Read Beier, Sara, and Stefan Bertilsson. Uncoupling of chitinase activity and uptake of hydrolysis products in freshwater bacterioplankton. Limnol. Oceanogr., 56(4), 2011, 1179­1188 text version

Limnol. Oceanogr., 56(4), 2011, 1179­1188 2011, by the American Society of Limnology and Oceanography, Inc. doi:10.4319/lo.2011.56.4.1179


Uncoupling of chitinase activity and uptake of hydrolysis products in freshwater bacterioplankton

Sara Beier1 and Stefan Bertilsson*

Department of Ecology & Genetics, Limnology, Uppsala University, Sweden


We investigated to what extent chitinolytic bacteria subsidize bacterial populations that do not produce chitinolytic enzymes but still use the products of chitin hydrolysis. Applying single-cell techniques to untreated and chitin-enriched lake water, we show that the number of planktonic cells taking up chitin hydrolysis products by far exceeds the number of cells expressing chitinases. Flavobacteria, Actinobacteria, and specifically members of the abundant and ubiquitous freshwater Ac1 cluster of the Actinobacteria, increased in abundance and were enriched in response to the chitin amendment. Flavobacteria were frequently observed in dense clusters on chitin particles, suggesting that they are actively involved in the hydrolysis and solubilization of chitin. In contrast, Actinobacteria were exclusively planktonic. We propose that planktonic Actinobacteria contain commensals specialized in the uptake of small hydrolysis products without expressing or possibly even possessing the machinery for chitin hydrolysis. More research is needed to assess the importance of such ``cheater'' substrate acquisition strategies in the turnover and degradation of polymeric organic matter in aquatic ecosystems.

Particulate and dissolved polymeric substances are major components of aquatic organic matter and can dominate the pool of carbon sources readily available for bacterial heterotrophs (Middelboe et al. 1995). To benefit from this potential nutrient and energy pool, microorganisms require exoenzymes acting outside of the cell in order to break down the polymers into smaller subunits that can be transported into cells for subsequent metabolic use. It has previously been shown that microbial hydrolysis of a variety of organic particles may exceed the capacity of the attached microbes to take up the released hydrolysis products (Smith et al. 1992). Such a scenario paves the way for the occurrence of commensals, as the resulting release of hydrolysis products into the surrounding medium may provide free-living bacteria with organic substrates for growth. It has indeed been shown that bacteria tend to cluster in plumes of dissolved organic carbon (DOC) that are released by sinking particles (Grossart 2010). More specifically, enzyme-mediated release of labile DOC has also been demonstrated for chitin particles (Kirchman and White 1999). Chitin is one of the most abundant biopolymers in aquatic systems. It is composed of (1R4)-b-linked N-acetyl-D-glucosamine (NAG) subunits and is an important component of both carbon and nitrogen cycles. The initial step in microbial chitin degradation is usually the chitinase-mediated hydrolysis of the polymer into monomers and oligomers. Our study focuses on the bacterial degradation of chitin, although fungi could also play an important role in chitin degradation in aquatic systems under certain circumstances (Wurzbacher et al. 2010). Experiments with cultured

* Corresponding author: [email protected]

1 Current address: UPMC Univ. Paris 06, UMR 7621, LOMIC, Observatoire Oceanologique, F-66650 Banyuls sur mer, France, ´ and CNRS, UMR 7621, LOMIC, Observatoire Oceanologique, ´ F-66650 Banyuls sur mer, France.

bacterial strains suggest that chitinase activity mainly produces chitin dimers (diNAG), but also the abovementioned chitin monomers (NAG) are being released (Horn et al. 2006). Apart from being a structural component of chitin, NAG also originates from murein in bacterial cell walls (Sharon 1965), whereas diNAG originates exclusively from degradation of chitin molecules. A previous study of a bacterial strain grown on chitin substrates in liquid medium showed that a minority of the cells expressed chitinase activity (Baty et al. 2000), suggesting that the genetically identical subpopulation lacking chitinase activity was feeding on released hydrolysis products. Such an intraspecific cross-feeding system, where only some individuals within a population express the energetically expensive chitinase machinery (Keyhani and Roseman 1999), could result in a more efficient populationlevel use of chitin as a nutrient source. Also, for particleattached microorganisms in natural ecosystems, it can be expected that dividing cells within a clonal population would remain in close proximity to each other even when surrounded by alien species. Such a metabolic strategy, where individual cells release excess hydrolysis products to feed members of its clonal population, would also provide an ecologically meaningful explanation for the overproduction of biopolymer hydrolysis products as outlined above (Smith et al. 1992). Also, commensal populations that do not express chitinases but incorporate released hydrolysis products would be favored by such excess polymer hydrolysis. Accordingly, there are early observations that some isolates obtained from chitin-containing particles can grow on NAG but not on polymeric chitin, and it was suggested that those organisms benefit from NAG produced by their chitinase-producing neighbors (Kaneko and Colwell 1978). It has furthermore been shown that a non-chitinolytic Escherichia coli strain can grow on diNAG as its sole carbon source (Keyhani and Roseman 1997) and more recent experiments on bacterial laboratory strains point to the existence of commensal or even



Beier and Bertilsson

parasitic relationships between chitin degraders and chitin users (Everuss et al. 2008; Jagmann et al. 2010). The ability to take up NAG appears to be widespread in aquatic bacteria (Nedoma et al. 1994; Riemann and Azam 2002). However, there is so far no evidence for organismlevel separation of NAG uptake and chitinolytic activity beyond the culturable minority of natural microbial communities. Hence, the quantitative importance of organisms using polymer hydrolysis products without producing the hydrolytic enzymes to generate these degradation intermediates is not known. In this study we quantified the ratio between chitinase-positive cells relative to those incorporating chitin hydrolysis products, by applying cultivation-independent single-cell methods. Our aim was to determine if the separation of chitin hydrolysis and substrate uptake is a significant heterotrophic growth strategy in freshwater systems. We used analogous methods to identify bacterial groups likely occupying the ecological niche as commensals, with a focus on quantitatively dominant taxa.

