- Email: [email protected]
Fluorescence in situ hybridization (CARD-FISH) of microorganisms in hydrocarbon contaminated aquifer sediment samples
Systematic and Applied Microbiology 35 (2012) 526–532
Contents lists available at SciVerse ScienceDirect
Systematic and Applied Microbiology journal homepage: www.elsevier.de/syapm
Fluorescence in situ hybridization (CARD-FISH) of microorganisms in hydrocarbon contaminated aquifer sediment samples Karolin Tischer a , Michael Zeder a,b,c , Rebecca Klug a , Jakob Pernthaler b , Martha Schattenhofer a , Hauke Harms a , Annelie Wendeberg a,∗ a
Helmholtz Centre for Environmental Research, Department of Environmental Microbiology, Permoserstr. 15, 04318 Leipzig, Germany University of Zurich, Department of Limnology, Seestr. 187, CH-8802 Kilchberg, Switzerland c Max Planck Institute for Marine Microbiology, Celciusstr. 1, 28359 Bremen, Germany b
a r t i c l e
i n f o
Article history: Received 30 September 2011 Received in revised form 11 January 2012 Accepted 24 January 2012 Keywords: CARD-FISH Hydrocarbon contamination Aquifer
a b s t r a c t Groundwater ecosystems are the most important sources of drinking water worldwide but they are threatened by contamination and overexploitation. Petroleum spills account for the most common source of contamination and the high carbon load results in anoxia and steep geochemical gradients. Microbes play a major role in the transformation of petroleum hydrocarbons into less toxic substances. To investigate microbial populations at the single cell level, fluorescence in situ hybridization (FISH) is now a well-established technique. Recently, however, catalyzed reporter deposition (CARD)-FISH has been introduced for the detection of microbes from oligotrophic environments. Nevertheless, petroleum contaminated aquifers present a worst case scenario for FISH techniques due to the combination of high background fluorescence of hydrocarbons and the presence of small microbial cells caused by the low turnover rates characteristic of groundwater ecosystems. It is therefore not surprising that studies of microorganisms from such sites are mostly based on cultivation techniques, fingerprinting, and amplicon sequencing. However, to reveal the population dynamics and interspecies relationships of the key participants of contaminant degradation, FISH is an indispensable tool. In this study, a protocol for FISH was developed in combination with cell quantification using an automated counting microscope. The protocol includes the separation and purification of microbial cells from sediment particles, cell permeabilization and, finally, CARD-FISH in a microwave oven. As a proof of principle, the distribution of Archaea and Bacteria was shown in 60 sediment samples taken across the contaminant plume of an aquifer (Leuna, Germany), which has been heavily contaminated with several ten-thousand tonnes of petroleum hydrocarbons since World War II. © 2012 Elsevier GmbH. All rights reserved.
Introduction Oil spills represent a major threat to worldwide aquatic and terrestrial ecosystems. It is widely acknowledged that natural microbial communities can transform contaminants into less toxic substances, and strategies exploiting microorganisms have been developed to clean up contaminated sites [10,30]. However, our understanding of the physiology and ecology of the natural microbial communities found at polluted sites is limited and bioremediation approaches are often based on trial and error. Most studies on the microbial diversity of petroleum contaminated sites are based on cultivation techniques, fingerprinting, and amplicon sequencing [12,29]. To investigate population dynamics
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. Wendeberg). 0723-2020/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. doi:10.1016/j.syapm.2012.01.004
and interspecies relationships at the single cell level, fluorescence in situ hybridization (FISH) of rRNA is now a widely used technique [11,40,46]. FISH can be combined with other methods, such as microautoradiography [28], mRNA FISH [45] and nanoSIMS [17], to study physiological traits of selected microbial populations and answer the question of “who is doing what” in a complex environment. However, only five studies on microbial communities in aquifer sediments have used FISH to date [18,23,33,35,37]. Although FISH has been more commonly applied to groundwater samples [16,22,33,34], more than 90% of the microbial biomass can be found attached to sediment particles [5,19], and it often exhibits higher activities than the planktonic community [1,2,20,27]. Therefore, detailed studies of aquifer sediments are more ecologically relevant than studies of groundwater samples, especially when evaluating the potential for natural attenuation [9]. The aim of this study was to develop a catalyzed reporter deposition (CARD)-FISH protocol for the detection of microorganisms in hydrocarbon contaminated aquifer sediment samples.
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
527
Fig. 1. Schematic of the sampling site showing the position of the Leuna aquifer in Germany and the top view of the Leuna industrial area with the hydrocarbon plume (light grey) and source (dark grey), the positions of the cores taken (C1–C8) and the cut-off trench (black line) of the pump-and-treat unit (not to scale).
For high-throughput quantification of CARD-FISH stained cells, a protocol was developed that could produce high quality cell preparations for automated FISH counting. The protocol included a cell detachment step using Tris–EDTA buffer and microwaving, detergent treatment and sonication, as well as a cell purification step using Nycodenz density gradient centrifugation. The population dynamics of Bacteria and Archaea were shown in 60 sediment samples across the petroleum hydrocarbon contaminant plume of the Leuna aquifer (Germany). Materials and methods Site description and sampling procedure The Leuna megasite has been a centre for chemical production since the beginning of the 20th century. The site has been associated with ammonia synthesis (since 1917) and coal hydrogenation (since 1927) that have led to heavy contamination due to bombing in World War II. In the late 1970s, production of methyl tertiary butyl ether (MTBE) and associated leakages led to additional subsurface contaminations. Today, large quantities of various gasoline components, especially benzene, toluene, ethyl benzene, xylene isomers (BTEX) and MTBE, reside at the site and create a contaminant plume of approximately 300 m in width and 500 m in length that reaches the cut-off trench of a pump-and-treat unit (Fig. 1). Direct-push cores were taken along the plume (C1–C5) and outside the plume (C8) in April 2008, September 2009 and February 2010 at the capillary fringe. The sediment cores were immediately transferred into an anoxic glove-box where all sub-sampling and oxygen measurements took place. To measure the O2 concentrations within each core, small holes at 4 cm intervals where poked into the plastic liners of the cores, and micro-optodes (PreSens, Regensburg, Germany) were inserted approximately 5 mm into the sediment. Cores were then sectioned into 4 cm thick sections, subsamples of each section were formaldehyde fixed (2% volume/volume [v/v] final concentration) and kept at 4 ◦ C for 24 h. The samples were washed twice with a 1:1 mix of 1 × PBS [137 mM NaCl, 2.7 mM KCl, 4 mM Na2 HPO4 , 2 mM KH2 PO4 , pH 7.6] and absolute ethanol by pelleting and resuspension, and then stored in absolute ethanol at −80 ◦ C. Sediment dry weight was determined by drying the samples at 105 ◦ C, and then samples were allowed
to cool to room temperature in an exsiccator and they were subsequently weighed. Cell separation Two-hundred microliters of sediment sample was mixed with 700 L TE buffer [10 mM TrisHCl, 5 mM EDTA, pH 9.0] and 100 L of Na-pyrophosphate [0.1 M] in a 1.5 mL reaction tube. This mixture was then placed into a water bath in a histological microwave oven (Microwave Research and Application Inc., Laurel, MD, USA) and heated to 55 ◦ C for 5 min at 200 W. After cooling the samples to room temperature (RT), 1 L of Tween 80 was added and each sample was agitated (level 3, 1060 rpm) on a vortexer (Vortex-Genie 2, Carl Roth GmbH, Karlsruhe, Germany) for 15 min at RT. Subsequently, the sample was sonicated on ice (Sonifier Model 250, Branson, Danbury, USA) with 10 pulses at 53 W. To separate dislodged cells from sediment particles, the samples were transferred into a 50 mL Falcon tube and mixed with 22.5 mL PBS and 2.5 mL Na-pyrophosphate [0.1 M]. Finally, a layer of Nycodenz (AXIS-SHIELD PoC, Oslo, Norway) solution [density 1.3 g mL−1 ; 60% (w/v), in MilliQ water] was carefully placed at the bottom of the Falcon tube using a syringe with a long needle, then the sediment was pelleted by centrifugation overnight at 4000 rpm (centrifuge 5810R, Eppendorf, Hamburg, Germany) and 4 ◦ C using a swing-out rotor (A-4-62, Eppendorf, Hamburg, Germany). After centrifugation, both the supernatant and the Nycodenz layer were collected, mixed, and filtered on white polycarbonate filters (type GTTP, pore size 0.2 m, diameter 47 mm, Millipore, Eschborn, Germany). The filter was then washed twice with autoclaved deionized water, air-dried and stored at −20 ◦ C until further processing. Cell permeabilization To prevent cell loss during the permeabilization procedure, filters were embedded in agarose. The filters were placed with the sample facing down into 200 L low-gelling-point agarose (0.1% [w/v] in MilliQ water) on Parafilm and left to dry at 35 ◦ C in a hybridization oven. Then, the filters were wetted with 50% ethanol and gently peeled off the Parafilm using forceps. The filters were allowed to air dry before sectioning, and filter sections were placed into a 1.5 mL reaction tube containing 1 mL of 1 × TE
528
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
Fig. 2. Effect of two cell dislodging procedures, (a) vortexing and (b) sonication, on total cell numbers of two selected sediment samples.
