-220.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Structure and Function of Bacterial Communities Emerging from Different Sources under Identical Conditions

Silke Langenheder,1* Eva S. Lindström,1,2 and Lars J. Tranvik1

Limnology/Department of Ecology and Evolution, Evolutionary Biology Centre, Uppsala University, Norbyvägen 20, 75236 Uppsala, Sweden,1 Département des Sciences Biologiques, Université du Québec à Montréal (UQÀM), Montréal, Québec, Canada2

Received 16 May 2005/ Accepted 5 October 2005


arrow
ABSTRACT
 
The aim of this study was to compare two major hypotheses concerning the formation of bacterial community composition (BCC) at the local scale, i.e., whether BCC is determined by the prevailing local environmental conditions or by "metacommunity processes." A batch culture experiment where bacteria from eight distinctly different aquatic habitats were regrown under identical conditions was performed to test to what extent similar communities develop under similar selective pressure. Differently composed communities emerged from different inoculum communities, as determined by terminal restriction fragment length polymorphism analysis of the 16S rRNA gene. There was no indication that similarity increased between communities upon growth under identical conditions compared to that for growth at the ambient sampling sites. This suggests that the history and distribution of taxa within the source communities were stronger regulating factors of BCC than the environmental conditions. Moreover, differently composed communities were different with regard to specific functions, such as enzyme activities, but maintained similar broad-scale functions, such as biomass production and respiration.


arrow
INTRODUCTION
 
Heterotrophic bacteria are major players in global biogeochemical cycles. In aquatic environments, they utilize dissolved organic matter (DOM), and the carbon flow depends on the fractions of the utilized organic carbon that are either respired directly or transferred to higher trophic levels within the food web. Information is limited on how the functional performance of heterotrophic bacterial communities is determined by community composition. There is evidence that different phylogenetic groups of bacteria preferably utilize certain components of the bulk DOM pool (9, 10), and changes in DOM composition may generate metabolic changes which are linked to altered community composition (e.g., see reference 19). On the other hand, large-scale ecosystem functions, such as heterotrophic respiration, may be independent of diversity or community composition, since the functional group "heterotrophic bacteria" is highly diverse and harbors extensive redundancy due to numerous coexisting species performing similar functions (28). In accordance with large-scale ecosystem functioning being independent of diversity or community composition, we have recently shown that there is only a weak link between bacterial community composition (BCC) and function in heterotrophic aquatic bacterial communities (40). This suggests that ecosystem functioning might be unrelated to bacterial community structure and diversity due to the presence of generalist species that are able to cope with a wide range of environmental conditions (40).

The assembly of communities at the local scale is influenced by both local environmental conditions (abiotic and biotic factors) and regional factors (e.g., climate, migration, and speciation), as shown, e.g., for zooplankton (7, 8). Similar processes have been suggested to be important in the regulation of bacterial community composition (32, 54). One view concerning the regulation of local bacterial community structure goes back to the famous statement "everything is everywhere—the environment selects" (1). Accordingly, based on the high local but low global diversity of microorganisms, the active community is selected from the total species pool by the prevailing environmental conditions (18). The idea that microorganisms are ubiquitously distributed is supported by studies of protozoan communities (17, 20) and studies reporting on the global distribution of closely related freshwater bacteria (e.g., see references 25, 29, 30, and 55). On the other hand, it is widely recognized that ecological differentiation and adaptation to certain environmental conditions occur even among major bacterial groups, since, e.g., beta-proteobacteria are dominant constituents of freshwater bacterioplankton but very rare in the ocean (24, 45, 55). Further on, the "everything is everywhere" concept is questioned by studies of soils and salt marshes showing increasing numbers of species with increasing sampled habitat sizes, in analogy with established species-area relationships demonstrated for higher organisms (26, 33).

Another view is that the local bacterial community structure is regulated by the size and diversity of the surrounding regional community, or "metacommunity" (11). Curtis and Sloan (11) also argue—based on the neutral theory of biodiversity (34)—that local bacterial community composition is a product of random events in connection to the recruitment of functionally equivalent bacterial taxa from the "meta- or source community." Hence, this concept does not assume that there is a one-to-one match between environment (or niche availability) and community structure.

The objective of this study was to compare these contrasting concepts of how bacterial communities are shaped and how bacterial community structure and function are related by using an experimental approach. Different bacterial communities were inoculated into incubation vessels with the same sterile medium, hence making it possible to investigate the importance of the origin of the source community (i.e., the inoculum) versus the environmental conditions (i.e., the medium) for the composition and functional performance of the emerging bacterial communities. Previous studies using dilution cultures often found much stronger effects of the medium than of the inoculum on functional parameters (21, 22, 53), which can be explained either by similarly composed communities being selected by identical media or by differently composed bacterial communities maintaining similar functions (40).

According to the "everything is everywhere—the environment selects" concept, initially different local communities are expected to converge towards similar composition upon growth under identical conditions. According to the "metacommunity concept," however, deviating communities consisting of functionally equivalent taxa should be enriched even under identical conditions.


arrow
MATERIALS AND METHODS
 
Sampling and experimental setup.
Grazer-free bacterial inocula obtained from eight sampling sites differing over a wide range of physicochemical conditions were grown in the same medium based on a natural dissolved organic matter extract. The rationale behind the selection of the study sites was to obtain inocula from habitats that differed widely in terms of water chemistry (oligotrophic versus eutrophic, high versus low dissolved organic carbon [DOC], alkaline versus acidic) (Table 1). Growth of bacteria in the cultures, where each treatment consisted of four replicates, was followed over a period up to 3 weeks, and community composition (terminal restriction fragment length polymorphism analysis [T-RFLP]) and general functional parameters (biomass production, growth rates, respiration, and growth efficiency) were measured. To assess more specific functions, activities of ectoenzymes involved in the degradation of polymers and proteins (ß-glucosidase and leucine-aminopeptidase, respectively) were measured, and the ability of the bacterial communities to degrade aromatic compounds was determined, using benzoic acid as a model substance.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Characteristics of habitats from which water for the preparation of inocula was retrieved