Fig. 1. Schematic illustration of the experimental setup: (A) experiment 1, (B) experiment 2.


Experimental setup--Pelagic surface water was sampled in mesotrophic Lake Erken (59u259N, 18u159E) and eutrophic Lake Ekoln (59u459N, 17u369E) on 31 October 2008 and 03 November 2008, respectively. Subsamples of lake water were enriched with 15 mg of colloidal chitin (dry weight) L21 for subsequent chitin-enrichment incubation experiments paralleled by non-chitin-amended control incubations. The remaining water was used directly after sampling for initial analyses of substrate uptake rates and bacterial community analyses. The added amount of chitin is within the range of total suspended organic matter previously observed in nutrient-rich surface waters (Lindstrom et al. 1999). ¨ Chitin-enriched water samples and non-enriched controls were incubated in 1-liter glass bottles on a rotary shaker. Incubations were maintained at 20uC in a 12 : 12 light : dark cycle (daylight, Osram 36W-12-950) in order to approximate diurnal solar radiation fluctuations. The progression of chitin degradation was assessed every 24­ 48 h by absorbance at 600 nm (Lambda-40 ultraviolet­ visible spectrophotometer; Perkin Elmer), using the nonenriched control incubations as a reference. In tests we could show that these absorbance readings were linearly related to the amount of suspended colloidal chitin (r2 5 0.99). After the values for absorbance reached approximately one-third of the initial value, the chitin-enriched incubations from Erken and Ekoln (9 and 10 d of incubation, respectively) were stopped. Samples were taken for substrate uptake rates and single-cell analyses, including catalyzed reporter deposition­fluorescence in situ hybridization (CARD-FISH), microautoradiography (MAR), and single-cell chitinase activity as detailed below. The same analyses were performed on the initial lake water samples immediately after sampling. The control treatments were sampled after 14 d (Erken) and 11 d (Ekoln) for CARD-FISH analysis (experiment [expt] 1, Fig. 1A). Experimental addition of colloidal chitin ensures a higher degree of chitin purity because associated proteins are removed by the harsh conditions used to break up crystalline structures as described below. Addition of noncrystalline, colloidal chitin also ensures that a broad and diverse chitinolytic community has the ability to respond to the biopolymer amendment; i.e., enzymemediated depolymerization of different forms of crystalline chitin often requires additional and contrasting enzymes (Svitil et al. 1997). Accordingly, a number of organisms grow selectively on either a-crystalline chitin or b-crystalline chitin (Ramaiah et al. 2000). Lake Erken water was sampled a second time on 13 March 2010 for a follow-up experiment to test if Actinobacteria and especially members of the AcI cluster that did not attach to colloidal chitin would instead attach to naturally abundant crystalline forms of chitin. The acrystalline form of chitin present in the carapace of zooplankton is likely the major chitin source in freshwater ecosystems. We therefore incubated freshly sampled lake water with 15 mg (dry weight) L21 particles derived from a Daphnia magna culture (daphnia enrichment). 1.4 liters of lake water with the added daphnia particles was maintained in a 2-liter glass bottle on a rotary shaker. Incubations were maintained at 15uC in the dark and sampled after 3, 5, and 8 d for CARD-FISH analyses (experiment [expt] 2, Fig. 1B). Preparation of colloidal chitin--Ten grams of powdered crab-shell chitin (Sigma-Aldrich) was incubated for , 12 h at 4uC with 100 mL of 85% phosphoric acid. Subsequently, 1 liter of tap water was added to the chitin and stirred for 24 h to obtain colloidal chitin particles. Next, the chitin was subjected to repeated centrifugation and rinsing with tap water until the pH was stable and similar to the original tap water. A final rinse with Milli-Q water was performed before the resulting colloidal chitin was sterilized by autoclaving.

Chitin use in bacterioplankton Preparation of daphnia particles--The daphnia were freeze-dried and disrupted mechanically by crushing them with a pestle. The powder obtained was suspended in MilliQ water and sequentially washed three times with alternating centrifugation and supernatant aspiration to keep only the particulate material. The chitin content of intact daphnia has been estimated at 7%, whereas the chitin content of the carapaces has been estimated at 18% (Cauchie et al. 2002). The chitin content of the prepared particulate daphnia material should fall within this range. In order to keep the surface structure of the zooplankton particles as natural as possible, they were not sterilized prior to the incubation. Microscopic inspection of the intact, freeze-dried daphnia indicated the presence of few bacteria attached to their carapace surface, demonstrating that the predominant part of the particles' surface area was available for de novo colonization. Incorporation rates--Community-level incorporation rates of the chitin hydrolysis products NAG and diNAG were measured along with glucose (Glc) included as a reference substance. Sample aliquots were incubated at near in situ temperature (10uC for initial lake water, 20uC for chitin-enrichment incubations) with 10 nmol L21 of the respective substrate (D-[U-14C]glucose, 317 mCi mmol21, batch B59; N-acetyl-D-[1-14C]glucosamine, 55 mCi mmol21, batch B219 A; [3H]diacetylchitobiose, 14 Ci mmol21, batch B1; General Electric Healthcare). Incubations were stopped after , 1 h by adding formaldehyde to 2% final concentration. Cells were captured on SuporH-200 membrane filters (0.2-mm pore size) by gentle vacuum filtration and washed three times with Q-grade water. Individual filters were placed in scintillation vials and overlaid with 5 mL of scintillation cocktail (Optiphase ``Hisafe'' 2; Perkin Elmer), and radioactivity was measured by liquid scintillation counting (Tri-Carb 2100tr; Packard-PerkinElmer). Two replicates and one killed control were used to estimate incorporation rates of the respective substrate in each incubation bottle. Incubations for single-cell analyses--Three replicates and one control, killed with paraformaldehyde added to a final concentration of 2% before addition of substrates, were incubated with 10 nmol L21 3H-labeled NAG (N-acetyl-D[1-3H]-glucosamine, 9 Ci mmol21, batch 57 A, GEHealthcare), diNAG ([ 3 H]diacetylchitobiose, 14 Ci mmol21, batch B1; General Electric Healthcare) or Glc (D-[6-3H]glucose, 24 Ci mmol21, batch 168; General Electric Healthcare), respectively. Initial water samples from Lakes Erken and Ekoln were incubated at 10uC and stopped after a 5-h period incubation by adding paraformaldehyde to 2% final concentration. Samples from respective chitin enrichments were incubated at 20uC for 2 h before stopping the incubation with paraformaldehyde. Samples were fixed for 2 h at room temperature and cells were subsequently collected on white polycarbonate filters (0.2-mm pore size). In order to test reproducibility, one filter was collected for each of the replicate incubation bottles from all four abovedescribed samples. The remaining water from the three