[10 mM TrisHCl, 5 mM EDTA, pH 9.0] and they were permeabilized by microwaving in a pre-heated waterbath at 65 ◦ C for 8 min at 1000 W (100% power output). The reaction tubes were then allowed to cool for 5 min at RT. Next, the cells were post-fixed in 990 L of 1 × TE and 120 L formaldehyde (37%) for 5 min at RT to stabilize the permeabilized cells for subsequent hybridization. Filters were then washed with 1 × TE, and endogenous peroxidases were inactivated with 0.1% H2 O2 in 1 × TE for 2 min at RT. The filters were then washed twice, as mentioned above, and transferred to the hybridization buffer (see below). CARD-FISH Horseradish peroxidise (HRP)-labelled probes (Biomers, Ulm, Germany) targeting Bacteria (EUB338I-III, [3,14]) and Archaea (ARCH915, [39]) were used for CARD-FISH. The hybridization was carried out in a 1.5 mL reaction tube containing up to 20 filter pieces, 1000 L hybridization buffer [0.9 M NaCl, 20 mM TrisHCl (pH 8), 10% dextran sulfate, 0.02% sodium dodecyl sulphate (SDS), 1% blocking reagent (Roche, Mannheim, Germany), 35% formamide (Fluka, Germany) and MilliQ water] and 3.3 L HRP probe working solution [50 ng L−1 ]. The tubes were placed in a pre-warmed water bath and microwaved in a histological microwave oven at 46 ◦ C for 40 min at 500 W. To wash off excess probe and to reactivate the HRP, filter sections were placed in 50 mL of 1 × PBS for 15 min at RT. Subsequently, the CARD reaction was undertaken by placing the filters in 1 mL amplification buffer [1 × PBS (pH 7.6), 2 M NaCl, 10% dextran sulphate, 0.1% blocking reagent (Roche, Mannheim, Germany) and MilliQ water], freshly amended with 0.0015% H2 O2 and 1 g mL−1 fluorescein-labelled tyramide (custom labelled; see [31]) for 15 min at 46 ◦ C in a conventional hybridization oven. Finally, filters were washed five times with deionized water, dehydrated in increasing ethanol concentrations (50%, 70%, and 100%) and allowed to air dry. For microscopic examination, filters were embedded in a DAPI amended mountant (see below). For dual hybridization of Archaea and Bacteria, two consecutive CARD-FISH stainings were carried out with an HRP inactivation step between the two (0.1% H2 O2 in 1 × TE for 2 min at RT), and Alexalabelled tyramides (custom labelled; [31]) were used for the CARD reaction. Total cell counts A general DNA stain (4 ,6-diamidino-2-phenylindole, DAPI) was added to the mountant and was used to determine the total cell
numbers. The mountant contained 1000 L Citifluor (Citifluor AF2, Citifluor Ltd., London, UK), 100 L of 1 × PBS and 1 L of a DAPI stock solution (1 mg mL−1 ). Automated counting FISH preparations were evaluated by automated microscopy [32,47–49] on an Axio Observer.Z1 (Carl Zeiss, Jena, Germany) using a 63 × Plan-Apochromat objective (NA = 1.4) and the filter sets 01 for DAPI and 38HE for FITC. The inverted microscope featured a motorized stage for four slides and an HXP 120 lamp (Carl Zeiss, Germany) for fluorescence illumination. Greyscale images were recorded on a digital camera (AxioCam MRm; Carl Zeiss). Imaging involved three steps: (i) whole slide overview images were acquired to localize the samples to be imaged, and to create a list of coordinates [48]; (ii) autonomous multi channel image acquisition was performed using a custom made Axio Vision 4.8 (Carl Zeiss, Germany) VBA macro (M. Zeder, unpublished); and (iii) images were checked for quality [50] and analyzed using the ACMEtool (available for download at www.technobiology.ch) in order to obtain and revise the bacterial cell counts. Automated counting results were corroborated by manual counting of 10 randomly chosen samples (r2 = 0.88). Results and discussion Sample preparation High concentrations of petroleum hydrocarbons in sediment samples (e.g. up to 2 mg g−1 dryweight benzene; Tischer et al., in prep., A [41]) caused unacceptably high levels of background fluorescence with both DAPI (a general DNA stain), and CARD-FISH, making it impossible to discern individual cells. Therefore, it was decided to separate and purify microbial cells from sediment particles for subsequent CARD-FISH and enumeration. Microorganisms are attached to soil particles by a variety of mechanisms, such as electrostatic interactions, binding through exopolysaccharides or proteins [25]. To detach cells from sediment particles, the sediment sample was therefore first microwaved in Tris–EDTA buffer and pyrophosphate in a histological microwave oven at 55 ◦ C. The slurry was then amended with the non-ionic surfactant Tween 80 and vortexed, followed by sonication. A time series for both steps was conducted to find the optimal balance between cell detachment, cell disruption, and increasing amounts of silt particles in the cell fraction, which caused high levels of background fluorescence (Fig. 2). Although 30 min vortexing resulted in
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
Fig. 3. Effect of 15 min and 30 min vortexing performed with all sediment samples. Groundwater table at 0 m.
529
Fig. 4. CARD-FISH detection rates using bacterial (EUB338 I–III) and archaeal (Arch915) probes. The sample used for dual CARD-FISH in Fig. 5 is indicated with an arrow.
530
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
the highest cell numbers for the two samples selected for the time series (Fig. 2a), in most of the 60 samples 15 min vortexing resulted in the highest cell numbers (Fig. 3) and was thus chosen for the final protocol. Additionally, it was found that vortexing for longer than 15 min caused higher levels of background fluorescence due to increasing amounts of silt in the cell fraction. Sonication to detach cells from soil particles is commonly used [8,13,15,21,25]. For the sonication treatment, 10 pulses of 1 s each were found to produce the highest numbers of dislodged cells (Fig. 2b), since sonication for longer than 10 pulses led to a decrease of cell numbers (Fig. 2), which was likely caused by disruption or by dislodging silt particles, thereby causing background fluorescence [6]. To separate cells subsequently from sediment particles, the slurry was subjected to Nycodenz density gradient centrifugation, as described by Barra Caracciolo et al. [6]. After density gradient centrifugation, cells could be detected in all three density layers (data not shown), in accordance with a previous report [6]. An attempt was made to collect all detached cells and, therefore, the entire supernatant was filtered onto membrane filters. The recovery of the microbial cells was comparable to previous studies [6,38] with an efficiency of 73% (average, SD = 12, tested on 8 different sediment samples with 3 replicate counts each). Additionally, a fingerprinting technique (terminal restriction length polymorphism, T-RFLP) was used to estimate a potential shift in community composition caused by cell dislodging and density gradient centrifugation. However, the community composition of both the pellet and the supernatant was found to be comparable (data not shown). Cell permeabilization Various cell permeabilization strategies to improve CARD-FISH detection rates were tested. Firstly, a time and temperature series of heating in Tris–EDTA buffer was performed and it was found that 8 min at 65 ◦ C resulted in the highest CARD-FISH detection rates (data not shown). It has been suggested that Tris binds to lipopolysaccharides and replaces stabilizing Ca2+ and Mg2+ , and therefore reduces the interaction between lipopolysaccharide molecules, while EDTA removes the divalent cations from their binding sites by chelation [43]. This results in the permeabilization of the outer membrane by release of a significant proportion of lipopolysaccharide from the cells [24]. Curiously, it was found that the TE permeabilization step in the microwave only worked in combination with hybridization. Hybridization in a conventional oven following TE permeabilization resulted in CARD-FISH signals below the detection limit. Furthermore, lysozyme, achromopeptidase, and proteinase K were tested for permeabilizing archaeal and bacterial cell walls. Although TE permeabilization and subsequent lysozyme and achromopeptidase treatment showed the highest detection rates for two samples (Table 1), in most of the 60 samples TE permeabilization resulted in a higher detection rate (Fig. 4). Additionally, cell morphologies were preserved better than after enzymatic treatment. CARD-FISH Filter sections were hybridized without delay, since storage after the permeabilization procedure sometimes led to lower detection rates (data not shown). Hybridization in a histological microwave oven increased both signal intensity and detection rate dramatically (2- to 20-fold), while decreasing the time needed for hybridization, when compared to hybridization in a conventional hybridization oven (Table 1). For more than two decades, it has been known that microwave irradiation can accelerate physical and chemical reactions [26]. When used for FISH, microwaves transfer high energy to the probes and can promote probe penetration into
Fig. 5. Dual CARD-FISH of core 3 12 cm: Archaea (Arch915, green), Bacteria (EUB338 I–III, red), and a general DNA stain (DAPI, blue). Size bar represents 10 m.