Sampling sites.
Surface water was sampled from the following eight habitats: (i) a rock pool (ROCK) located at the coast of the Hållnäs peninsula in the northern part of the province of Uppland in central Sweden (60°30'N, 18°00'E); the pool is small and shallow, situated in a rock crevice located approximately 100 m from the coastline, and contains rainwater and some minor impact of sea spray, with a salinity of 0.02 psu; (ii) a groundwater spring (GROW) close to the city of Uppsala (59°50'N, 17°46'E), where water is welling up to the surface, surrounded by agricultural area; (iii) Lake Eckarfjärden (ECKA), an oligotrophic hardwater lake close to the Baltic Sea (60°22'N, 18°12'E); (iv) a small highly stained forest stream inlet to Lake Lumpen (LUMP) in the Uppland region (59°58'N, 17°17'E); (v) Lake Vättern (VATT) (58°28'N, 14°55'E), the second largest lake in Sweden, a clearwater lake with a 60-year water retention time (http://www.vattern.org); (vi) Lake Ljustjärnen (LJUS), a small oligotrophic clearwater forest lake located in the area of Bergslagen, approximately 200 km northwest of Uppsala (59°55'N, 15°27'E); (vii) Lake Skottjärnen (SKOT), a small oligotrophic brownwater forest lake located in the Bergslagen area (59°57'N, 15°24'E); and (viii) a bog pool (BOG) in Uppland (60°03'N, 17°20'E) dominated by Sphagnum mats.

The sampling sites are up to 400 km apart. All water samples were taken on 19 April 2004, except for those from Lake Eckarfjärden, which was sampled on 16 April 2004. Two liters of water was sampled from a depth of ≤1m, depending on the depth of the water body. The water was stored at +4°C until the setup of the experiment (on 22 April). Physicochemical parameters (temperature, pH, conductivity, alkalinity, total phosphorus, total nitrogen, and chlorophyll a) were measured for each of the eight waters according to standard procedures. The DOC concentration was analyzed according to Langenheder et al. (39).

Experimental setup.
Bacterial cultures were set up in autoclavable 840-ml polycarbonate tubes. Each incubation vessel was closed without headspace with a polycarbonate piston with silicone sealings. The piston had two stainless steel pipes (one inlet and one outlet) with an inner diameter of 4 mm that were connected to Teflon tubing and sterile three-way valves. For sample collection, the outlet valve was opened and water was pressed out by moving the piston with the help of a hand lever.

The cultures were filled with autoclaved artificial lake water medium prepared according to the method of Bastviken et al. (3), using Nordic reservoir natural organic matter from the International Humic Substances Society (1R108N), diluted to a final DOC concentration of approximately 10 mg liter–1. A stock solution with a DOC concentration of 520 mg liter–1 was prepared from the freeze-dried powder and filtered twice through a sterile 0.2-µm Mini Filter polysulfone capsule (Gelman Sciences). Prior to addition to the cultures, the stock solution was filtered again through sterile 0.2-µm Supor syringe filters (Gelman Sciences). All filter units were extensively rinsed with autoclaved Milli-Q water prior to use to avoid contamination with external DOC. The pH in the cultures was 7.0 ± 0.03 (n = 3), and the alkalinity was 0.34 ± 0.02 meq liter–1 (n = 3).

For preparation of the inocula, bacteria from each of the eight sites described earlier were filtered through GF/F filters (Whatman; precombusted for 8 h at 450°C). Since the concentration of bacteria was substantially lower in the groundwater than in the other samples, groundwater bacteria were concentrated. To achieve this, the water volume of the GF/F filtrate was decreased by filtration through autoclaved 0.2-µm-pore-size 47-mm polycarbonate filters, resulting in an approximately sixfold increase in bacterial abundance compared to the value given in Table 1. Ten milliliters of inoculum was added to each incubation vessel, and the resulting batch cultures were incubated at 19°C in the dark. Bacterial growth in the cultures was monitored, and the experiment was stopped when no further net increase in bacterial abundance was observed. Hence, the experiment was stopped between 242 and 505 h after inoculation, depending on the treatment. Three controls with only artificial lake water and natural organic matter extract were prepared and remained sterile throughout the entire experiment. Moreover, by use of microscopy, flagellates were confirmed to be virtually absent from the filtered water samples at the final stage of the experiment.

Generally, all glass- and plasticware coming into contact with the samples was soaked for several hours in 1 M HCl and rinsed with excessive amounts of Milli-Q water afterwards.

Oxygen consumption.
The net change in the dissolved oxygen concentration in the cultures was measured according to a spectrophotometric modification of the Winkler titrimetric method (48) in 30-ml Winkler flasks. The absorbance at 450 nm was measured and transformed into dissolved oxygen concentration using a calibration curve given by Mille-Lindblom and Tranvik (46). Bacterial respiration (BR) was defined as O2 utilization in the cultures and was transformed into carbon units by assuming a respiratory quotient of 1 (12).

Bacterial abundance, biomass, growth, and growth efficiency.
Subsamples of 10 ml were taken at 20- to 26-h intervals (up to 48-h intervals at later stages of the experiment), preserved and stained as described previously (15), and analyzed flow cytometrically using a FACScan flow cytometer (Becton Dickinson), with minor modifications of the original protocol (13). The cell volume and carbon content per cell were determined and calculated as described earlier (15). The bacterial biomass yield (YB) was obtained by multiplying the obtained cellular carbon content and the respective value of maximum cell abundance in the cultures. The bacterial growth efficiency (BGE) was calculated according to the following formula: BGE = [YB/(YB + BR)] x 100. The maximum intrinsic growth rate (µ) was calculated as the slope of the linear regression curve within the interval of linear increase of ln-transformed bacterial abundance plotted as a function of time, including at least three data points. The lag phase was defined as the period between inoculation and the observable increase in bacterial cell numbers in the cultures.