replicate incubation bottles for each sample was pooled prior to collection of cells by filtration. Filters were stored at 220uC until further analyses by CARD-FISH and MAR. Incubations for single-cell chitinase-activity combined with MAR were performed as described above with radioactively labeled substrates (NAG, diNAG). One hour before reactions were stopped, 10 mmol L21 ELFH97 chitinase-N-acetylglucosaminidase substrate (Invitrogen) (ELFH97-NAG) dissolved in dimethylsulfoxide was added to a final concentration of 10 mmol L21. Incubations were terminated by 10-min fixation with 0.2% paraformaldehyde at room temperature and subsequent collection of cells on white polycarbonate filters (0.2-mm pore size). Results from previous tests had shown that paraformaldehyde fixation as described in the previous paragraph (2%, 1 h) increases the background noise, as it induces formation of artificial ELFH97 signals. Filters were stored at 220uC until further analyses. CARD-FISH--CARD-FISH was performed as previously described (Pernthaler et al. 2004). Cells on the filters were permeabilized with lysozyme (10 mg L21, 37uC, 60 min) and achromopeptidase (0.12 U mL21, 37uC, 30 min). Hybridization was carried out overnight at 35uC using an array of horseradish peroxidase­labeled probes (formamide concentrations applied in the hybridization in parentheses): EUB338I-III (55%) for detection of Eubacteria (Amann et al. 1990; Daims et al. 1999), ALF986 (45%) for detection of Alphaproteobacteria (Neef 1997), BET42a (55%) for detection of Betaproteobacteria (Manz et al. 1992), GAM42a (55%) for detection of Gammaproteobacteria (Manz et al. 1992), CF319a (55%) for detection of Flavobacteria (Manz et al. 1996), HGC69a (35%) for detection of Actinobacteria (Roller et al. 1994), AcI852 (55%) for detection of Cluster I freshwater Actinobacteria (Warnecke et al. 2005), and NON338 (20%) as a negative control (Wallner et al. 1993). The CARD-FISH signal amplification step was performed with Alexa FluorH 488 (Invitrogen) and the cells on the filters were counterstained with 49,6-diamidino-2-phenylindole (DAPI; 1 mg mL21) immediately after the amplification step. MAR--MAR was performed in combination with CARD-FISH and ELFH97-NAG­treated samples as previously described (Alonso and Pernthaler 2005). The respective filter pieces were glued onto microscopy glass slides using Uhu plus (Uhu GmbH) and subsequently covered with a 46uC pre-warmed photographic emulsion (NTB2; Kodak) containing 0.05% agarose. The emulsion on the filters was solidified by placing the slides for 10 min on an ice-cold metal bar. Subsequently, filters were stored at 4uC in darkness until development. Filters hybridized with EUB338 were used to experimentally determine the optimal exposure time for each sample and each substrate. When the respective optimal exposure time was reached (4­ 7 d), filters were developed for 3 min in D19 developer (Kodak), rinsed 30 s in MilliQ water, and fixed 3 min in Tmax fixer (Kodak). After a subsequent washing step for 5 min with distilled water, slides with attached filters were


Beier and Bertilsson

Table 1. Incorporation rates (nmol L21 h21) of Glc, NAG, and diNAG at a concentration of 10 nmol L21, with the standard deviation indicated in parentheses. Erken Glc NAG diNAG 10.7(60.4) 3.3(60.5) 1.8(60.1) Erken-chitin 133.3(62.0) 66.9(65.1) 140.5(60.1) Ekoln 20.7(62.3) 4.2(60.01) 3.6(60.1) Ekoln-chitin 219.3(61.9) 147.2(63.6) 185.1(68.2)

Fig. 2. Expt 1: Relative increase of incorporation rates for diNAG, NAG, and Glc after chitin incubation of lake water samples. The demonstrated increase factor is the ratio of the respective incorporation rates of the chitin enrichments after incubation to that in the initial water samples. Error bars indicate the standard deviation.

dried overnight at room temperature in a desiccator prior to microscopic analyses. Microscopic counts and spatial organization of cells-- Freely dispersed cells on the filters were counted manually by microscopy (Nikon Y-FL; ocular: CFIUW 103/25; objective: 1003/1.3 oil). A minimum of 10 fields containing at least 1000 DAPI-stained cells was counted. Particles and particle-attached cells were not included in the quantitative counts. Cells positive for CARD-FISH and ELFH97 signals are given relative to total DAPI counts, and the fraction of active cells is given relative to the respective CARD-FISH counts. In addition, the identity of cells attached to particles in the chitin treatments was assessed qualitatively. Pictures for visualizing the spatial organization of cells were obtained in the Rudbeck laboratory (Cell Analysis Core Facility) on a Zeiss 510 Meta confocal microscope (laser: Argon/2 for Alexa FluorH 488, Laserdiode 405 for DAPI; emission filters: 505­530 nm for Alexa FluorH 488, 420­480 nm for DAPI; objective: 633/1.4 oil) using the system-embedded software (LSM510, version 4.2).