the cells [26]. Weise et al. [44] speculated that microwave irradiation may act as a steric modifier, allowing the DNA probe to find the target in a more efficient and faster way than in a normal FISH assay which depends on diffusion of probes to the target. The detection rates in Leuna aquifer sediment samples ranged between 15% and 100% for Bacteria and Archaea together, while unspecific signals (hybridized with NON338) were below 1% (data not shown). The highest detection rates were found in samples with lower concentrations of hydrocarbons: in the deepest samples of core 3 (up to 100%) and core 5 (up to 70%) and the whole of core 8 (between 36 and 60%), the latter originating from an uncontaminated part of the aquifer (Fig. 1). The lowest detection rates were found in samples that were saturated with hydrocarbons, originating either from the source zone (cores 1 and 2), or from the area around the water table, where the oil phase was situated (upper parts of core 3 and core 5; Fig. 1). Detection rates in core 2 ranged between 13 and 19%, but core 1 had to be omitted from FISH analysis, since cell numbers were below the detection limit. Low CARD-FISH detection rates in the highly contaminated zones were most likely due to toxic levels of petroleum hydrocarbons, in addition to an insufficient amount of bioavailable water in the oil phase, which may have resulted in low activity of microorganisms and thus in low ribosome copy numbers per cell. It was concluded that microorganisms in that zone were not actively taking part in the degradation of petroleum hydrocarbons, which corresponded to the plume-fringe concept, where highest activities of contaminant degradation can be found at the fringes of contaminant plumes [4]. To date, of the five studies that have used FISH on microbial communities in aquifer sediments [18,23,33,35,37], only one of them has achieved a detection rate of 96% for the eubacterial probe, but only one sample was analyzed [18]. In the other four studies, the detection rates reported were relatively low with 23% (Kleikemper et al. 2005), 33% [35], 72% [Pombo et al. 2005], and 17–75% as the sum of Eubacteria and Archaea [36,37]. Probe detection rates in this study of up to 100% were obtained for Eubacteria and Archaea together, which are considerably higher than previously reported. It was observed that the abundances of Bacteria and Archaea were remarkably similar in cores 2, 3 and 5. The specificity of these FISH results was validated by performing a dual CARDFISH staining for Archaea and Bacteria (Fig. 5). Cores 2, 3 and 5 were from the methanogenic part of the plume (Tischer et al., in prep, B [42]), where a syntrophic lifestyle is more favorable and close coupling between bacterial contaminant degraders
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
531
Table 1 Detection rates of probes targeting Bacteria (Eub338 I-III) and Archaea (Arch915) comparing various cell dislodging, permeabilization, and hybridization procedures, performed on two selected sediment samples (core 3 4 cm and core 3 56 cm). Abbreviations: TE, microwaving in TE buffer; Lys, lysozyme digestions; Achr, achromopeptidase digestion; PK , proteinase K digestion; MW, hybridization in a microwave oven; Conv., hybridization in a conventional oven; ON, overnight. Dislodging time [min]
Permeabilization
Hybridization
Bacteria
Archaea
C3–4
15 15 30 30 30
Lys, Achr TE TE TE, Lys, Achr TE, Lys, Achr, PK
Conv., ON MW, 45 min MW, 45 min MW, 45 min MW, 45 min
C3–56
C3–4
C3–56
%DAPI
SD
%DAPI
SD
%DAPI
SD
%DAPI
SD
2 5 6 7 4
0.28 0.95 0.62 2.44 2
14 36 44 43 13
2.87 2.31 2.88 1.62 5
1 3.78 2 3 3
0.3 0.77 0.46 0.43 0.64
2 39 57 47 4
0.51 6.36 5.75 8.48 1.57
and archaeal methanogens is plausible. Such a relationship has also been observed by Becker et al. [7] in 3-chlorbenzoateamended and 2-chlorophenol-amended methanogenic sediment mesocosms, where increases in Syntrophus-like rRNA were followed by increases in archaeal rRNA. Automated counting In this study, a recently developed automated counting routine [32,50] was modified for CARD-FISH in aquifer sediment samples. The combination of these two techniques allowed a reliable highthroughput analysis of microbial population dynamics in samples with high background fluorescence and, occasionally, a large proportion of very small cells with low FISH and/or DAPI signals. Conclusions The CARD-FISH protocol presented here enabled sedimentattached microbial cells to be recovered and quantified from push core samples of an aquifer that was highly contaminated with hydrocarbons. Controlled microwave irradiation during the permeabilization and hybridization procedures led to a substantial improvement of CARD-FISH detection in these samples. In combination with automated quantification, large numbers of samples from vertical and horizontal transects through the aquifer could be analyzed. Follow-up studies on population dynamics of selected microorganisms investigated at high spatial resolution could help shed light on the ecology of contaminant degraders. References [1] Albrechtsen, H.-J., Winding, A. (1992) Microbial biomass and activity in subsurface sediments from Vejen, Denmark. Microb. Ecol. 23, 303–317. [2] Albrechtsen, H.J., Christensen, T.H. (1994) Evidence for microbial iron reduction in a landfill leachate-polluted aquifer (Vejen, Denmark). Appl. Environ. Microbiol. 60, 3920–3925. [3] Amann, R.I., et al. (1990) Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919–1925. [4] Anneser, B., et al. (2008) High-resolution monitoring of biogeochemical gradients in a tar oil-contaminated aquifer. Appl. Geochem. 23, 1715–1730. [5] Anneser, B., et al. (2010) High resolution analysis of contaminated aquifer sediments and groundwater—what can be learned in terms of natural attenuation? Geomicrobiol. J. 27, 130–142. [6] Barra Caracciolo, A., Grenni, P., Cupo, C., Rossetti, S. (2005) In situ analysis of native microbial communities in complex samples with high particulate loads. FEMS Microbiol. Lett. 253, 55–58, doi:10.1016/j.femsle.2005.09.018. [7] Becker, J.G., Berardesco, G., Rittmann, B.E., Stahl, D.A. (2004) The role of syntrophic associations in sustaining anaerobic mineralization of chlorinated organic compounds. Environ. Health Perspect. 113. [8] Böckelmann, U., Szewzyk, U., Grohmann, E. (2003) A new enzymatic method for the detachment of particle associated soil bacteria. J. Microbiol. Methods 55, 201–211, doi:10.1016/s0167-7012(03)00144-1. [9] Brad, T., van Breukelen, B.M., Braster, M., van Straalen, N.M., Roling, W.F. (2008) Spatial heterogeneity in sediment-associated bacterial and eukaryotic communities in a landfill leachate-contaminated aquifer. FEMS Microbiol. Ecol. 65, 534–543, doi:10.1111/j.1574-6941.2008.00533.x, FEM533 [pii]. [10] Brar, S.K., et al. (2006) Bioremediation of hazardous wastes—a review. J. Hazard. Toxic Radioactive Waste 10, 59.