Activities of ectoenzymes.
The activities of two extracellular enzymes, leucine aminopeptidase and ß-glucosidase, were measured at the end of the incubation period (381 h for the SKOT and BOG inocula and 216 h for all other treatments). For leucine aminopeptidase activity, L-leucine 7-amido-4-methylcoumarin hydrochloride (Sigma) was dissolved in 5 mM bicarbonate buffer (pH 8), added to 200 µl of a sample from the cultures at a final concentration of 0.2 mM, and incubated at room temperature. At this concentration, the Vmax of the enzyme was measured (50). The release of the fluorescent product 7-amino-4-methylcoumarin (AMC) was measured once per hour for a total period of 10 h, using a MicroWell plate reader connected to a FluroMax-2 instrument (ISA Horiba Group) at a 380-nm excitation wavelength and a 440-nm emission wavelength. Enzyme activities were calculated from the increase in fluorescence for the time interval with a linear increase and converted into AMC concentrations with the help of a standard curve. ß-Glucosidase activity was measured following a similar procedure. 4-Methylumbelliferyl-ß-D-glucopyranoside (Sigma) was dissolved in methylcellosolve (high-performance liquid chromatography grade; Sigma) and added to 200-µl subsamples at a final concentration of 0.2 mM. The release of the fluorescent product 4-methylumbelliferyl was followed for 10 h at 365-nm excitation and 445-nm emission, and extracellular enzyme activities were calculated as described above. ß-Glucosidase activity decreased during the first 5 h (including the sterile controls) and started to increase linearly after this, perhaps because the 4-methylumbelliferyl was not completely dissolved when samples were added. ß-Glucosidase activities were calculated for the period of linear increase after 5 h of incubation, but the results should generally be treated with care.

[14C]benzoic acid uptake.
The utilization of radioactively labeled benzoic acid was measured at the end of the incubation period (381 h for the SKOT and BOG inocula and 216 h for all other treatments). Triplicate subsamples of 5 ml plus a blank where bacteria were killed by the addition of formaldehyde were incubated in test tubes sealed with rubber septum plugs at 20°C for 6 h with 14C-ring benzoic acid (8.0 mCi mmol–1; Sigma) at a final concentration of 1.25 µM. The incubation was terminated by adding 250 µl of formaldehyde through the rubber septum with a syringe. 14CO2 was driven out of the samples by adding 250 µl of 4 M HCl and purging with air for 3 min, and the 14CO2 was collected in 2.5 ml of Carbo-Sorb S (Packard BioScience) in a scintillation vial, as described by Tranvik (52). After that, 2.5 ml of Permafluor E+ (Perkin-Elmer) was added, and radioactivity was counted in a Packard Tri-Carb 2100TR liquid scintillation counter. Particulate 14C was collected on 0.2-µm cellulose nitrate filters (Sartorius, Göttingen, Germany) that were presoaked in benzoic acid solution (1 g liter–1), and the filters were subsequently rinsed five times with 2 ml of the same benzoic acid solution. Radioactivity was assessed by liquid scintillation counting after the addition of 3.5 ml Filtercount (Packard BioScience). Total uptake was calculated as the sum of both parameters.

Statistical analysis of functional parameters.
Differences in functional parameters depending on the inoculum source were tested using one-way multivariate analysis of variance (MANOVA; Pillai's trace test), and a subsequent Tukey’s honestly significant difference (HSD) test was performed on results from univariate ANOVAs for each single functional parameter. Two different MANOVAs were performed: in the first analysis data from all treatments were included, whereas in the second one data from three treatments (LJUS, SKOT, and BOG) were excluded because they showed retarded growth and differed clearly from the rest. Hence, a new MANOVA was run including only the five treatments that were similar in their overall growth patterns. Principal component analysis (PCA) was performed on a correlation matrix of normalized data. Log transformation was generally used to normalize data, but values for BGE were arcsine square-root transformed. All statistical analyses were done using Statistica 6.0 (StatSoft, Inc.).

DNA extraction and terminal restriction fragment polymorphism length analysis.
At the end of the experiment, community compositions were analyzed by T-RFLP analysis as described previously (40). Briefly, DNAs from samples of 4.5 ml, collected by centrifugation, were extracted using a DNeasy kit (QIAGEN, Hilden, Germany), and PCRs were set up and performed as described earlier (40), with minor modifications. Three replicate PCR mixtures for each sample were pooled, treated with mung bean nuclease (14), and purified and concentrated using a QIAquick PCR purification kit (QIAGEN, Hilden, Germany). PCR products (~75 ng) were digested using the restriction enzymes HhaI, HaeIII, and AluI (Invitrogen, Carlsbad, Calif.), and hexachlorofluorescein-labeled fragments were separated and detected with an ABI 3700 96-capillary sequencer running in GeneScan mode (Applied Biosystems). T-RFLP electropherograms were inspected with the free software GenScan View 4 (CRIBI group [http://www.bmr.cribi.unipd.it]). The T-RFLP patterns obtained from the three different enzymes were pooled for each sample and further processed using a program in Visual Basic for Microsoft Excel (51). The data set for subsequent multivariate statistical analysis was constructed taking into account peaks with a size of 40 to 520 base pairs and a relative peak height of >1% of the total signal, and peaks that were fewer than 0.5 bases from a larger peak were merged. To account for small differences in running time among samples, peaks with <0.5-bp differences were considered to be of the same length. These settings were obtained by analyzing five replicates of the same sample, for which independent DNA extractions, PCRs, and restriction digests were made.

T-RFLP patterns were compared through calculation of a similarity matrix, using the Jaccard similarity coefficient, which calculates the similarity between a pair of samples based on binary data as follows: Sjxy = nxy/(nxy + {Delta}xy), in which nxy is the number of peaks common to both samples x and y and {Delta}xy is the number of peaks found only in x or in y. The matrix was calculated using the programs LecPCR and DistAFLP, which are part of the ADE-4 package (http://pbil.univ-lyon1.fr/ADE-4/microb/), and was analyzed by using the UPGMA (unweighted-pair group method using average linkages) clustering algorithm of Statistica 6.0 (StatSoft, Inc.).