bacterial cells incorporating Glc, NAG, and diNAG was 2­ 3-fold higher than the initial lake water (Fig. 3). For the combined bacterial domain, the fraction of bacteria incorporating the chitin monomer NAG was always slightly higher than the fraction of bacteria taking up diNAG. ELFH97-NAG activity, targeting expression of chitinase activity at a single-cell level, was below the detection level in the initial lake water and varied between 0.3% and 0.5% of total DAPI counts in the chitin-enriched incubations (Fig. 3). With the conservative assumption that only cells that hybridize with the probes EUB338I-III represent ELFH97-NAG active cells, the fraction of bacterial cells expressing chitinase activity would be 0.3% and 0.6%, respectively. The ELFH97-NAG approach was further combined with MAR, revealing that 75% (Erken) and 67% (Ekoln) of ELFH97-NAG active cells incorporated NAG, whereas the corresponding percentages for diNAG was 90% (Erken) and 75% (Ekoln), respectively. Bacterial community structure in the chitin incubations-- Actinobacteria was the most abundant planktonic bacterial phylum in all four samples, making up between 18% and 44% of DAPI counts. Application of more specific probes revealed that species belonging to the freshwater Ac1 cluster of Actinobacteria (Warnecke et al. 2004) were the most abundant group in this phylum, accounting for 87­ 95% of the total Actinobacteria. Flavobacteria (10.1­ 20.9%), Betaproteobacteria (11.0­16.9%), and Alphaproteobacteria (6.0­13.6%) were also significant components of the combined bacterial communities in the four samples in


Incorporation rates--For both lakes, incorporation rates of all three compounds studied (NAG, diNAG, and Glc), were higher in the chitin-enriched treatments compared to the initial lake water. The smallest increase caused by the chitin enrichment was observed for Glc, with , 10-fold higher rates in the chitin treatment compared to the lake water samples. The increase in NAG incorporation rates was lower for Erken than for Ekoln (20-fold and 35-fold, respectively), whereas the increase in diNAG incorporation rates due to the experimental chitin enrichment was higher (78-fold and 51-fold, respectively). Hence, in both lakes the increase of diNAG incorporation rates clearly exceeded the increase of NAG incorporation rates (Fig. 2). Raw data of the rate measurements are given in Table 1. Single-cell chitinase activity and metabolite uptake--In the initial lake water, the fraction of bacterial cells incorporating Glc was approximately twice as high as the fraction of bacteria taking up NAG and diNAG (6­7%) (Fig. 3). In the chitin-enriched treatments, the fraction of

Fig. 3. Expt 1: Proportion of bacteria (EUB338-II-III) taking up NAG, diNAG, Glc, or expressing chitinase activity (ELFH97-NAG, assuming that all detected ELFH97-NAG­ positive cells are bacteria) in the chitin incubation experiment. Error bars indicate standard deviation for samples where triplicate incubation bottles were processed. (Ekoln and Erken indicate the initial lake water samples, and Erken- and Ekoln-chitin the respective chitin enrichments.)

Chitin use in bacterioplankton


Fig. 4. Expt 1: Relative contribution of bacterial phyla and families within the Proteobacteria to DAPI counts in the chitin incubation experiment. The difference of the sum of the phyla to the counts with the probe EUBI-III is described as unidentified bacteria. (Ekoln and Erken indicate the initial lake water samples, and Erken- and Ekoln-chitin the respective chitin enrichments; ``bacteria'' in phylogenetic group names is abbreviated with ``b.'')

expt 1. Planktonic Gammaproteobacteria were always rare, contributing from 1.1% to 1.9% of the DAPI counts (Fig. 4). Signals of the NON338 probe never exceeded 0.7% of DAPI counts. The overall shift in bacterial community composition resulting from the incubations with chitin supplementation followed the same trend in both lakes: whereas the relative contribution of Proteobacteria families stagnated or decreased, an increase in Actino- and Flavobacteria was observed (Fig. 4). Still, as total cell counts (DAPI) increased in both incubations with chitin supplementation (Erken: from 3.1 to 5.6 3 106 mL21, Ekoln: from 2.5 to 9.5 3 106 mL21), the absolute abundance of all studied bacterial groups increased compared to initial samples, with Alphaproteobacteria in Erken as the sole exception. In contrast to the chitin-enriched incubations, Actinobacteria and Flavobacteria decreased markedly in the control incubations without chitin addition (Erken control incubation: Actinobacteria: 4.1%, Flavobacteria: 2.1%; Ekoln control incubation: Actinobacteria: 2.5%, Flavobacteria: 2.5%). Qualitative observations of bacteria on particles in the chitin enrichments revealed that all targeted phylogenetic groups except Actinobacteria accumulated on the chitin particles. Notably, Actinobacteria cells never aggregated on particles (. 50 particles inspected) but were instead evenly distributed over the entire filter (Fig. 5A,B). We could further show, with confocal microscopy, that Actinobacteria were located on the filter surface rather than on top of the chitin particles (Fig. 5B). Based on these observations, we conclude that Actinobacteria, in contrast to the other phyla, were not physically associated with particulate

Fig. 5. Expt 1: Three-dimensional location of Actinobacteria (green) and DAPI-stained cells (red) on a particle in the chitin incubation experiment: (A) x­y axis, (B) x­z axis; picture by confocal microscopy). The individual z-levels and channels were merged using the following adjustments: threshold: 10%, ramp: 50%, maximum opacity: 50%.