[11] Breuker, A., Koweker, G., Blazejak, A., Schippers, A. (2011) The deep biosphere in terrestrial sediments in the Chesapeake Bay area, Virginia, USA. Front Microbiol. 2, 156, doi:10.3389/fmicb.2011.00156. [12] Cavalca, L., Dell’Amico, E., Andreoni, V. (2004) Intrinsic bioremediability of an aromatic hydrocarbon-polluted groundwater: diversity of bacterial population and toluene monoxygenase genes. Appl. Microbiol. Biotechnol. 64, 576–587, doi:10.1007/s00253-003-1449-6. [13] Courtois, S., et al. (2001) Quantification of bacterial subgroups in soil: comparison of DNA extracted directly from soil or from cells previously released by density gradient centrifugation. Environ. Microbiol. 3, 431–439, doi:10.1046/j.1462-2920.2001.00208.x. [14] Daims, H., Bruhl, A., Amann, R., Schleifer, K.H., Wagner, M. (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, 434–444. [15] Fazi, S., Amalfitano, S., Pernthaler, J., Puddu, A. (2005) Bacterial communities associated with benthic organic matter in headwater stream microhabitats. Environ. Microbiol. 7, 1633–1640, doi:10.1111/j.1462-2920.2005.00857.x. [16] Ficker, M., Krastel, K., Orlicky, S., Edwards, E. (1999) Molecular characterization of a toluene-degrading methanogenic consortium. Appl. Environ. Microbiol. 65, 5576–5585. [17] Fike, D.A, Gammon, C.L., Ziebis, W., Orphan, V.J. (2008) Micron-scale mapping of sulfur cycling across the oxycline of a cyanobacterial mat: a paired nanoSIMS and CARD-FISH approach. ISME J. 2, 749–759, doi:10.1038/ismej.2008.39, ismej200839 [pii]. [18] Grenni, P., et al. (2009) A new fluorescent oligonucleotide probe for in situ detection of s-triazine-degrading Rhodococcus wratislaviensis in contaminated groundwater and soil samples. Water Res. 43, 2999–3008, doi:10.1016/j.watres.2009.04.022. [19] Griebler, C., Mindl, B., Slezak, D., Geiger-Kaiser, M. (2002) Distribution patterns of attached and suspended bacteria in pristine and contaminated shallow aquifers studied with an in situ sediment exposure microcosm. Aquat. Microbial. Ecol. 28, 117–129, doi:10.3354/ame028117. [20] Holm, P.E., Nielsen, P.H., Albrechtsen, H.J., Christensen, T.H. (1992) Importance of unattached bacteria and bacteria attached to sediment in determining potentials for degradation of xenobiotic organic contaminants in an aerobic aquifer. Appl. Environ. Microbiol. 58, 3020–3026. [21] Khammar, N., Malhautier, L., Degrange, V., Lensi, R., Fanlo, J.-L. (2004) Evaluation of dispersion methods for enumeration of microorganisms from peat and activated carbon biofilters treating volatile organic compounds. Chemosphere 54, 243–254, doi:10.1016/s0045-6535(03)00721-5. [22] Kleikemper, J., Pelz, O., Schroth, M.H., Zeyer, J. (2002) Sulfate-reducing bacterial community response to carbon source amendments in contaminated aquifer microcosms. FEMS Microbiol. Ecol. 42, 109–118, doi:10.1111/j.15746941.2002.tb01000.x, FEM109. [23] Kleikemper, J., et al. (2005) Activity and diversity of methanogens in a petroleum hydrocarbon-contaminated aquifer. Appl. Environ. Microbiol. 71, 149–158, doi:10.1128/aem.71.1.149-158.2005. [24] Leive, L. (1968) Studies on the permeability change produced in coliform bacteria by ethylenediaminetetraacetate. J. Biol. Chem. 243, 2373–2380. [25] Lindahl, V., Bakken, L.R. (1995) Evaluation of methods for extraction of bacteria from soil. FEMS Microbiol. Ecol. 16, 135–142, doi:10.1111/j.15746941.1995.tb00277.x. [26] Lu, Y.-J., Cheng, S.-J., Ning, J., Dong, X.-Y. (1997) An accelerated in situ hybridization procedure using microwave irradiation. Chromosome Res. 5, 147–149, doi:10.1023/a:1018478410547. [27] Madsen, E.L., Ghiorse, W.C. (1993) In: Ford, T.E. (Ed.), Aquatic Microbiology—An Ecological Approach, Blackwell Scientific Publication, pp. 167–213. [28] Miura, Y., Okabe, S. (2008) Quantification of cell specific uptake activity of microbial products by uncultured Chloroflexi by microautoradiography combined with fluorescence in situ hybridization. Environ. Sci. Technol. 42, 7380–7386. [29] Müller, S., Vogt, C., Laube, M., Harms, H., Kleinsteuber, S. (2009) Community dynamics within a bacterial consortium during growth on toluene under sulfate-reducing conditions. FEMS Microbiol. Ecol. 70, 586–596, doi:10.1111/j.1574-6941.2009.00768.x. [30] Perelo, L.W. (2010) In situ and bioremediation of organic pollutants in aquatic sediments. J. Hazard. Mater. 177, 81–89.
532
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
[31] Pernthaler, A., Pernthaler, J., Amann, R. (2004) In: Kowalchuk, G., et al. (Eds.), Molecular Microbial Ecology Manual, Kluwer Academic Press. [32] Pernthaler, J., Pernthaler, A., Amann, R. (2003) Automated enumeration of groups of marine picoplankton after fluorescence in situ hybridization. Appl. Environ. Microbiol. 69, 2631–2637. [33] Pombo, S.A, Kleikemper, J., Schroth, M.H., Zeyer, J. (2005) Field-scale isotopic labeling of phospholipid fatty acids from acetate-degrading sulfate-reducing bacteria. FEMS Microbiol. Ecol. 51, 197–207, doi:10.1016/j.femsec.2004.08.010. [34] Pombo, S.A., Pelz, O., Schroth, M.H., Zeyer, J. (2002) Field-scale 13Clabeling of phospholipid fatty acids (PLFA) and dissolved inorganic carbon: tracing acetate assimilation and mineralization in a petroleum hydrocarboncontaminated aquifer. FEMS Microbiol. Ecol. 41, 259–267, doi:10.1111/j.15746941.2002.tb00987.x. [35] Richardson, R.E., James, C.A., Bhupathiraju, V.K., Alvarez-Cohen, L. (2002) Microbial activity in soils following steam treatment. Biodegradation 13, 285–295. [36] Rogers, S.W., Moorman, T.B., Ong, S.K. (2007) Fluorescent in situ hybridization and microautoradiography applied to ecophysiology in soil. Soil Sci. Soc. Am. J. 71, 620–631. [37] Rogers, S.W., Ong, S.K., Moorman, T.B. (2007) Mineralization of PAHs in coal-tar impacted aquifer sediments and associated microbial community structure investigated with FISH. Chemosphere 69, 1563–1573, doi:10.1016/j.chemosphere.2007.05.058, S0045-6535(07)00716-3 [pii]. [38] Rossetti, S., Aulenta, F., Majone, M., Crocetti, G., Tandoi, V. (2008) Structure analysis and performance of a microbial community from a contaminated aquifer involved in the complete reductive dechlorination of 1,1,2,2-tetrachloroethane to ethene. Biotechnol. Bioeng. 100, 240–249, doi:10.1002/bit.21776. [39] Stahl, D.A., Amann, R. (1991) In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics, Wiley & Sons Ltd, pp. 205–248. [40] Suarez-Suarez, A., et al. (2011) Response of sulfate-reducing bacteria to an artificial oil-spill in a coastal marine sediment. Environ. Microbiol. 13, 1488–1499, doi:10.1111/j.1462-2920.2011.02451.x.
[41] Tischer, K., Kleinsteuber, S., Schleinitz, K., Fetzer, I., Spott, O., Stange, F., Lohse, U., Gerling, S., Schmidt, C., Harms, H., Wendeberg, A., (in press). Microbial communities along biogeochemical gradients in a hydrocarbon contaminated aquifer. [42] Tischer, K., Wendeberg A., (in press). Microbial population dynamics along biogeochemical gradients in a hydrocarbon contaminated aquifer as revealed by CARD-FISH. [43] Vaara, M. (1992) Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56, 395–411. [44] Weise, A., Liehr, T., Claussen, U., Halbhuber, K.-J. (2005) Increased efficiency of fluorescence in situ hybridization (FISH) using the microwave. J. Histochem. Cytochem. 53, 1301–1303, doi:10.1369/jhc.5B6632.2005. [45] Wendeberg, A., Zielinski, F.U., Borowski, C., Dubilier, N. (2011) Expression patterns of mRNAs for methanotrophy and thiotrophy in symbionts of the hydrothermal vent mussel Bathymodiolus puteoserpentis. ISME J., doi:10.1038/ismej.2011.81. [46] Witzig, M., Boguhn, J., Kleinsteuber, S., Fetzer, I., Rodehutscord, M. (2010) Influence of the maize silage to grass silage ratio and feed particle size of rations for ruminants on the community structure of ruminal Firmicutes in vitro. J. Appl. Microbiol. 109, 1998–2010, doi:10.1111/j.1365-2672.2010.04829.x, JAM4829 [pii]. [47] Zeder, M. (2009) Development of methods and technology for automated high-throughput multi-parameter analysis of single cells by fluorescence microscopy. PhD thesis, Zürich. [48] Zeder, M., Ellrott, A., Amann, R. (2011) Automated sample area definition for high-throughput microscopy. Cytometry A 79A, 306–310, doi:10.1002/cyto.a.21034. [49] Zeder, M., Kohler, E., Pernthaler, J. (2010) Automated quality assessment of autonomously acquired microscopic images of fluorescently stained bacteria. Cytometry A 77A, 76–85, doi:10.1002/cyto.a.20810. [50] Zeder, M., Pernthaler, J. (2009) Multispot live-image autofocusing for highthroughput microscopy of fluorescently stained bacteria. Cytometry A 75A, 781–788, doi:10.1002/cyto.a.20770.