arrow
RESULTS
 
Bacterial community structure.
The bacterial communities in the eight different habitats from which the source communities were obtained were clearly differently composed, and Jaccard similarities were generally <0.3 (Fig. 1A). Bacterial communities emerging in the cultures with the eight different inocula were also clearly different (Fig. 1B), and Jaccard similarities were again 0.3 or lower in all cases. Therefore, dissimilarities among the communities that developed from different inocula in the same medium were as great as the dissimilarities among the ambient communities at the original sampling sites (Fig. 1A and B), i.e., bacterial communities did not become more similar to each other upon growth under identical conditions. Among the ambient sampling sites, the lakes (ECKA, LJUS, VATT, and SKOT) and LUMP shared the most peaks, whereas the rock pool community (ROCK), the groundwater community (GROW), and the bog pool community (BOG) were clearly different (Fig. 1A). This pattern was only partly maintained in the cultures. The most obvious deviation was that, apart from the GROW community, bacteria originating from acidic sampling sites (LJUS, SKOT, and BOG) were most distinct in their compositions from the rest of the communities (Fig. 1B). The similarity between communities growing in the cultures and those in the ambient sampling sites was generally low (Jaccard similarity, 0.15 ± 0.07; n = 30), and there was a clear reduction in peak numbers due to cultivation. Hence, peaks were observed in 35 ± 8 (n = 30) positions for the cultures, in contrast to 67 ± 24 (n = 8) positions for the original samples when pooling the three enzymes and using a detection threshold of 1%. For the culture experiment, among-replicate variability was considerable and was highest for the treatments LJUS, SKOT, and BOG (Fig. 1B).


Figure 1
View larger version (16K):
[in this window]
[in a new window]
  		FIG. 1. UPGMA analysis based on T-RFLP patterns obtained from eight ambient sampling sites (A) and the final stages of the culture experiment (B). See Table 1 for assignments of treatment abbreviations to the different habitats from which the source communities were obtained. For the culture experiment, three or four replicates were included for each treatment.

Functional parameters.
The bacteria in the majority of the incubations (ROCK, GROW, ECKA, LUMP, and VATT) developed following typical batch culture growth curves consisting of an initial lag phase, an exponential growth phase, and a stationary phase (Fig. 2). Cultures inoculated with bacteria from the remaining habitats (LJUS, SKOT, and BOG) showed deviating patterns. The most apparent difference in the growth patterns of these last treatments was the conspicuously longer lag phase (Fig. 2). Whereas the bacteria in the ROCK, GROW, ECKA, LUMP, and VATT samples started to grow within 24 to 48 h, it took 123 ± 23 h for LJUS, 182 ± 45 h for SKOT, and 328 ± 10 h for BOG before detectable growth was observed. Moreover, in these cultures there was considerable variability among the growth patterns of replicate cultures since growth started at different times after inoculation. The bacterial communities developing from LJUS, SKOT, and BOG had not entered stationary growth phase when the experiment was stopped (Fig. 2). The length of the lag phase, i.e., the time it took for the different inocula before an increase in cell numbers was observable, seemed to be related to the pH at the ambient sampling site, i.e., the lower the pH, the longer the lag phase (Fig. 3).


Figure 2
View larger version (23K):
[in this window]
[in a new window]
  		FIG. 2. Bacterial abundances in batch cultures inoculated with different starting communities. Note the different scaling of the x axis for SKOT and BOG curves. All symbols represent mean values ± standard errors (SE) (n = 4).


Figure 3
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 3. Length of lag phase (adaptation time) is dependent on the pH at the sampling sites from which the inocula were retrieved. All symbols represent mean values ± standard deviations (n = 4).

PCA including all functional parameters showed a clear separation among the different treatments (Fig. 4), with the slowly growing treatments (LJUS, SKOT, and BOG) being clearly separated from the other treatments along the first principal component. There was a significant effect of the source of the inoculum on the functional performance of bacterial communities growing in the cultures (one-way MANOVA; F70,140 = 3.09; P < 0.00001) when all eight inocula were included in the analysis. Univariate ANOVAs showed that all functional parameters except the biomass yield differed significantly between cultures that had received different inocula (data not shown). When LJUS, SKOT, and BOG (i.e., the treatments showing retarded growth) were excluded from the analysis (see Materials and Methods for details), there was no significant inoculum effect on biomass yield, respiration, and bacterial growth efficiency (Table 2; Fig. 5). However, there was a significant effect of the source of the inoculum on the maximum intrinsic growth rate as well as on ectoenzyme activities and benzoic acid utilization (Tables 2 and 3).


Figure 4
View larger version (15K):
[in this window]
[in a new window]
  		FIG. 4. PCA with functional parameters. Parameters included in the analysis were bacterial abundance (abund), biomass yield (YB), bacterial respiration (BR), bacterial growth efficiency (BGE), length of lag phase (lag), maximum intrinsic growth rate (µ), beta-glucosidase activity (beta), amino-leucine peptidase activity (amino), and benzoic acid uptake (benz). All values were log(x + 1) transformed, except for µ (untransformed) and BGE, which was arcsine square-root transformed. (a) Distribution of treatments (mean ± SE; n = 4) along the first two principal components. (b) Loadings of parameters included in the analysis.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Results from one-way ANOVA to test the effect of the origin of the bacterial inoculum on functional parametersa


Figure 5
View larger version (34K):
[in this window]
[in a new window]
  		FIG. 5. Functional parameters measured during or at the end of the experiments. Only results for ROCK, GROW, ECKA, LUMP, and VATT are shown. YB, bacterial biomass yield; BR, bacterial respiration; BGE, bacterial growth efficiency; µ, maximum intrinsic growth rate. All bars represent mean values ± SE calculated from four replicate cultures. A post hoc test revealed no significant differences between treatments for YB, BR, and BGE. For VATT, µ was significantly lower than that for LUMP.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Activities of ß-glucosidase, leucine-aminopeptidase, and 14C-ring benzoic acid uptake in ROCK, GROW, ECKA, LUMP, and VATTa


arrow
DISCUSSION
 
Development of BCC depending on source community and environmental conditions.
The first aim of this study was to test whether identical environmental conditions would lead to the selection of bacterial communities that were similar in structure and function. The different inocula resulted in different communities of heterotrophic bacteria under identical conditions. Thus, the history and distribution of taxa within the source community (i.e., the inoculum) were important factors regulating the composition of the selected communities, whereas environmental conditions (i.e., the medium) seemed to be less important. This agrees with previous studies suggesting that both environmental conditions and source community determine BCC (38, 40). If certain bacterial taxa selected from a single ubiquitous species pool were superior for coping with the conditions in the cultures, communities emerging from different source communities under identical conditions would show larger similarities to each other than those between the communities at the original sampling sites, where environmental conditions were dissimilar and thus selected for differently composed communities. This was not the case, since in both cases the similarity between different treatments or habitats was 0.3 or lower (Fig. 1).