chitin. Morphologically, Actinobacteria were exclusively small coccoid or selenoid cells, whereas many Flavobacteria, and to some extent also Proteobacteria, formed filaments in the chitin treatments. Abundance and habitat preference of Actinobacteria in the daphnia enrichment--Addition of daphnia particles to lake water in expt 2 resulted in a pronounced disproportional increase of Actinobacteria and the AcI cluster within this phylum, starting from 4% and 3% of DAPI counts, respectively, at the third day of incubation and reaching 27% and 24%, respectively, after 8 d of incubation (Fig. 6). The total abundance of planktonic bacteria decreased slightly within the study period (day 3: 4.1 3 106, day 5: 3.3 3 106, day 8: 2.9 3 106). Regardless, Actinobacteria and members of the AcI cluster were clearly replicating during the observed time period: their total cell number doubled between days 3 and 5, and increased again , 43 between days 5 and 8. Cell staining with DAPI revealed that the added zooplankton carapace particles were intensively colonized


Beier and Bertilsson with particles, the Ac1 bacteria were usually observed on the filter plane below the particles rather than on top of the particle surface. In contrast, parallel hybridization with the general Actinobacteria probe HGC69a revealed a few aggregates of three to five cells on particles. These cells were larger and more brightly fluorescing than the majority of the planktonic Actinobacteria. (For both probes and all days . 50 particles were inspected.) Hydrolysis product uptake of single bacterial phyla-- Among the targeted phylogenetic groups and for all four samples of expt 1, Actinobacteria had the highest fraction of cells incorporating NAG (Fig. 7A). In the initial lake water, between 10% (Erken) and 14% (Ekoln) of the Actinobacteria were active in NAG uptake, whereas NAG uptake activity reached 31% in Erken and 22% in Ekoln for chitin-enriched incubations. The active fraction of the abundant freshwater AcI cluster within the Actinobacteria usually had values close to those of total Actinobacteria (Fig. 7A). With the exception of Betaproteobacteria in the Erken-chitin enrichment, all other phyla were less active than the bacterial average in NAG uptake (Fig. 7A). Also for diNAG uptake, Actinobacteria were among the most active groups, at least in the initial lake water samples (Fig. 7B). In Erken, 11% of the Actinobacteria cells were active in diNAG uptake, whereas the corresponding fraction was 7% for Ekoln. A different pattern emerged for chitin-

Fig. 6. Expt 2: Relative contribution of Actinobacteria and members of the AcI cluster to DAPI counts in the daphnia enrichment experiment after 3, 5, and 8 d of incubation.

after 3 d of incubation. Analogous to the incubation experiments with colloidal chitin, members of the AcI cluster were evenly distributed across the membrane filter surface and never formed clusters on particles. When they coincided

Fig. 7. Expt 1: Fraction of cells within the targeted bacterial phylogenetic groups, taking up chitin hydrolysis products in the chitin incubation experiment. Error bars indicate standard deviation for treatments where samples from triplicate cultures were counted. The gray bars in the background describe the fraction of active bacteria (EUB338-II-III) as already shown in Fig. 2. (A) NAG uptake, (B) diNAG uptake. (Ekoln and Erken indicate the initial lake water samples, and Erken- and Ekoln-chitin the respective chitin enrichments.)

Chitin use in bacterioplankton


Fig. 8. Expt 1: Sum of active cells (A) NAG, and (B) diNAG in each of the targeted bacterial phyla compared to the active cells detected among the bacteria (Ekoln and Erken indicate the initial lake water samples, and Erken- and Ekoln-chitin the respective chitin enrichments; ``bacteria'' in phylogenetic group names is abbreviated with ``b.'')

enriched treatments. In these incubations Actinobacteria were proportionally less active in diNAG uptake (14% and 17% in Erken and Ekoln, respectively) compared to the bacterial average. Instead, a large proportion of Flavobacteria were active in diNAG uptake (21% and 25% in Erken and Ekoln, respectively). Also, Betaproteobacteria and Gammaproteobacteria were more active in the uptake of this substrate (Fig. 7B). In brief, clear differences in substrate preferences were observed for individual phyla and families (Fig. 7). The sum of the active cells in each of the individual targeted phylogenetic groups corresponded well to the active fraction counted for the total bacterial domain using probe set EUB338I-III (Fig. 8). The background signals for NAG and diNAG uptake in killed controls never exceeded 0.7% of the bacterial cells.


The large increase in diNAG incorporation rates after chitin addition illustrates the quantitative dominance of diNAG compared to NAG as a chitin degradation intermediate. This has already been demonstrated for cultures (Horn

