Microb Ecol (2008) 56:505–512 DOI 10.1007/s00248-008-9369-8
ORIGINAL ARTICLE
Elevated Lytic Phage Production as a Consequence of Particle Colonization by a Marine Flavobacterium (Cellulophaga sp.) Lasse Riemann & Hans-Peter Grossart
Received: 10 October 2007 / Accepted: 25 January 2008 / Published online: 18 March 2008 # Springer Science + Business Media, LLC 2008
Abstract Bacteria growing on marine particles generally have higher densities and cell-specific activities than freeliving bacteria. Since rapidity of phage adsorption is dependent on host density, while infection productivity is a function of host physiological status, we hypothesized that marine particles are sites of elevated phage production. In the present study, organic-matter-rich agarose beads and a marine phage–host pair (Cellulophaga sp., ΦSM) were used as a model system to examine whether bacterial colonization of particles increases phage production. While no production of phages was observed in plain seawater, the presence of beads enhanced attachment and growth of bacteria, as well as phage production. This was observed because of extensive lysis of bacteria in the presence of beads and a subsequent increase in phage abundance both on beads and in the surrounding water. After 12 h, extensive phage lysis reduced the density of attached bacteria; however, after 32 h, bacterial abundance increased again. Reexposure to phages and analyses of bacterial isolates suggested that this regrowth on particles was by phage-resistant clones. The present demonstration of elevated lytic phage production associated with model particles illustrates not only that a marine phage has the ability to successfully infect and lyse surface-attached bacteria but also that acquisition of resistance may affect L. Riemann (*) Department of Natural Sciences, University of Kalmar, 39182 Kalmar, Sweden e-mail:
[email protected] H.-P. Grossart Department of Limnology of Stratified Lakes, Leibniz-Institute of Freshwater Ecology and Inland Fisheries, Alte Fischerhuette 2, 16775 Stechlin, Germany e-mail:
[email protected]
temporal phage–host dynamics on particles. These findings from a model system may have relevance to the distribution of phage production in environments rich in particulate matter (e.g., in coastal areas or during phytoplankton blooms) where a significant part of phage production may be directly linked to these nutrient-rich “hot spots.”
Introduction Macroscopic organic aggregates (marine snow) constitute up to 63% of the particulate organic matter in the ocean [2]. They are important for carbon and nutrient cycling, even in oligotrophic waters [37], and serve as “hot spots” for microbial colonization, activity, and growth ([3, 44] and references therein). Given that the particle itself provides a favorable environment for the bacteria, e.g., elevated substrate availability, it is not surprising that an elevated cell-specific activity is often found (e.g., [16]). In addition, surface attachment is advantageous, as a given bacterium is kept in a location where dissolved nutrients are constantly replenished due to enzymatic hydrolysis and advective flow [22]. For instance, it has been shown that just the introduction of inert glass beads to a dilute seawater medium increased bacterial growth efficiency by 14% and rates of proteolytic enzyme activity and cell-specific [3H] leucine incorporation into protein by factors of 6.8 and 2.2, respectively [48]. Phages and heterotrophic nanoflagellates are the main mortality factors for marine bacterioplankton [14]. Phages are believed to lyse up to 10–20% of the bacterial community per day [47], and it has been estimated that up to a quarter of the photosynthetically fixed carbon in the ocean is recycled back to dissolved organic material by viral lysis [50]. The nutritional or metabolic status of a
506
bacterial host, however, is critical for viral infection and proliferation. It has been shown, for example, that low nutrient availability results in long latent periods and reduced burst size [1, 23], while maximum proliferation rates and yield of phages are observed at optimum growth conditions for the host [51]. Hence, it is conceivable that the presence of particles in a given environment will lead to elevated bacterial activity and, thereby, to an increased production of phages. Similar to lysis of single-species biofilms [12], enhanced growth of mono-specific microcolonies on particles therefore could allow for efficient phage infection, lysis, and release into the surrounding water. Only a few studies have examined viruses on particles in aquatic environments. In an early study, Proctor and Fuhrman [38] examined particles from sediment traps using transmission electron microscopy. They reported that phages were often aggregated and estimated that 2–37% of the particle-associated bacteria were killed by phages. In later studies, viral abundance does appear higher on particles relative to ambient water [27, 36]; however, it is unclear to what extent viruses are produced in these loci or merely adsorbed and thereby inactivated due to the loci surface charge [33]. For instance, viruses readily attach to freshly produced transparent exopolymeric particles (TEP) [7]. Due to an elevated viral density, it is conceivable that TEP constitute hot spots for viral infection of bacteria [29]. Furthermore, it has been suggested that the two-dimensional particle surface allows for increased encounter of hosts and thereby for efficient viral infection and lysis of non-resistant bacterial strains. This will also lead to an increasing abundance of resistant bacteria [18]. Though conceivable, these ideas suffer from a lack of understanding of the mechanisms underlying phage proliferation on particles. In the present study, we posed the hypothesis that the increased growth associated with bacterial colonization of nutrient-rich particles elicits an increased production and release of phages. To test for this, we used a simple experimental approach based on a model phage–host pair, the marine bacterium Cellulophaga sp., a specific lytic phage ΦSM [31], and small agar beads with known size [21]. This experimental setup allowed us to quantify abundance and growth of attached bacteria, as well as the lytic production of phages, and examine potential resistance acquisition among attached bacteria.