Previous experiments provided convincing support of the "everything is everywhere—the environment selects" concept for eukaryotic microbes (17). These studies included a range of enrichment strategies and incubations extended up to roughly 4 months in order to provoke the growth of very rare cryptic species. In contrast, our experiments lasted 242 to 505 h. Based on the assumption that rare or cryptic populations have the potential to become dominant under the right circumstances, a longer incubation time would possibly also reflect a higher degree of ubiquity for the bacteria in our experiment, if ubiquitous species in our inocula were rare and needed extensive enrichment to reach detectable quantities. It is possible to gauge the hypothetical time (t) it takes for such rare populations to reach 1% of the total bacterial concentration (assumed to be 109 cells liter–1), i.e., a fraction of the total community that is likely to be easily detected by T-RFLP. This is done by resolving the exponential growth equation for t, i.e., Nt = Ni x eµt, where µ is the exponential growth rate, Ni is the initial abundance, and Nt is the number of cells at which the population becomes detectable. Thus, for a rare population with an initial abundance of 100 cells liter–1 in a culture growing with a doubling time of 11 h (corresponding to the average community growth rate found in our experiment), it would take roughly 180 h for the population to reach a detectable size of 107 cells liter–1. The actual incubation times in the experiment were 242 to 505 h, suggesting that rare populations would in fact be able to reach detectable cell densities unless they were growing substantially slower than the average community growth rate. In general terms, the calculations above illustrate that the successful proliferation of rare or cryptic species enabling them to reach high levels of abundance requires substantial time under suitable conditions. Hence, their window of opportunity may be restricted to environments with stable environmental conditions at scales that are relevant for bacteria. Hence, in frequently disturbed communities such as pelagic environments, time might simply not be sufficient for cryptic species to become dominant members of the community. Thus, the composition of a local bacterial community should not exclusively be regulated either by selection by the local environmental condition or by "metacommunity processes." In contrast, the two mechanisms are not mutually exclusive and occur simultaneously but apply to different time scales, and their relative importance might vary between habitats differing with regard to their disturbance regimen (5, 41).

At a rough glance, the growth of differently composed communities under identical conditions seems to contradict the previously suggested global distribution of freshwater bacteria (25, 29, 30, 55, 56). However, due to the lack of information about the identity of community members enriched in the culture experiments, our results cannot be used to deduce that freshwater planktonic bacteria are not globally dispersed. Even though communities seemed to be different from each other depending on the origin of the source community, they might still contain highly related, though different, representatives belonging to the same bacterial cluster. Accordingly, studies of sediments and soils suggest that the degree of ubiquity of microbial taxa depends on the level of phylogenetic resolution (6, 33). Additionally, it has also been shown that bacteria with identical 16S rRNA gene sequences (which by operational definition would belong to the same species) can have clearly different ecophysiological and phenotypic traits (30, 35). Hence, even if there is a global dispersal of bacterial species, the local adaptation of genotypes within these species might lead to various responses by the same species (as defined by 16S rRNA sequences) to changes in environmental conditions.

There was some resemblance between the clustering patterns between culture communities and ambient sites. The most apparent was that the GROW (groundwater) and BOG (bog pool) samples had the highest fractions of unique peaks in the original samples as well as in the cultures. Groundwater might be unique because it lacks direct contact with airborne bacteria, and its connectivity to other habitats is very limited. The bog pool provides hostile conditions (pH < 4) for many bacteria and might therefore also harbor a highly adapted community. Hence, GROW and BOG cultures developed clearly distinct communities because they were inoculated with clearly distinct source communities. More unexpected was the observation that source communities from two lakes (LJUS and SKOT) also resulted in the growth of quite distinct bacterial communities (Fig. 1B). It appeared, however, that pH was generally important in regulating the overall growth patterns of selected bacterial communities (Fig. 2). Hence, all bacterial source communities from acidic sites (LJUS, SKOT, and BOG) had extended lag phases of several days (in the case of LJUS and SKOT; Fig. 2) or even weeks (BOG) before growth could be observed. Possibly, a long apparent lag phase is due to a "shock response" of bacteria exposed to strong deviations in pH, resulting in either a very long adaptation time or a very small effective inoculum of specific bacteria that are able to grow under the culture conditions.

Although the variability among replicate cultures with the same inoculum was always less than the variability between cultures with different inocula (Fig. 1B), there was a high fraction (30% or more) of peaks that were unique to each replicate culture. Recently, Hewson and Fuhrman (31) could show for another fingerprinting method (automated ribosomal intergenic spacer analysis) that it is highly reproducible as long as only the largest peaks, i.e., those comprising >0.5% of the total signal, are taken into account. Assuming that the same accounts for T-RFLP analysis, it seems likely that the observed among-replicate variability in BCC resembles real differences in taxon composition and is not a methodological artifact. The size of the source community (i.e., the total number of individuals), its diversity, and the distribution of individuals among taxa are crucial for community assemblage patterns (11). Highly similar BCC among replicate cultures requires that replicate inocula (i.e., the replicate 10-ml subsamples of the source community) also show a high degree of similarity. Deviating patterns in BCC among replicate cultures might thus simply reflect that the inocula were not identical. Long and Azam (43) demonstrated that variations in BCC are minor from the microliter to the liter scale, but this accounted only for the most abundant community members. Considering that some of the taxa that grew in the cultures may have been rare in the source community, it seems likely that differences in the presence or absence of rare species and the absolute abundance of taxa caused the observed variability in community assembly patterns among replicate cultures. Together with size-selective removal of cells during filtration and cultivation-induced shifts in community structure and diversity (16, 44), the random exclusion of rare taxa from the inocula also might have limited the potential of the different source communities to converge in BCC upon growth under identical conditions.