et al. 2006), and here we extend these findings to natural aquatic microbial communities. In contrast to the rate measurements, the importance of diNAG as a chitin degradation intermediate was not reflected in the fraction of planktonic bacteria incorporating the respective substrate; i.e., in chitin treatments, the fraction of cells incorporating diNAG was lower than the fraction of cells incorporating NAG (Fig. 7). However, the number of cells incorporating a certain substance is not directly related to the total uptake activity of the community because the amount of substrate taken up by individual cells can vary (Sintes and Herndl 2006). Moreover, bulk rate measurements include the activity of particle-attached cells, while our quantitative single-cell analyses included only planktonic cells. In our experiments, the number of free-living planktonic cells incorporating NAG exceeded the number of chitinase positive cells by 40- to 60-fold (Fig. 3). However, this quantitative comparison does not take the particle-associated bacteria into account. Additionally, the ELFH97-NAG signal only indicates momentarily active chitinases, whereas this enzymatic activity might be a highly regulated process with strong temporal dynamics. Preliminary time series experiments testing the dynamics of chitinase activity patterns in a chitin-enriched lake water incubation indeed revealed strong fluctuations in the number of chitinase-positive bacteria within 12 d, ranging from 0.3% to 4% in the planktonic fraction (data not shown). It has also been assumed in an earlier study that the ELFH97-NAG approach underestimates the number of chitinase positive cells (Baty et al. 2000). Due to the pronounced difference in the fraction of ELFH97NAG positive cells and those taking up hydrolysis products, our results still clearly support the hypothesis that the uptake of hydrolysis products without simultaneous expression of the respective hydrolytic enzymes is a common trait in aquatic bacteria (Fig. 3). Our values for single-cell NAG uptake are within the earlier reported range, with 4­40% of lake bacteria incorporating NAG (Nedoma et al. 1994). It is known that the ability to degrade chitin and take up the hydrolysis products is widespread among bacterial phyla (Cottrell and Kirchman 2000). Also in the present study, all phyla analyzed were taking up the chitin hydrolysis products to some extent (Fig. 7). Furthermore, the majority of the groups quantified by CARD-FISH increased their total cell numbers after chitin addition, and there were no extreme shifts in the composition of the targeted community. However, the actively proliferating members of Actinobacteria and Flavobacteria clearly profited disproportionately from the addition of chitin. Because the contribution of these groups to the total cell counts decreased in control incubations without chitin addition and because both groups were highly active in the consumption of the hydrolysis products, we attribute the success of these two phyla directly to the added chitin. The involvement of Flavobacteria in the hydrolysis of macromolecules, such as chitin and cellulose, has been suggested before (Kirchman 2002). Even if the majority of planktonic Flavobacteria taking up NAG and diNAG were not ELFH97-NAG positive, the occurrence of this phyla in dense clusters on chitin particles implies that they were


Beier and Bertilsson from zooplankton. Actinobacteria were also highly active in accumulating chitin hydrolysis products, suggesting that chitin might be an important nutrient and energy source for members of this phylum. We suggest that members of freshwater Actinobacteria profit from a metabolic strategy as ``cheaters,'' wherein they use chitin hydrolysis products without investing in expression of the extensive enzymatic machinery needed for the degradation of such polymeric organic matter. This substrate acquisition strategy might be frequent in aquatic ecosystems. More research is needed to assess whether or not this is a common feature in the degradation of other polymeric substrates, considering that it has been shown that degradation of particulate organic matter (POM), regardless of its origin, follows similar general patterns (Biddanda and Pomeroy 1988). The possible involvement of several complementary ``cheating'' species with a wide range of different substrate uptake affinities, that are likely higher than those of true POM degraders, might lead to a near-complete utilization of substrate. In an early study (Jacobsen and Azam 1984), a considerable fraction of labeled fecal pellets was recovered in the size fraction of free-living cells. The authors attributed this to the detachment of cells from particles. However, our experiments show that also permanently free-living cells contribute substantially to the degradation of particulate organic matter. The presence of such ``cheating'' species would influence the processing and degradation of internally produced and terrigenous organic particles. Bacterial communities would respond more rapidly to such particle inputs, and this could eventually lead to a less efficient burial of carbon in sediments.

Acknowledgments We thank Laura Alonso-Saez, Ramiro Logares, and two anonymous reviewers for constructive comments on the manuscript. We also thank Dirk Pacholsky for his help with the confocal microscopy, and Anja Wenzel and Tobias Vrede for providing the zooplankton culture. The work was funded by the Swedish Research Council Formas, Uppsala Microbiomics Centre (UMC), the Swedish Research Council (grant to S. Bertilsson), and the Ohlsson-Borgh Foundation (grant to S. Beier).

actively contributing to the hydrolysis of chitin. The high numbers of planktonic Flavobacteria might accordingly be a result of active detachment of cells from particles or complete dissolution of particles. In contrast, members of the freshwater Actinobacteria cluster AcI identified by CARD-FISH seemed to be exclusively free-living cells, independently of whether or not pure chitin or the chitinous zooplankton carapace was added. This was apparent from microscopic inspection of particles (Fig. 5), and is in agreement with earlier studies suggesting that members of the AcI cluster have a planktonic lifestyle (Allgaier et al. 2007). It is unlikely that exoenzymes mediating hydrolysis of water-insoluble polymers would be an ecologically meaningful trait for organisms with a strictly planktonic lifestyle because their habitat is spatially distant from the site of the enzymatic activity. Therefore, results from our experiment raise the question of whether or not NAG- and diNAG-accumulating Actinobacteria are commensals that feed on the hydrolysis products released by other species but are not able to hydrolyze chitin themselves. A further indication that the metabolically active Actinobacteria do not produce chitinases is the observation that a higher fraction of cells in the chitin treatment is incorporating NAG rather than diNAG. Because diNAG seems to be the main product of chitinases, chitin degraders are more likely to invest in transporters and uptake mechanisms for the dimer rather than the monomer. This has been observed for chitindegrading organisms that were not even able to metabolize NAG (Ramaiah et al. 2000) and is also reflected in our results, where a higher fraction of ELFH97-NAG positive cells incorporated diNAG than NAG. True chitin degraders are, therefore, likely shifting the ratio of released hydrolysis products toward NAG by specifically or preferentially incorporating diNAG. Furthermore, NAG does not originate only from chitin, but also from other biogenic molecules, such as peptidoglycan. Thus, it might be favorable for free-living, non-chitinolytic cells to invest in the NAG, rather than the diNAG uptake system, at least in environments with sufficient nutrient availability. This might serve as an explanation for the lower amount of Actinobacteria incorporating diNAG compared to those incorporating NAG in the nutrient-rich Ekoln water and the two chitin-enriched samples, respectively (Fig. 7). Actinobacteria often make up the majority of planktonic bacterial communities in lakes, where they may contribute up to 70% of the total cells (Warnecke et al. 2005). A few members of the freshwater Actinobacteria have been isolated or highly enriched (Hahn 2009), but there are no pure cultures of the abundant AcI cluster available, yet. MAR-FISH experiments for environmental samples have demonstrated that freshwater Actinobacteria incorporate a variety of small organic compounds (Buck et al. 2009), but it has so far not been possible to identify the sources of nutrients and energy that freshwater Actinobacteria rely on for proliferation. Here we demonstrate enhanced actinobacterial growth upon exposure to chitin or chitinous particles derived