Contents lists available at SciVerse ScienceDirect
Systematic and Applied Microbiology journal homepage: www.elsevier.de/syapm
Fluorescence in situ hybridization (CARD-FISH) of microorganisms in hydrocarbon contaminated aquifer sediment samples Karolin Tischer a , Michael Zeder a,b,c , Rebecca Klug a , Jakob Pernthaler b , Martha Schattenhofer a , Hauke Harms a , Annelie Wendeberg a,∗ a
Helmholtz Centre for Environmental Research, Department of Environmental Microbiology, Permoserstr. 15, 04318 Leipzig, Germany University of Zurich, Department of Limnology, Seestr. 187, CH-8802 Kilchberg, Switzerland c Max Planck Institute for Marine Microbiology, Celciusstr. 1, 28359 Bremen, Germany b
a r t i c l e
i n f o
Article history: Received 30 September 2011 Received in revised form 11 January 2012 Accepted 24 January 2012 Keywords: CARD-FISH Hydrocarbon contamination Aquifer
a b s t r a c t Groundwater ecosystems are the most important sources of drinking water worldwide but they are threatened by contamination and overexploitation. Petroleum spills account for the most common source of contamination and the high carbon load results in anoxia and steep geochemical gradients. Microbes play a major role in the transformation of petroleum hydrocarbons into less toxic substances. To investigate microbial populations at the single cell level, fluorescence in situ hybridization (FISH) is now a well-established technique. Recently, however, catalyzed reporter deposition (CARD)-FISH has been introduced for the detection of microbes from oligotrophic environments. Nevertheless, petroleum contaminated aquifers present a worst case scenario for FISH techniques due to the combination of high background fluorescence of hydrocarbons and the presence of small microbial cells caused by the low turnover rates characteristic of groundwater ecosystems. It is therefore not surprising that studies of microorganisms from such sites are mostly based on cultivation techniques, fingerprinting, and amplicon sequencing. However, to reveal the population dynamics and interspecies relationships of the key participants of contaminant degradation, FISH is an indispensable tool. In this study, a protocol for FISH was developed in combination with cell quantification using an automated counting microscope. The protocol includes the separation and purification of microbial cells from sediment particles, cell permeabilization and, finally, CARD-FISH in a microwave oven. As a proof of principle, the distribution of Archaea and Bacteria was shown in 60 sediment samples taken across the contaminant plume of an aquifer (Leuna, Germany), which has been heavily contaminated with several ten-thousand tonnes of petroleum hydrocarbons since World War II. © 2012 Elsevier GmbH. All rights reserved.
Introduction Oil spills represent a major threat to worldwide aquatic and terrestrial ecosystems. It is widely acknowledged that natural microbial communities can transform contaminants into less toxic substances, and strategies exploiting microorganisms have been developed to clean up contaminated sites [10,30]. However, our understanding of the physiology and ecology of the natural microbial communities found at polluted sites is limited and bioremediation approaches are often based on trial and error. Most studies on the microbial diversity of petroleum contaminated sites are based on cultivation techniques, fingerprinting, and amplicon sequencing [12,29]. To investigate population dynamics
∗ Corresponding author. E-mail addresses: [email protected], [email protected] (A. Wendeberg). 0723-2020/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. doi:10.1016/j.syapm.2012.01.004
and interspecies relationships at the single cell level, fluorescence in situ hybridization (FISH) of rRNA is now a widely used technique [11,40,46]. FISH can be combined with other methods, such as microautoradiography [28], mRNA FISH [45] and nanoSIMS [17], to study physiological traits of selected microbial populations and answer the question of “who is doing what” in a complex environment. However, only five studies on microbial communities in aquifer sediments have used FISH to date [18,23,33,35,37]. Although FISH has been more commonly applied to groundwater samples [16,22,33,34], more than 90% of the microbial biomass can be found attached to sediment particles [5,19], and it often exhibits higher activities than the planktonic community [1,2,20,27]. Therefore, detailed studies of aquifer sediments are more ecologically relevant than studies of groundwater samples, especially when evaluating the potential for natural attenuation [9]. The aim of this study was to develop a catalyzed reporter deposition (CARD)-FISH protocol for the detection of microorganisms in hydrocarbon contaminated aquifer sediment samples.
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
527
Fig. 1. Schematic of the sampling site showing the position of the Leuna aquifer in Germany and the top view of the Leuna industrial area with the hydrocarbon plume (light grey) and source (dark grey), the positions of the cores taken (C1–C8) and the cut-off trench (black line) of the pump-and-treat unit (not to scale).
For high-throughput quantification of CARD-FISH stained cells, a protocol was developed that could produce high quality cell preparations for automated FISH counting. The protocol included a cell detachment step using Tris–EDTA buffer and microwaving, detergent treatment and sonication, as well as a cell purification step using Nycodenz density gradient centrifugation. The population dynamics of Bacteria and Archaea were shown in 60 sediment samples across the petroleum hydrocarbon contaminant plume of the Leuna aquifer (Germany). Materials and methods Site description and sampling procedure The Leuna megasite has been a centre for chemical production since the beginning of the 20th century. The site has been associated with ammonia synthesis (since 1917) and coal hydrogenation (since 1927) that have led to heavy contamination due to bombing in World War II. In the late 1970s, production of methyl tertiary butyl ether (MTBE) and associated leakages led to additional subsurface contaminations. Today, large quantities of various gasoline components, especially benzene, toluene, ethyl benzene, xylene isomers (BTEX) and MTBE, reside at the site and create a contaminant plume of approximately 300 m in width and 500 m in length that reaches the cut-off trench of a pump-and-treat unit (Fig. 1). Direct-push cores were taken along the plume (C1–C5) and outside the plume (C8) in April 2008, September 2009 and February 2010 at the capillary fringe. The sediment cores were immediately transferred into an anoxic glove-box where all sub-sampling and oxygen measurements took place. To measure the O2 concentrations within each core, small holes at 4 cm intervals where poked into the plastic liners of the cores, and micro-optodes (PreSens, Regensburg, Germany) were inserted approximately 5 mm into the sediment. Cores were then sectioned into 4 cm thick sections, subsamples of each section were formaldehyde fixed (2% volume/volume [v/v] final concentration) and kept at 4 ◦ C for 24 h. The samples were washed twice with a 1:1 mix of 1 × PBS [137 mM NaCl, 2.7 mM KCl, 4 mM Na2 HPO4 , 2 mM KH2 PO4 , pH 7.6] and absolute ethanol by pelleting and resuspension, and then stored in absolute ethanol at −80 ◦ C. Sediment dry weight was determined by drying the samples at 105 ◦ C, and then samples were allowed
to cool to room temperature in an exsiccator and they were subsequently weighed. Cell separation Two-hundred microliters of sediment sample was mixed with 700 L TE buffer [10 mM TrisHCl, 5 mM EDTA, pH 9.0] and 100 L of Na-pyrophosphate [0.1 M] in a 1.5 mL reaction tube. This mixture was then placed into a water bath in a histological microwave oven (Microwave Research and Application Inc., Laurel, MD, USA) and heated to 55 ◦ C for 5 min at 200 W. After cooling the samples to room temperature (RT), 1 L of Tween 80 was added and each sample was agitated (level 3, 1060 rpm) on a vortexer (Vortex-Genie 2, Carl Roth GmbH, Karlsruhe, Germany) for 15 min at RT. Subsequently, the sample was sonicated on ice (Sonifier Model 250, Branson, Danbury, USA) with 10 pulses at 53 W. To separate dislodged cells from sediment particles, the samples were transferred into a 50 mL Falcon tube and mixed with 22.5 mL PBS and 2.5 mL Na-pyrophosphate [0.1 M]. Finally, a layer of Nycodenz (AXIS-SHIELD PoC, Oslo, Norway) solution [density 1.3 g mL−1 ; 60% (w/v), in MilliQ water] was carefully placed at the bottom of the Falcon tube using a syringe with a long needle, then the sediment was pelleted by centrifugation overnight at 4000 rpm (centrifuge 5810R, Eppendorf, Hamburg, Germany) and 4 ◦ C using a swing-out rotor (A-4-62, Eppendorf, Hamburg, Germany). After centrifugation, both the supernatant and the Nycodenz layer were collected, mixed, and filtered on white polycarbonate filters (type GTTP, pore size 0.2 m, diameter 47 mm, Millipore, Eschborn, Germany). The filter was then washed twice with autoclaved deionized water, air-dried and stored at −20 ◦ C until further processing. Cell permeabilization To prevent cell loss during the permeabilization procedure, filters were embedded in agarose. The filters were placed with the sample facing down into 200 L low-gelling-point agarose (0.1% [w/v] in MilliQ water) on Parafilm and left to dry at 35 ◦ C in a hybridization oven. Then, the filters were wetted with 50% ethanol and gently peeled off the Parafilm using forceps. The filters were allowed to air dry before sectioning, and filter sections were placed into a 1.5 mL reaction tube containing 1 mL of 1 × TE
528
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
Fig. 2. Effect of two cell dislodging procedures, (a) vortexing and (b) sonication, on total cell numbers of two selected sediment samples.