Methods The Phage–Host System The bacterium (Cellulophaga sp., Flavobacteriaceae, Bacteroidetes, GenBank AF497997) and its lytic phage (ΦSM; genome size, ~50 kb) was originally isolated from
L. Riemann, H.-P. Grossart
surface waters in the southern Kattegat, Denmark [19, 31]. No temperate phages could be induced from Cellulophaga sp. by mitomycin C induction (data not shown). The phage stock used was generated by the soft agar overlay method [40]. Briefly, several agar-overlay plates were created with the appropriate phage dilution producing almost confluent lysis. The overlay was shredded using a sterile loop, 5 ml phage buffer (0.1 M NaCl, 0.008 M MgSO4, 0.05 Tris– HCl, 2 mM CaCl2, 0.1% gelatine, 0.5 μM tryptophan, 5% glycerol, pH 7.5) was added, and the plates were incubated on a shaker for >1 h. The agar/phage buffer mixture was transferred to a sterile tube and centrifuged at 10,000×g for 5 min. Finally, the supernatant was 0.22-μm filtered to remove any remaining bacteria. The stock was stored at 4°C in the dark for 10 days before the experiment. Experimental design Before the experiment, Cellulophaga sp. was grown in marine liquid broth (MLB, 0.005 g casein hydrolysate, 0.005 g peptone, 0.005 g yeast extract, 0.03 ml 87% glycerol, 0.5 l MQ water, and 0.5 l GF/C (Whatman) filtered seawater, SW), diluted in 6 l SW to a concentration of ~104 cells ml−1, and incubated at 20°C overnight. The SW used was obtained from Kattegat (depth, 45 m; salinity, 33 PSU). Before use, the SW was 0.45-μm filtered (Millipak 40, Milipore) and autoclaved. Model agarose particles were enriched with MLB using a modified protocol as described by Kiørboe et al. [21]. Briefly, warm agarose (2%) in MLB was dripped into sterile seawater covered by a thin layer of mineral oil. Agarose beads (~0.15-cm radius) formed in the oil and sank into the seawater. They were then thoroughly washed in sterile seawater and fixed on thin glass threads. The agarose beads were precolonized with bacteria before the experiment by keeping the beads suspended on thin glass threads overnight in a bacterial suspension growing in MLB (initial concentration of 5.5×104 cells ml−1). The experimental setup consisted of four treatments in triplicates: (a) free-living bacteria (plus autoclaved suspension of phages), (b) free-living bacteria and phages, (c) freeliving bacteria and beads (plus autoclaved suspension of phages), and (d) free-living bacteria and phages and beads. Five hundred milliliters of pre-incubated overnight bacterial suspension (see above) was aliquoted into 12 clear polypropylene containers (1 l vol., #216-2640, VWR). Live phages (B and D) and autoclaved phages (A and C) were added at a final concentration of ~5.5×105 phages ml−1. A total of 50 beads were suspended on glass threads in each of the containers of treatments C and D. Containers were incubated at 20°C in the dark at slow rotation (~50 rpm) for a total of 69 h. Samples for bacterial and phage enumeration were obtained at 0, 4, 12, 21, 33, 45, 57, and
Elevated Production of Phages on Model Particles
69 h. Statistical analyses were done using analysis of covariance, with time as the covariate and treatment type as the nominal predictor. All statistical analyses were performed with the software JMP 4.02 using average values. Significance was given at p15 fields per filter) were counted. Only a small fraction of the bacteria on the beads detached when using the methanol-ultrasonication method (and these were therefore counted directly on the beads; see above), while a pronounced detachment of discernible phages was observed. Still, our abundance estimates for both attached bacteria and phages should be considered conservative. Resistance Experiments Two tests were performed to examine whether the bacterial regrowth on particles was by bacterial clones resistant to phage infection. At 45 h, the presence of phage-resistant cells on beads was examined. Nine beads from treatments C and D, respectively, were submerged for 24 h in autoclaved seawater containing a dense suspension of phage ΦSM (107 plaque forming units per milliliter). Then, the effect of phage lysis on the attached bacteria was qualitatively examined by microscopy as described above. At 69 h, six beads from treatments C and D, respectively, were vortexed in sterile seawater, and dilutions were spread onto ZoBell agar plates [5 g peptone, 1 g yeast extract, 15 g agar, 800 ml GF/C (Whatman) filtered seawater, 200 ml distilled water, autoclaved at 121°C, 20 min]. After 3 days, 20 colonies from each treatment were clean-streaked and tested for resistance by exposure to the ΦSM phage using the soft agar overlay method. To make sure that the bacteria resistant to ΦSM were not contaminants, their 16S rRNA genes were partly sequenced. Bacterial DNA was extracted using the EZNA tissue DNA kit (Omega) and the 16S rRNA gene was PCRamplified using primers 27f and 1492r as previously described [19]. The partial 16S rRNA was sequenced (commercially by Macrogen, Korea) using the primer 27f.
Results The abundance of free-living bacteria in treatments A and B did not differ significantly (p=0.263, F=1.289). In these control experiments, the abundance doubled within the first 12 h but then remained relatively stable at ~3×105 cells per ml for the rest of the experiment (Fig. 1a). Similarly, the concentration of free phages in treatment B remained stable until the last sampling point (Fig. 1b). The presence of beads and live phages significantly affected bacterial growth. In treatments C and D, an increase in abundance of free-living and attached bacteria was observed during the first 12 h (Fig. 1a,c). For free-living bacteria, the initial concentrations and the increase during the first 12 h were similar to the control treatments; however, later in the experiment, the
508
Figure 1 Dynamics of a free-living bacteria, b free phages, and c bacteria and phages associated with agar beads. Values are averages of triplicate treatments ± standard deviation
presence of beads (treatments C and D) had significant effects on the abundance of free-living bacteria (B vs. D, p