¨ ALLGAIER, M., S. BRUCKNER, E. JASPERS, AND H. P. GROSSART. 2007. Intra- and inter-lake variability of free-living and particle-associated Actinobacteria communities. Environ. Microbiol. 9: 2728­2741, doi:10.1111/j.1462-2920.2007.01385.x ALONSO, C., AND J. PERNTHALER. 2005. Incorporation of glucose under anoxic conditions by bacterioplankton from coastal North Sea surface waters. Appl. Environ. Microbiol. 71: 1709­1716, doi:10.1128/AEM.71.4.1709-1716.2005 AMANN, R. I., B. J. BINDER, R. J. OLSON, S. W. CHISHOLM, R. DEVEREUX, AND D. A. STAHL. 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56: 1919­1925. BATY, A. M., C. C. EASTBURN, Z. DIWU, S. TECHKARNJANARUK, A. E. GOODMAN, AND G. G. GEESEY. 2000. Differentiation of chitinaseactive and non-chitinase-active subpopulations of a marine bacterium during chitin degradation. Appl. Environ. Microbiol. 66: 3566­3573, doi:10.1128/AEM.66.8.3566-3573.2000

Chitin use in bacterioplankton

BIDDANDA , B. A., AND L. R. P OMEROY . 1988. Microbial aggregation and degradation of phytoplankton-derived detritus in seawater. 1. Microbial succession. Mar. Ecol. Prog. Ser. 42: 79­88, doi:10.3354/meps042079 BUCK, U., H. P. GROSSART, R. AMANN, AND J. PERNTHALER. 2009. Substrate incorporation patterns of bacterioplankton populations in stratified and mixed waters of a humic lake. Environ. Microbiol. 11: 1854­1865, doi:10.1111/j.1462-2920. 2009.01910.x CAUCHIE, H.-M., M.-F. JASPAR-VERSALI, L. HOFFMANN, AND J.-P. ´ THOME. 2002. Potential of using Daphnia magna (Crustacea) developing in an aerated waste stabilisation pond as a commercial source of chitin. Aquaculture 205: 103­117, doi:10.1016/S0044-8486(01)00674-3 COTTRELL, M. T., AND D. L. KIRCHMAN. 2000. Natural assemblages of marine proteobacteria and members of the Cytophaga-Flavobacter cluster consuming low- and highmolecular-weight dissolved organic matter. Appl. Environ. Microbiol. 66: 1692­1697, doi:10.1128/AEM.66.4.1692-1697. 2000 ¨ DAIMS, H., A. BRUHL, R. AMANN, K. H. SCHLEIFER, AND M. WAGNER. 1999. The domain-specific probe EUB338 is insufficient for the detection of all Bacteria: Development and evaluation of a more comprehensive probe set. Syst. Appl. Microbiol. 22: 434­444. EVERUSS, K. J., M. W. DELPIN, AND A. E. GOODMAN. 2008. Cooperative interactions within a marine bacterial dual species biofilm growing on a natural biodegradable substratum. Aquat. Microb. Ecol. 53: 191­199, doi:10.3354/ame01235 GROSSART, H. P. 2010. Ecological consequences of bacterioplankton lifestyles: Changes in concepts are needed. Environ. Microbiol. Rep. 2: 706­714, doi:10.1111/j.1758-2229.2010. 00179.x HAHN, M. W. 2009. Description of seven candidate species affiliated with the phylum Actinobacteria, representing planktonic freshwater bacteria. Int. J. Syst. Evol. Microbiol. 59: 112­117, doi:10.1099/ijs.0.001743-0 HORN, S. J., A. SØRBOTTEN, B. SYNSTAD, P. SIKORSKI, M. SØRLIE, ° K. M. VARUM, AND V. G. H. EIJSINK. 2006. Endo/exo mechanism and processivity of family 18 chitinases produced by Serratia marcescens. FEBS J. 273: 491­503, doi:10.1111/ j.1742-4658.2005.05079.x JACOBSEN, T. R., AND F. AZAM. 1984. Role of bacteria in copepod fecal pellet decomposition--colonization, growth-rates and mineralization. Bull. Mar. Biol. 35: 495­502. JAGMANN, N., H. P. BRACHVOGEL, AND B. PHILIPP. 2010. Parasitic growth of Pseudomonas aeruginosa in co-culture with the chitinolytic bacterium Aeromonas hydrophila. Environ. Microbiol. 12: 1787­1802, doi:10.1111/j.1462-2920.2010.02271.x KANEKO, T., AND R. R. COLWELL. 1978. Annual cycle of Vibrio Parahaemolyticus in Chesapeake Bay. Microb. Ecol. 4: 135­155, doi:10.1007/BF02014284 KEYHANI, N. O., AND S. ROSEMAN. 1997. Wild-type Escherichia coli grows on the chitin disaccharide, N,N9-diacetylchitobiose, by expressing the cel operon. Proc. Natl. Acad. Sci. USA 94: 14367­14371, doi:10.1073/pnas.94.26.14367 ------, AND ------. 1999. Physiological aspects of chitin catabolism in marine bacteria. Biochim. Biophys. Acta 1473: 108­122, doi:10.1016/S0304-4165(99)00172-5 KIRCHMAN, D. L. 2002. The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol. Ecol. 39: 91­100, doi:10.1111/j.1574-6941.2002.tb00910.x ------, AND J. WHITE. 1999. Hydrolysis and mineralization of chitin in the Delaware Estuary. Aquat. Microb. Ecol. 18: 187­196, doi:10.3354/ame018187