[10 mM TrisHCl, 5 mM EDTA, pH 9.0] and they were permeabilized by microwaving in a pre-heated waterbath at 65 ◦ C for 8 min at 1000 W (100% power output). The reaction tubes were then allowed to cool for 5 min at RT. Next, the cells were post-fixed in 990 L of 1 × TE and 120 L formaldehyde (37%) for 5 min at RT to stabilize the permeabilized cells for subsequent hybridization. Filters were then washed with 1 × TE, and endogenous peroxidases were inactivated with 0.1% H2 O2 in 1 × TE for 2 min at RT. The filters were then washed twice, as mentioned above, and transferred to the hybridization buffer (see below). CARD-FISH Horseradish peroxidise (HRP)-labelled probes (Biomers, Ulm, Germany) targeting Bacteria (EUB338I-III, [3,14]) and Archaea (ARCH915, [39]) were used for CARD-FISH. The hybridization was carried out in a 1.5 mL reaction tube containing up to 20 filter pieces, 1000 L hybridization buffer [0.9 M NaCl, 20 mM TrisHCl (pH 8), 10% dextran sulfate, 0.02% sodium dodecyl sulphate (SDS), 1% blocking reagent (Roche, Mannheim, Germany), 35% formamide (Fluka, Germany) and MilliQ water] and 3.3 L HRP probe working solution [50 ng L−1 ]. The tubes were placed in a pre-warmed water bath and microwaved in a histological microwave oven at 46 ◦ C for 40 min at 500 W. To wash off excess probe and to reactivate the HRP, filter sections were placed in 50 mL of 1 × PBS for 15 min at RT. Subsequently, the CARD reaction was undertaken by placing the filters in 1 mL amplification buffer [1 × PBS (pH 7.6), 2 M NaCl, 10% dextran sulphate, 0.1% blocking reagent (Roche, Mannheim, Germany) and MilliQ water], freshly amended with 0.0015% H2 O2 and 1 g mL−1 fluorescein-labelled tyramide (custom labelled; see [31]) for 15 min at 46 ◦ C in a conventional hybridization oven. Finally, filters were washed five times with deionized water, dehydrated in increasing ethanol concentrations (50%, 70%, and 100%) and allowed to air dry. For microscopic examination, filters were embedded in a DAPI amended mountant (see below). For dual hybridization of Archaea and Bacteria, two consecutive CARD-FISH stainings were carried out with an HRP inactivation step between the two (0.1% H2 O2 in 1 × TE for 2 min at RT), and Alexalabelled tyramides (custom labelled; [31]) were used for the CARD reaction. Total cell counts A general DNA stain (4 ,6-diamidino-2-phenylindole, DAPI) was added to the mountant and was used to determine the total cell
numbers. The mountant contained 1000 L Citifluor (Citifluor AF2, Citifluor Ltd., London, UK), 100 L of 1 × PBS and 1 L of a DAPI stock solution (1 mg mL−1 ). Automated counting FISH preparations were evaluated by automated microscopy [32,47–49] on an Axio Observer.Z1 (Carl Zeiss, Jena, Germany) using a 63 × Plan-Apochromat objective (NA = 1.4) and the filter sets 01 for DAPI and 38HE for FITC. The inverted microscope featured a motorized stage for four slides and an HXP 120 lamp (Carl Zeiss, Germany) for fluorescence illumination. Greyscale images were recorded on a digital camera (AxioCam MRm; Carl Zeiss). Imaging involved three steps: (i) whole slide overview images were acquired to localize the samples to be imaged, and to create a list of coordinates [48]; (ii) autonomous multi channel image acquisition was performed using a custom made Axio Vision 4.8 (Carl Zeiss, Germany) VBA macro (M. Zeder, unpublished); and (iii) images were checked for quality [50] and analyzed using the ACMEtool (available for download at www.technobiology.ch) in order to obtain and revise the bacterial cell counts. Automated counting results were corroborated by manual counting of 10 randomly chosen samples (r2 = 0.88). Results and discussion Sample preparation High concentrations of petroleum hydrocarbons in sediment samples (e.g. up to 2 mg g−1 dryweight benzene; Tischer et al., in prep., A [41]) caused unacceptably high levels of background fluorescence with both DAPI (a general DNA stain), and CARD-FISH, making it impossible to discern individual cells. Therefore, it was decided to separate and purify microbial cells from sediment particles for subsequent CARD-FISH and enumeration. Microorganisms are attached to soil particles by a variety of mechanisms, such as electrostatic interactions, binding through exopolysaccharides or proteins [25]. To detach cells from sediment particles, the sediment sample was therefore first microwaved in Tris–EDTA buffer and pyrophosphate in a histological microwave oven at 55 ◦ C. The slurry was then amended with the non-ionic surfactant Tween 80 and vortexed, followed by sonication. A time series for both steps was conducted to find the optimal balance between cell detachment, cell disruption, and increasing amounts of silt particles in the cell fraction, which caused high levels of background fluorescence (Fig. 2). Although 30 min vortexing resulted in
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
Fig. 3. Effect of 15 min and 30 min vortexing performed with all sediment samples. Groundwater table at 0 m.
529
Fig. 4. CARD-FISH detection rates using bacterial (EUB338 I–III) and archaeal (Arch915) probes. The sample used for dual CARD-FISH in Fig. 5 is indicated with an arrow.
530
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
the highest cell numbers for the two samples selected for the time series (Fig. 2a), in most of the 60 samples 15 min vortexing resulted in the highest cell numbers (Fig. 3) and was thus chosen for the final protocol. Additionally, it was found that vortexing for longer than 15 min caused higher levels of background fluorescence due to increasing amounts of silt in the cell fraction. Sonication to detach cells from soil particles is commonly used [8,13,15,21,25]. For the sonication treatment, 10 pulses of 1 s each were found to produce the highest numbers of dislodged cells (Fig. 2b), since sonication for longer than 10 pulses led to a decrease of cell numbers (Fig. 2), which was likely caused by disruption or by dislodging silt particles, thereby causing background fluorescence [6]. To separate cells subsequently from sediment particles, the slurry was subjected to Nycodenz density gradient centrifugation, as described by Barra Caracciolo et al. [6]. After density gradient centrifugation, cells could be detected in all three density layers (data not shown), in accordance with a previous report [6]. An attempt was made to collect all detached cells and, therefore, the entire supernatant was filtered onto membrane filters. The recovery of the microbial cells was comparable to previous studies [6,38] with an efficiency of 73% (average, SD = 12, tested on 8 different sediment samples with 3 replicate counts each). Additionally, a fingerprinting technique (terminal restriction length polymorphism, T-RFLP) was used to estimate a potential shift in community composition caused by cell dislodging and density gradient centrifugation. However, the community composition of both the pellet and the supernatant was found to be comparable (data not shown). Cell permeabilization Various cell permeabilization strategies to improve CARD-FISH detection rates were tested. Firstly, a time and temperature series of heating in Tris–EDTA buffer was performed and it was found that 8 min at 65 ◦ C resulted in the highest CARD-FISH detection rates (data not shown). It has been suggested that Tris binds to lipopolysaccharides and replaces stabilizing Ca2+ and Mg2+ , and therefore reduces the interaction between lipopolysaccharide molecules, while EDTA removes the divalent cations from their binding sites by chelation [43]. This results in the permeabilization of the outer membrane by release of a significant proportion of lipopolysaccharide from the cells [24]. Curiously, it was found that the TE permeabilization step in the microwave only worked in combination with hybridization. Hybridization in a conventional oven following TE permeabilization resulted in CARD-FISH signals below the detection limit. Furthermore, lysozyme, achromopeptidase, and proteinase K were tested for permeabilizing archaeal and bacterial cell walls. Although TE permeabilization and subsequent lysozyme and achromopeptidase treatment showed the highest detection rates for two samples (Table 1), in most of the 60 samples TE permeabilization resulted in a higher detection rate (Fig. 4). Additionally, cell morphologies were preserved better than after enzymatic treatment. CARD-FISH Filter sections were hybridized without delay, since storage after the permeabilization procedure sometimes led to lower detection rates (data not shown). Hybridization in a histological microwave oven increased both signal intensity and detection rate dramatically (2- to 20-fold), while decreasing the time needed for hybridization, when compared to hybridization in a conventional hybridization oven (Table 1). For more than two decades, it has been known that microwave irradiation can accelerate physical and chemical reactions [26]. When used for FISH, microwaves transfer high energy to the probes and can promote probe penetration into
Fig. 5. Dual CARD-FISH of core 3 12 cm: Archaea (Arch915, green), Bacteria (EUB338 I–III, red), and a general DNA stain (DAPI, blue). Size bar represents 10 m.