Coupling between community structure and function.
The second major aim of this study was to test if there is a link between BCC and the functioning of the system. The pH-affected communities (LJUS, SKOT, and BOG) were functionally clearly different from those within the remaining treatment groups, i.e., ROCK, GROW, ECKA, LUMP, and VATT (Fig. 2 and 4). Similar effects of pH on bacterial community composition and functioning have been observed previously (40, 47) and agree with findings from field studies suggesting that pH is a strong regulating factor of BCC in lakes (42, 54) and a major threat to the survival of allochthonous bacteria in aquatic systems (2). On the other hand, broad-scale functions such as bacterial biomass production and respiration were similar in treatments ROCK, GROW, ECKA, LUMP, and VATT, independently of BCC, suggesting functional redundancy. This agrees with the results of a previous full-factorial switch experiment (40), where we investigated the importance of the source of the growth medium versus the inoculum as shaping factors of community composition and broad-scale functioning and found a loose connection between community composition and function. Other culture studies also found that functioning was rather independent of the origins of inocula (21, 22, 53).

On the other hand, there are studies showing that community structure matters for function (4, 37, 38). It is interesting to observe, however, that these studies mostly measured "narrow" functions, such as pesticide-induced or semilabile DOC degradation and enzyme activities, while studies supporting that function is independent of BCC typically address "broad" functions, such as respiration and biomass yield (e.g., see reference 40). In this study, different bacterial communities showed at least partly differing ectoenzyme activities (Table 3) and various potentials to utilize benzoic acid (Table 3), which was used as a model for aromatic ring structures. This indicates that different components within the total DOM pool were degraded in cultures with different inocula. Thus, specific functions such as enzyme activities differed in treatments with different BCC, whereas there were no observable effects on broader functions such as biomass production and community respiration. This idea is compatible with the finding of functional equivalency of different fungal species with respect to carbon mineralization (49) and that well-defined narrow niche functions are more sensitive to a reduction in diversity (27) or perturbation (23) than broad-scale functions carried out by a wide range of organisms.

To summarize, differently composed bacterial communities developed from different source communities under identical conditions. These structural differences, however, were only partly reflected in differences in community function, and the functions that appeared sensitive to community composition were relatively specific, while aggregated functions (e.g., community biomass and respiration) were insensitive to community composition. Very similar findings have been made for zooplankton communities (36), suggesting that the overall mechanisms behind community structure and function might be similar for microbes and higher organisms.


arrow
ACKNOWLEDGMENTS
 
We thank Jan Johansson and Kristiina Nygren for their assistance and help during field work and implementation of the experiment. We are grateful to Eddie von Wachenfeldt and Therese Carlsson for help with the collection of the water samples. We greatly appreciate help from Jan Bengtsson, Kalle Mälson, and Allan Rohde with the study site selection. We thank Helmut Hillebrand, Paul del Giorgio, Ingvar Sundh, and several members of the Microbial Ecology group at our department for fruitful discussions and constructive comments on earlier drafts of the manuscript. Constructive comments by an anonymous reviewer also greatly improved the manuscript.

This work was supported by grants from the Swedish Research Council to L.J.T. and the Helge Ax:son Johnson and Malméns foundations to S.L.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Limnology/Department of Ecology and Evolution, Evolutionary Biology Centre, Uppsala University, Norbyvägen 20, 75236 Uppsala, Sweden. Phone: 46 18 471 2726. Fax: 46 18 531134. E-mail: silke.langenheder{at}ebc.uu.se Back