¨ ° LINDSTRO M, M., L. HA KANSON, O. ABRAHAMSSON, AND H. JOHANSSON. 1999. An empirical model for prediction of lake water suspended particulate matter. Ecol. Model. 121: 185­198, doi:10.1016/S0304-3800(99)00081-2 MANZ, W., R. AMANN, W. LUDWIG, M. VANCANNEYT, AND K.-H. SCHLEIFER. 1996. Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiology 142: 1097­1106, doi:10.1099/ 13500872-142-5-1097 ------, ------, ------, M. WAGNER, AND K. H. SCHLEIFER. 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: Problems and solutions. Syst. Appl. Microbiol. 15: 593­600. MIDDELBOE, M., M. SØNDERGAARD, Y. LETARTE, AND N. H. BORCH. 1995. Attached and free-living bacteria--production and polymer hydrolysis during a diatom bloom. Microb. Ecol. 29: 231­248, doi:10.1007/BF00164887 ´ NEDOMA, J., J. VRBA, J. HEJZLAR, K. SIMEK, AND V. STRASKRABOVA. 1994. N-acetylglucosamine dynamics in freshwater environments: Concentration of amino sugars, extracellular enzyme activities, and microbial uptake. Limnol. Oceanogr. 39: 1088­1100, doi:10.4319/lo.1994.39.5.1088 NEEF, A. 1997. Anwendung der in situ Einzelzell-Identifizierung von Bakterien zur Populationsanalyse in komplexen mikrobiellen Biozonosen. Ph.D. thesis. Technische Univ. Munchen. ¨ ¨ [Application of single cell in situ identification of bacteria for population level analyses in complex microbial biocenoses.] PERNTHALER, A., J. PERNTHALER, AND R. AMANN. 2004. Sensitive multi-color fluorescence in situ hybridization for the identification of environmental microorganisms, p. 711­726. In G. A. Kowalchuk, F. J. De Bruin, I. M. Head, A. D. L. Akkermans, and J. D. Van Elsas [eds.], Molecular microbial ecology manual. Kluwer. RAMAIAH, N., R. T. HILL, J. CHUN, J. RAVEL, M. H. MATTE, W. L. STRAUBE, AND R. R. COLWELL. 2000. Use of a chiA probe for detection of chitinase genes in bacteria from the Chesapeake Bay. FEMS Microbiol. Ecol. 34: 63­71, doi:10.1111/j.15746941.2000.tb00755.x RIEMANN, L., AND F. AZAM. 2002. Widespread N-acetyl-Dglucosamine uptake among pelagic marine bacteria and its ecological implications. Appl. Environ. Microbiol. 68: 5554­5562, doi:10.1128/AEM.68.11.5554-5562.2002 ROLLER, C., M. WAGNER, R. AMANN, W. LUDWIG, AND K. H. SCHLEIFER. 1994. In situ probing of gram-positive bacteria with high DNA G+C content using 23S rRNA targeted oligonucleotides. Microbiology 140: 2849­2858, doi:10.1099/ 00221287-140-10-2849 SHARON, N. 1965. Distribution of amino sugars in microorganisms, plants and invertebrates, p. 1­44. In E. A. Balazs and R. W. Jeanloz [eds.], The amino sugars--distribution and biological role. Academic Press. SINTES, E., AND G. J. HERNDL. 2006. Quantifying substrate uptake by individual cells of marine bacterioplankton by catalyzed reporter deposition fluorescence in situ hybridization combined with micro autoradiography. Appl. Environ. Microbiol. 72: 7022­7028, doi:10.1128/AEM.00763-06 SMITH, D. C., M. SIMON, A. L. ALLDREDGE, AND F. AZAM. 1992. Intense hydrolytic enzyme activity on marine aggregates and implications for rapid particle dissolution. Nature 359: 139­142, doi:10.1038/359139a0 SVITIL, A. L., S. M. N. CHADHAIN, J. A. MOORE, AND D. L. KIRCHMAN. 1997. Chitin degradation proteins produced by the marine bacterium Vibrio harveyi growing on different forms of chitin. Appl. Environ. Microbiol. 63: 408­413.


Beier and Bertilsson

WURZBACHER, C. M., F. BARLOCHER, AND H. P. GROSSART. 2010. Fungi in lake ecosystems. Aquat. Microb. Ecol. 59: 125­149, doi:10.3354/ame01385

WALLNER, G., R. AMANN, AND W. BEISKER. 1993. Optimizing fluorescent in situ hybridization with rRNA-targeted oligonucleotide probes for flow cytometric identification of microorganisms. Cytometry 14: 136­143, doi:10.1002/cyto.990140205 WARNECKE, F., R. AMANN, AND J. PERNTHALER. 2004. Actinobacterial 16S rRNA genes from freshwater habitats cluster in four distinct lineages. Environ. Microbiol. 6: 242­253, doi:10.1111/j.1462-2920.2004.00561.x ------, R. SOMMARUGA , R. SEKAR, J. S. HOFER , AND J. PERNTHALER. 2005. Abundances, identity, and growth state of Actinobacteria in mountain lakes of different UV transparency. Appl. Environ. Microbiol. 71: 5551­5559, doi:10.1128/AEM.71.9.5551-5559.2005

Associate editor: Peter Hernes Received: 08 November 2010 Accepted: 07 March 2011 Amended: 18 March 2011


Beier, Sara, and Stefan Bertilsson. Uncoupling of chitinase activity and uptake of hydrolysis products in freshwater bacterioplankton. Limnol. Oceanogr., 56(4), 2011, 1179­1188

10 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


Notice: fwrite(): send of 207 bytes failed with errno=104 Connection reset by peer in /home/ on line 531