the cells [26]. Weise et al. [44] speculated that microwave irradiation may act as a steric modifier, allowing the DNA probe to find the target in a more efficient and faster way than in a normal FISH assay which depends on diffusion of probes to the target. The detection rates in Leuna aquifer sediment samples ranged between 15% and 100% for Bacteria and Archaea together, while unspecific signals (hybridized with NON338) were below 1% (data not shown). The highest detection rates were found in samples with lower concentrations of hydrocarbons: in the deepest samples of core 3 (up to 100%) and core 5 (up to 70%) and the whole of core 8 (between 36 and 60%), the latter originating from an uncontaminated part of the aquifer (Fig. 1). The lowest detection rates were found in samples that were saturated with hydrocarbons, originating either from the source zone (cores 1 and 2), or from the area around the water table, where the oil phase was situated (upper parts of core 3 and core 5; Fig. 1). Detection rates in core 2 ranged between 13 and 19%, but core 1 had to be omitted from FISH analysis, since cell numbers were below the detection limit. Low CARD-FISH detection rates in the highly contaminated zones were most likely due to toxic levels of petroleum hydrocarbons, in addition to an insufficient amount of bioavailable water in the oil phase, which may have resulted in low activity of microorganisms and thus in low ribosome copy numbers per cell. It was concluded that microorganisms in that zone were not actively taking part in the degradation of petroleum hydrocarbons, which corresponded to the plume-fringe concept, where highest activities of contaminant degradation can be found at the fringes of contaminant plumes [4]. To date, of the five studies that have used FISH on microbial communities in aquifer sediments [18,23,33,35,37], only one of them has achieved a detection rate of 96% for the eubacterial probe, but only one sample was analyzed [18]. In the other four studies, the detection rates reported were relatively low with 23% (Kleikemper et al. 2005), 33% [35], 72% [Pombo et al. 2005], and 17–75% as the sum of Eubacteria and Archaea [36,37]. Probe detection rates in this study of up to 100% were obtained for Eubacteria and Archaea together, which are considerably higher than previously reported. It was observed that the abundances of Bacteria and Archaea were remarkably similar in cores 2, 3 and 5. The specificity of these FISH results was validated by performing a dual CARDFISH staining for Archaea and Bacteria (Fig. 5). Cores 2, 3 and 5 were from the methanogenic part of the plume (Tischer et al., in prep, B [42]), where a syntrophic lifestyle is more favorable and close coupling between bacterial contaminant degraders
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
531
Table 1 Detection rates of probes targeting Bacteria (Eub338 I-III) and Archaea (Arch915) comparing various cell dislodging, permeabilization, and hybridization procedures, performed on two selected sediment samples (core 3 4 cm and core 3 56 cm). Abbreviations: TE, microwaving in TE buffer; Lys, lysozyme digestions; Achr, achromopeptidase digestion; PK , proteinase K digestion; MW, hybridization in a microwave oven; Conv., hybridization in a conventional oven; ON, overnight. Dislodging time [min]
Permeabilization
Hybridization
Bacteria
Archaea
C3–4
15 15 30 30 30
Lys, Achr TE TE TE, Lys, Achr TE, Lys, Achr, PK
Conv., ON MW, 45 min MW, 45 min MW, 45 min MW, 45 min
C3–56
C3–4
C3–56
%DAPI
SD
%DAPI
SD
%DAPI
SD
%DAPI
SD
2 5 6 7 4
0.28 0.95 0.62 2.44 2
14 36 44 43 13
2.87 2.31 2.88 1.62 5
1 3.78 2 3 3
0.3 0.77 0.46 0.43 0.64
2 39 57 47 4
0.51 6.36 5.75 8.48 1.57
and archaeal methanogens is plausible. Such a relationship has also been observed by Becker et al. [7] in 3-chlorbenzoateamended and 2-chlorophenol-amended methanogenic sediment mesocosms, where increases in Syntrophus-like rRNA were followed by increases in archaeal rRNA. Automated counting In this study, a recently developed automated counting routine [32,50] was modified for CARD-FISH in aquifer sediment samples. The combination of these two techniques allowed a reliable highthroughput analysis of microbial population dynamics in samples with high background fluorescence and, occasionally, a large proportion of very small cells with low FISH and/or DAPI signals. Conclusions The CARD-FISH protocol presented here enabled sedimentattached microbial cells to be recovered and quantified from push core samples of an aquifer that was highly contaminated with hydrocarbons. Controlled microwave irradiation during the permeabilization and hybridization procedures led to a substantial improvement of CARD-FISH detection in these samples. In combination with automated quantification, large numbers of samples from vertical and horizontal transects through the aquifer could be analyzed. Follow-up studies on population dynamics of selected microorganisms investigated at high spatial resolution could help shed light on the ecology of contaminant degraders. References [1] Albrechtsen, H.-J., Winding, A. (1992) Microbial biomass and activity in subsurface sediments from Vejen, Denmark. Microb. Ecol. 23, 303–317. [2] Albrechtsen, H.J., Christensen, T.H. (1994) Evidence for microbial iron reduction in a landfill leachate-polluted aquifer (Vejen, Denmark). Appl. Environ. Microbiol. 60, 3920–3925. [3] Amann, R.I., et al. (1990) Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919–1925. [4] Anneser, B., et al. (2008) High-resolution monitoring of biogeochemical gradients in a tar oil-contaminated aquifer. Appl. Geochem. 23, 1715–1730. [5] Anneser, B., et al. (2010) High resolution analysis of contaminated aquifer sediments and groundwater—what can be learned in terms of natural attenuation? Geomicrobiol. J. 27, 130–142. [6] Barra Caracciolo, A., Grenni, P., Cupo, C., Rossetti, S. (2005) In situ analysis of native microbial communities in complex samples with high particulate loads. FEMS Microbiol. Lett. 253, 55–58, doi:10.1016/j.femsle.2005.09.018. [7] Becker, J.G., Berardesco, G., Rittmann, B.E., Stahl, D.A. (2004) The role of syntrophic associations in sustaining anaerobic mineralization of chlorinated organic compounds. Environ. Health Perspect. 113. [8] Böckelmann, U., Szewzyk, U., Grohmann, E. (2003) A new enzymatic method for the detachment of particle associated soil bacteria. J. Microbiol. Methods 55, 201–211, doi:10.1016/s0167-7012(03)00144-1. [9] Brad, T., van Breukelen, B.M., Braster, M., van Straalen, N.M., Roling, W.F. (2008) Spatial heterogeneity in sediment-associated bacterial and eukaryotic communities in a landfill leachate-contaminated aquifer. FEMS Microbiol. Ecol. 65, 534–543, doi:10.1111/j.1574-6941.2008.00533.x, FEM533 [pii]. [10] Brar, S.K., et al. (2006) Bioremediation of hazardous wastes—a review. J. Hazard. Toxic Radioactive Waste 10, 59.