arrow
REFERENCES
 
    1
  1. Baas-Becking, L. G. M. 1934. Geologie of Inleidning tot de Milieukunde. W.P. Van Stokum, The Hague, The Netherlands.
    2. 2
  2. Barcina, I., P. Lebaron, and J. Vives-Rego. 1997. Survival of allochthonous bacteria in aquatic systems: a biological approach. FEMS Microbiol. Ecol. 23:1-9.
    3. 3
  3. Bastviken, D., L. Persson, G. Odham, and L. J. Tranvik. 2004. Degradation of dissolved organic mater in oxic and anoxic lake water. Limnol. Oceanogr. 49:109-116.
    4. 4
  4. Carlson, C. A., S. J. Giovannoni, D. A. Hansell, S. J. Goldberg, R. Parsons, and K. Vergin. 2004. Interactions among dissolved organic carbon, microbial processes, and community structure in the mesopelagic zone of the northwestern Sargasso Sea. Limnol. Oceanogr. 49:1073-1083.
    5. 5
  5. Chase, J. M. 2003. Community assembly: when should history matter? Oecologia 136:489-498.[CrossRef][Medline]
    6. 6
  6. Cho, J. C., and J. M. Tiedje. 2000. Biogeography and degree of endemicity of fluorescent Pseudomonas strains in soil. Appl. Environ. Microbiol. 66:5448-5456.[Abstract/Free Full Text]
    7. 7
  7. Cottenie, K., and L. De Meester. 2004. Metacommunity structure: synergy of biotic interactions as selective agents and dispersal as fuel. Ecology 85:114-119.
    8. 8
  8. Cottenie, K., E. Michels, N. Nuytten, and L. De Meester. 2003. Zooplankton metacommunity structure: regional versus local processes in highly interconnected ponds. Ecology 84:991-1000.
    9. 9
  9. Cottrell, M. T., and D. L. Kirchman. 2000. Natural assemblages of marine proteobacteria and members of the Cytophaga-Flavobacter cluster consuming low- and high-molecular-weight dissolved organic matter. Appl. Environ. Microbiol. 66:1692-1697.[Abstract/Free Full Text]
    10. 10
  10. Covert, J. S., and M. A. Moran. 2001. Molecular characterization of estuarine bacterial communities that use high- and low-molecular weight fractions of dissolved organic carbon. Aquat. Microb. Ecol. 25:127-139.
    11. 11
  11. Curtis, T. P., and W. T. Sloan. 2004. Prokaryotic diversity and its limits: microbial community structure in nature and implications for microbial ecology. Curr. Opin. Biotechnol. 7:221-226.
    12. 12
  12. del Giorgio, P. A., and J. J. Cole. 1998. Bacterial growth efficiency in natural aquatic systems. Annu. Rev. Ecol. Syst. 29:503-541.[CrossRef]
    13. 13
  13. del Giorgio, P. A., J. M. Gasol, D. Vaque, P. Mura, S. Agusti, and C. M. Duarte. 1996. Bacterioplankton community structure: protists control net production and the proportion of active bacteria in a coastal marine community. Limnol. Oceanogr. 41:1169-1179.
    14. 14
  14. Egert, M., and M. W. Friedrich. 2003. Formation of pseudo-terminal restriction fragments, a PCR-related bias affecting terminal restriction fragment length polymorphism analysis of microbial community structure. Appl. Environ. Microbiol. 69:2555-2562.[Abstract/Free Full Text]
    15. 15
  15. Eiler, A., S. Langenheder, S. Bertilsson, and L. J. Tranvik. 2003. Heterotrophic bacterial growth efficiency and community structure at different natural organic carbon concentrations. Appl. Environ. Microbiol. 69:3701-3709.[Abstract/Free Full Text]
    16. 16
  16. Eilers, H., J. Pernthaler, F. O. Glockner, and R. Amann. 2000. Culturability and in situ abundance of pelagic bacteria from the North Sea. Appl. Environ. Microbiol. 66:3044-3051.[Abstract/Free Full Text]
    17. 17
  17. Fenchel, T., G. F. Esteban, and B. J. Finlay. 1997. Local versus global diversity of microorganisms: cryptic diversity of ciliated protozoa. Oikos 80:220-225.[CrossRef]
    18. 18
  18. Fenchel, T., and B. J. Finlay. 2004. The ubiquity of small species: patterns of local and global diversity. BioScience 54:777-784.[CrossRef]
    19. 19
  19. Findlay, S. E. G., R. L. Sinsabaugh, W. V. Sobczak, and M. Hoostal. 2003. Metabolic and structural responses of hyporheic microbial communities to variations in supply of dissolved organic matter. Limnol. Oceanogr. 48:1608-1617.
    20. 20
  20. Finlay, B. J., and K. J. Clarke. 1999. Ubiquitous dispersal of microbial species. Nature 400:828.[CrossRef]
    21. 21
  21. Fuchs, B. M., M. V. Zubkov, K. Sahm, P. H. Burkill, and R. Amann. 2000. Changes in community composition during dilution cultures of marine bacterioplankton as assessed by flow cytometric and molecular biological techniques. Environ. Microbiol. 2:191-201.[CrossRef][Medline]
    22. 22
  22. Gasol, J. M., M. Comerma, J. M. Garcia, J. Armengol, E. O. Casamayor, P. Kojecka, and K. Simek. 2002. A transplant experiment to identify the factors controlling bacterial abundance, activity, production, and community composition in a eutrophic canyon-shaped reservoir. Limnol. Oceanogr. 47:62-77.
    23. 23
  23. Girvan, M. S., C. D. Campbell, K. Killham, J. I. Prosser, and L. A. Glover. 2005. Bacterial diversity promotes community stability and functional resilience after perturbation. Environ. Microbiol. 7:301-313.[CrossRef][Medline]
    24. 24
  24. Glöckner, F. O., B. M. Fuchs, and R. Amann. 1999. Bacterioplankton compositions of lakes and oceans: a first comparison based on fluorescence in situ hybridization. Appl. Environ. Microbiol. 65:3721-3726.[Abstract/Free Full Text]
    25. 25
  25. Glöckner, F. O., E. Zaichikov, N. Belkova, L. Denissova, J. Pernthaler, A. Pernthaler, and R. Amann. 2000. Comparative 16S rRNA analysis of lake bacterioplankton reveals globally distributed phylogenetic clusters including an abundant group of actinobacteria. Appl. Environ. Microbiol. 66:5053-5065.[Abstract/Free Full Text]
    26. 26
  26. Green, J. L., A. J. Holmes, M. Westoby, I. Olicer, D. Briscoe, M. Dangerfield, M. Gillings, and A. J. Beattie. 2004. Spatial scaling of microbial eukaryote diversity. Nature 432:747-750.[CrossRef][Medline]
    27. 27
  27. Griffiths, B. S., K. Ritz, R. D. Bardgett, R. Cook, S. Christensen, F. Ekelund, S. J. Sorensen, E. Baath, J. Bloem, P. C. de Ruiter, J. Dolfing, and B. Nicolardot. 2000. Ecosystem response of pasture soil communities to fumigation-induced microbial diversity reductions: an examination of the biodiversity-ecosystem function relationship. Oikos 90:279-294.[CrossRef]
    28. 28
  28. Groffman, P. M., and P. J. Bohlen. 1999. Soil and sediment biodiversity. Cross-system comparisons and large-scale effects. BioScience 49:139-148.[CrossRef]
    29. 29
  29. Hahn, M. W. 2003. Isolation of strains belonging to the cosmopolitan Polynucleobacter necessarius cluster from freshwater habitats located in three climatic zones. Appl. Environ. Microbiol. 69:5248-5254.[Abstract/Free Full Text]