[11] Breuker, A., Koweker, G., Blazejak, A., Schippers, A. (2011) The deep biosphere in terrestrial sediments in the Chesapeake Bay area, Virginia, USA. Front Microbiol. 2, 156, doi:10.3389/fmicb.2011.00156. [12] Cavalca, L., Dell’Amico, E., Andreoni, V. (2004) Intrinsic bioremediability of an aromatic hydrocarbon-polluted groundwater: diversity of bacterial population and toluene monoxygenase genes. Appl. Microbiol. Biotechnol. 64, 576–587, doi:10.1007/s00253-003-1449-6. [13] Courtois, S., et al. (2001) Quantification of bacterial subgroups in soil: comparison of DNA extracted directly from soil or from cells previously released by density gradient centrifugation. Environ. Microbiol. 3, 431–439, doi:10.1046/j.1462-2920.2001.00208.x. [14] Daims, H., Bruhl, A., Amann, R., Schleifer, K.H., Wagner, M. (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, 434–444. [15] Fazi, S., Amalfitano, S., Pernthaler, J., Puddu, A. (2005) Bacterial communities associated with benthic organic matter in headwater stream microhabitats. Environ. Microbiol. 7, 1633–1640, doi:10.1111/j.1462-2920.2005.00857.x. [16] Ficker, M., Krastel, K., Orlicky, S., Edwards, E. (1999) Molecular characterization of a toluene-degrading methanogenic consortium. Appl. Environ. Microbiol. 65, 5576–5585. [17] Fike, D.A, Gammon, C.L., Ziebis, W., Orphan, V.J. (2008) Micron-scale mapping of sulfur cycling across the oxycline of a cyanobacterial mat: a paired nanoSIMS and CARD-FISH approach. ISME J. 2, 749–759, doi:10.1038/ismej.2008.39, ismej200839 [pii]. [18] Grenni, P., et al. (2009) A new fluorescent oligonucleotide probe for in situ detection of s-triazine-degrading Rhodococcus wratislaviensis in contaminated groundwater and soil samples. Water Res. 43, 2999–3008, doi:10.1016/j.watres.2009.04.022. [19] Griebler, C., Mindl, B., Slezak, D., Geiger-Kaiser, M. (2002) Distribution patterns of attached and suspended bacteria in pristine and contaminated shallow aquifers studied with an in situ sediment exposure microcosm. Aquat. Microbial. Ecol. 28, 117–129, doi:10.3354/ame028117. [20] Holm, P.E., Nielsen, P.H., Albrechtsen, H.J., Christensen, T.H. (1992) Importance of unattached bacteria and bacteria attached to sediment in determining potentials for degradation of xenobiotic organic contaminants in an aerobic aquifer. Appl. Environ. Microbiol. 58, 3020–3026. [21] Khammar, N., Malhautier, L., Degrange, V., Lensi, R., Fanlo, J.-L. (2004) Evaluation of dispersion methods for enumeration of microorganisms from peat and activated carbon biofilters treating volatile organic compounds. Chemosphere 54, 243–254, doi:10.1016/s0045-6535(03)00721-5. [22] Kleikemper, J., Pelz, O., Schroth, M.H., Zeyer, J. (2002) Sulfate-reducing bacterial community response to carbon source amendments in contaminated aquifer microcosms. FEMS Microbiol. Ecol. 42, 109–118, doi:10.1111/j.15746941.2002.tb01000.x, FEM109. [23] Kleikemper, J., et al. (2005) Activity and diversity of methanogens in a petroleum hydrocarbon-contaminated aquifer. Appl. Environ. Microbiol. 71, 149–158, doi:10.1128/aem.71.1.149-158.2005. [24] Leive, L. (1968) Studies on the permeability change produced in coliform bacteria by ethylenediaminetetraacetate. J. Biol. Chem. 243, 2373–2380. [25] Lindahl, V., Bakken, L.R. (1995) Evaluation of methods for extraction of bacteria from soil. FEMS Microbiol. Ecol. 16, 135–142, doi:10.1111/j.15746941.1995.tb00277.x. [26] Lu, Y.-J., Cheng, S.-J., Ning, J., Dong, X.-Y. (1997) An accelerated in situ hybridization procedure using microwave irradiation. Chromosome Res. 5, 147–149, doi:10.1023/a:1018478410547. [27] Madsen, E.L., Ghiorse, W.C. (1993) In: Ford, T.E. (Ed.), Aquatic Microbiology—An Ecological Approach, Blackwell Scientific Publication, pp. 167–213. [28] Miura, Y., Okabe, S. (2008) Quantification of cell specific uptake activity of microbial products by uncultured Chloroflexi by microautoradiography combined with fluorescence in situ hybridization. Environ. Sci. Technol. 42, 7380–7386. [29] Müller, S., Vogt, C., Laube, M., Harms, H., Kleinsteuber, S. (2009) Community dynamics within a bacterial consortium during growth on toluene under sulfate-reducing conditions. FEMS Microbiol. Ecol. 70, 586–596, doi:10.1111/j.1574-6941.2009.00768.x. [30] Perelo, L.W. (2010) In situ and bioremediation of organic pollutants in aquatic sediments. J. Hazard. Mater. 177, 81–89.
532
K. Tischer et al. / Systematic and Applied Microbiology 35 (2012) 526–532
[31] Pernthaler, A., Pernthaler, J., Amann, R. (2004) In: Kowalchuk, G., et al. (Eds.), Molecular Microbial Ecology Manual, Kluwer Academic Press. [32] Pernthaler, J., Pernthaler, A., Amann, R. (2003) Automated enumeration of groups of marine picoplankton after fluorescence in situ hybridization. Appl. Environ. Microbiol. 69, 2631–2637. [33] Pombo, S.A, Kleikemper, J., Schroth, M.H., Zeyer, J. (2005) Field-scale isotopic labeling of phospholipid fatty acids from acetate-degrading sulfate-reducing bacteria. FEMS Microbiol. Ecol. 51, 197–207, doi:10.1016/j.femsec.2004.08.010. [34] Pombo, S.A., Pelz, O., Schroth, M.H., Zeyer, J. (2002) Field-scale 13Clabeling of phospholipid fatty acids (PLFA) and dissolved inorganic carbon: tracing acetate assimilation and mineralization in a petroleum hydrocarboncontaminated aquifer. FEMS Microbiol. Ecol. 41, 259–267, doi:10.1111/j.15746941.2002.tb00987.x. [35] Richardson, R.E., James, C.A., Bhupathiraju, V.K., Alvarez-Cohen, L. (2002) Microbial activity in soils following steam treatment. Biodegradation 13, 285–295. [36] Rogers, S.W., Moorman, T.B., Ong, S.K. (2007) Fluorescent in situ hybridization and microautoradiography applied to ecophysiology in soil. Soil Sci. Soc. Am. J. 71, 620–631. [37] Rogers, S.W., Ong, S.K., Moorman, T.B. (2007) Mineralization of PAHs in coal-tar impacted aquifer sediments and associated microbial community structure investigated with FISH. Chemosphere 69, 1563–1573, doi:10.1016/j.chemosphere.2007.05.058, S0045-6535(07)00716-3 [pii]. [38] Rossetti, S., Aulenta, F., Majone, M., Crocetti, G., Tandoi, V. (2008) Structure analysis and performance of a microbial community from a contaminated aquifer involved in the complete reductive dechlorination of 1,1,2,2-tetrachloroethane to ethene. Biotechnol. Bioeng. 100, 240–249, doi:10.1002/bit.21776. [39] Stahl, D.A., Amann, R. (1991) In: Stackebrandt, E., Goodfellow, M. (Eds.), Nucleic Acid Techniques in Bacterial Systematics, Wiley & Sons Ltd, pp. 205–248. [40] Suarez-Suarez, A., et al. (2011) Response of sulfate-reducing bacteria to an artificial oil-spill in a coastal marine sediment. Environ. Microbiol. 13, 1488–1499, doi:10.1111/j.1462-2920.2011.02451.x.
[41] Tischer, K., Kleinsteuber, S., Schleinitz, K., Fetzer, I., Spott, O., Stange, F., Lohse, U., Gerling, S., Schmidt, C., Harms, H., Wendeberg, A., (in press). Microbial communities along biogeochemical gradients in a hydrocarbon contaminated aquifer. [42] Tischer, K., Wendeberg A., (in press). Microbial population dynamics along biogeochemical gradients in a hydrocarbon contaminated aquifer as revealed by CARD-FISH. [43] Vaara, M. (1992) Agents that increase the permeability of the outer membrane. Microbiol. Rev. 56, 395–411. [44] Weise, A., Liehr, T., Claussen, U., Halbhuber, K.-J. (2005) Increased efficiency of fluorescence in situ hybridization (FISH) using the microwave. J. Histochem. Cytochem. 53, 1301–1303, doi:10.1369/jhc.5B6632.2005. [45] Wendeberg, A., Zielinski, F.U., Borowski, C., Dubilier, N. (2011) Expression patterns of mRNAs for methanotrophy and thiotrophy in symbionts of the hydrothermal vent mussel Bathymodiolus puteoserpentis. ISME J., doi:10.1038/ismej.2011.81. [46] Witzig, M., Boguhn, J., Kleinsteuber, S., Fetzer, I., Rodehutscord, M. (2010) Influence of the maize silage to grass silage ratio and feed particle size of rations for ruminants on the community structure of ruminal Firmicutes in vitro. J. Appl. Microbiol. 109, 1998–2010, doi:10.1111/j.1365-2672.2010.04829.x, JAM4829 [pii]. [47] Zeder, M. (2009) Development of methods and technology for automated high-throughput multi-parameter analysis of single cells by fluorescence microscopy. PhD thesis, Zürich. [48] Zeder, M., Ellrott, A., Amann, R. (2011) Automated sample area definition for high-throughput microscopy. Cytometry A 79A, 306–310, doi:10.1002/cyto.a.21034. [49] Zeder, M., Kohler, E., Pernthaler, J. (2010) Automated quality assessment of autonomously acquired microscopic images of fluorescently stained bacteria. Cytometry A 77A, 76–85, doi:10.1002/cyto.a.20810. [50] Zeder, M., Pernthaler, J. (2009) Multispot live-image autofocusing for highthroughput microscopy of fluorescently stained bacteria. Cytometry A 75A, 781–788, doi:10.1002/cyto.a.20770.