    30. 30
  30. Hahn, M. W., H. Lunsdorf, Q. L. Wu, M. Schauer, M. G. Höfle, J. Boenigk, and P. Stadler. 2003. Isolation of novel ultramicrobacteria classified as actinobacteria from five freshwater habitats in Europe and Asia. Appl. Environ. Microbiol. 69:1442-1451.[Abstract/Free Full Text]
    31. 31
  31. Hewson, I., and J. A. Fuhrman. 2004. Richness and diversity of bacterioplankton species along an estuarine gradient in Moreton Bay, Australia. Appl. Environ. Microbiol. 70:3425-3433.[Abstract/Free Full Text]
    32. 32
  32. Horner-Devine, M. C., K. M. Carney, and B. J. M. Bohannan. 2004. An ecological perspective on bacterial biodiversity. Proc. R. Soc. Lond. B 271:113-122.[Medline]
    33. 33
  33. Horner-Devine, M. C., M. Lage, J. B. Hughes, and B. J. M. Bohannan. 2004. A taxa-area relationship for bacteria. Nature 432:750-753.[CrossRef][Medline]
    34. 34
  34. Hubbell, S. P. 2001. A unified neutral theory of biodiversity and biogeography. Princeton University Press, Princeton, N.J.
    35. 35
  35. Jaspers, E., and J. Overmann. 2004. Ecological significance of microdiversity: identical 16S rRNA gene sequences can be found in bacteria with highly divergent genomes and ecophysiologies. Appl. Environ. Microbiol. 70:4831-4839.[Abstract/Free Full Text]
    36. 36
  36. Jenkins, D. G., and A. L. J. Buikema. 1998. Do similar communities develop in similar sites? A test with zooplankton structure and function. Ecol. Monogr. 68:421-443.[CrossRef]
    37. 37
  37. Johnson, A., N. Lewellyn, J. Smith, C. van der Gast, A. Lilley, A. Singer, and I. Thompson. 2004. The role of microbial community composition and groundwater chemistry in determining isoproturon degradation potential in UK aquifers. FEMS Microbiol. Ecol. 49:71-82.
    38. 38
  38. Kirchman, D. L., A. I. Dittel, S. E. G. Findlay, and D. Fischer. 2004. Changes in bacterial activity and community structure in response to dissolved organic matter in the Hudson River, New York. Aquat. Microb. Ecol. 35:243-257.
    39. 39
  39. Langenheder, S., V. Kisand, J. Wikner, and L. J. Tranvik. 2003. Salinity as a structuring factor for the composition and performance of bacterioplankton degrading riverine DOC. FEMS Microbiol. Ecol. 45:189-202.[CrossRef]
    40. 40
  40. Langenheder, S., E. S. Lindström, and L. J. Tranvik. 2005. Weak coupling between community composition and functioning in aquatic bacteria. Limnol. Oceanogr. 50:957-967.
    41. 41
  41. Lindström, E. S., and A. K. Bergström. 2004. Influence of inlet bacteria on bacterioplankton assemblage composition in lakes of different hydraulic retention time. Limnol. Oceanogr. 49:125-136.
    42. 42
  42. Lindström, E. S., M. P. Kamst-van Agterveld, and G. Zwart. 2005. Distribution of typical freshwater bacterial groups is associated with pH, temperature, and lake water retention time. Appl. Environ. Microbiol. 71:8201-8206.
    43. 43
  43. Long, R. A., and F. Azam. 2001. Microscale patchiness of bacterioplankton assemblage richness in seawater. Aquat. Microb. Ecol. 26:103-113.
    44. 44
  44. Massana, R., C. Pedrós-Alió, E. O. Casamayor, and J. M. Gasol. 2001. Changes in marine bacterioplankton phylogenetic composition during incubations designed to measure biogeochemically significant parameters. Limnol. Oceanogr. 46:1181-1188.
    45. 45
  45. Methe, B. A., W. D. Hiorns, and J. P. Zehr. 1998. Contrasts between marine and freshwater bacterial community composition: analyses of communities in Lake George and six other Adirondack lakes. Limnol. Oceanogr. 43:368-374.
    46. 46
  46. Mille-Lindblom, C., and L. J. Tranvik. 2003. Antagonism between bacteria and fungi on decomposing aquatic plant litter. Microb. Ecol. 45:173-182.[CrossRef][Medline]
    47. 47
  47. Princic, A., I. Mahne, F. Megusar, E. A. Paul, and J. M. Tiedje. 1998. Effects of pH and oxygen and ammonium concentrations on the community structure of nitrifying bacteria from wastewater. Appl. Environ. Microbiol. 64:3584-3590.[Abstract/Free Full Text]
    48. 48
  48. Roland, F., N. F. Caraco, and J. J. Cole. 1999. Rapid and precise determination of dissolved oxygen by spectrophotometry: evaluation of interference from color and turbidity. Limnol. Oceanogr. 44:1148-1154.
    49. 49
  49. Setälä, H., and M. A. McLean. 2004. Decomposition rate of organic substrates in relation to the species diversity of soil saprophytic fungi. Oecologia 139:98-107.[CrossRef][Medline]
    50. 50
  50. Stepanauskas, R., H. Edling, and L. J. Tranvik. 1999. Differential dissolved organic nitrogen availability and bacterial aminopeptidase activity in limnic and marine waters. Microb. Ecol. 38:264-272.[CrossRef][Medline]
    51. 51
  51. Stepanauskas, R., M. A. Moran, B. A. Bergamaschi, and J. T. Hollibaugh. 2003. Covariance of bacterioplankton composition and environmental variables in a temperate delta system. Aquat. Microb. Ecol. 31:85-98.
    52. 52
  52. Tranvik, L. J. 1993. Microbial transformation of labile dissolved organic matter into humic-like matter in seawater. FEMS Microbiol. Ecol. 12:177-183.[CrossRef]
    53. 53
  53. Tranvik, L. J., and M. G. Höfle. 1987. Bacterial growth in mixed cultures on dissolved organic carbon from humic and clear waters. Appl. Environ. Microbiol. 53:482-488.[Abstract/Free Full Text]
    54. 54
  54. Yannarell, A. C., and E. W. Triplett. 2005. Geographic and environmental sources of variation in lake bacterial community composition. Appl. Environ. Microbiol. 71:227-239.[Abstract/Free Full Text]
    55. 55
  55. Zwart, G., B. C. Crump, M. Agterveld, F. Hagen, and S. K. Han. 2002. Typical freshwater bacteria: an analysis of available 16S rRNA gene sequences from plankton of lakes and rivers. Aquat. Microb. Ecol. 28:141-155.
    56. 56
  56. Zwart, G., E. J. van Hannen, M. P. Kamst-van Agterveld, K. Van der Gucht, E. S. Lindström, J. Van Wichelen, T. Lauridsen, B. C. Crump, S. K. Han, and S. Declerck. 2003. Rapid screening for freshwater bacterial groups by using reverse line blot hybridization. Appl. Environ. Microbiol. 69:5875-5883.[Abstract/Free Full Text]

Applied and Environmental Microbiology, January 2006, p. 212-220, Vol. 72, No. 1
0099-2240/06/$08.00+0     doi:10.1128/AEM.72.1.212-220.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.
Agricola
Right arrow Articles by Langenheder, S.
Right arrow Articles by Tranvik, L. J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»