Technical Reference #1639

1639.

In Situ Characterization of Nitrifying Biofilm: Minimizing Biomass Loss and Preserving Persepctive Robert Delatolla; Nathalie Tufenkji; Yves Comeau; Danidel Lamarre; Alain Gadbois; Dimitrios Berk, McGill University, Water Research, 43(1639), (2009)

Abstract:
Methods for characterizing nitrifying bacteria within biofilms are of key importance to understand and optimize the nitrification kinetics of attached growth treatment facilities.In this work we propose an analytical protocol based upon environmental s

Keywords:
biofilm; nitirification; wastewater; ESEM; CLSM; FISH; nitrobacter; nitrosomonas; biomass

Materials & Methods:
2.1. Biofilm samples The biofilms analyzed in this study were sampled from two laboratory-scale nitrifying BAF up-flow reactors. The nitrifying biofilms were grown on 3.8 1.2 mm diameter hydrophobic polystyrene bead substrata (Veolia Water Paris France) within laboratory reactors. The temperature within the reactors was 20 C; the hydraulic retention time (HRT) ranged from 1 to 4 h depending on the influent concentration of ammonia the superficial velocity was maintained at 21 mm/min with a recirculation pump the pH was between 7.2 and 7.8 and the dissolved oxygen (DO) ranged between 6 and 7 mg/L. The design of the laboratory reactors allowed the removal of the substratum from the reactor with minimum disturbance to the biofilm. Synthetic wastewater (SWW) solution was used to provide nutrients and ammonia to the biofilm within the reactors. The SWW was composed of Na2CO3$H2O (44 mg/L) FeSO4$7H2O (80 mg/L) ethylenediamine tetraacetic acid (EDTA) (178 mg/L) KH2PO4 (5833 mg/L) NaH2PO4 (463 mg/L) CaCl2$2H2O (30 mg/L) MgSO4$7H2O (178 mg/L) MnCl2$4H2O (0.1 mg/L) Na2MoO4$2H2O (0.05 mg/L) CuSO4$5H2O (0.01 mg/L) ZnSO4$7H2O (0.05 mg/L). The ammonia concentration within the laboratory reactors was maintained at 5 mg-N/L while the feed stream was devoid of NO2 NO3 and carbon. The laboratory reactors were allowed to run for a period of approximately 6 months before the analytical examination of the biofilm commenced. It should be noted that for simplicity different beads were used for the ESEM analysis and the CLSM FISH analysis. However the protocol presented here could readily be carried out such that beads analyzed by ESEM are subsequently analyzed by CLSM FISH. 2.2. SEM sample preparation and image acquisition Ten polystyrene beads with attached biofilm were fixed with 2.5% gluteraldehyde for 3 h. The samples were then rinsed in phosphate-buffered saline (PBS) and placed consecutively in vials containing 50% 70% and 96% ethanol for 20 min at each concentration. Following dehydration the samples were critical point dried using carbon dioxide as the transitional fluid instead of air drying to maintain the stability of the fixed structures. The samples were attached onto viewing stages and sputter coated with gold in order to prevent static buildup during viewing of the samples. The samples were imaged on the SEM microscope (Philips XL30 FESEM). 2.3. ESEM image acquisition Ten polystyrene beads with attached biofilm were removed from the laboratory reactor and 100 ESEM images were taken without storage or preparation (Hitachi S-4300SE/N ESEM). The samples were placed directly onto a viewing stage and subsequently placed inside the microscope vapor chamber. The gaseous phase within the vapor chamber was water vapor. Samples were analyzed for up to 20 min at 20 Pa before the biofilm began to dry within the vapor chamber and the appearance of the biofilm was affected. Secondary electron detection mode was used to capture high resolution topographic images of the biofilm while backscattered electron detection mode was used to capture images of exposed bead surface and biofilm with higher contrast. The two detection modes were used interchangeably balancing resolution and contrast. Images between the 250 and 500 optical magnifications proved ideal for quantification of the biofilm percent coverage and thickness. 2.4. ESEM image analysis The percent coverage of the biofilm on the bead and the thickness of the biofilm were quantified from the ESEM images using the image analysis software Vision Assistant 7.1 (National Instruments LabView 8.0). The integrated threshold function was used to segment the pixels of a grayscale ESEM image into two binary categories a biofilm category and the background (bead surface) category. All pixels that fell within the grayscale range defined as the biofilm interval were assigned a value of one. All other pixels within the background interval were set to zero. The thickness of the biofilm on the beads was also quantified from the ESEM images. Sample images were captured such that the curvature of the bead allows the final extent of the biofilm to be digitally measured relative to the surface of the bead. Vision Assistant 7.1 was used to measure the biofilm thickness. 2.5. FISH reference cells Pure cultures of the ammonia oxidizing bacterium (AOB) Nitrosomonas europaea (ATCC no. 25978) and of the nitrite oxidizing bacterium (NOB) Nitrobacter sp. (ATCC no. 25384) were obtained from the American Type Culture Collection (ATCC) Rockville MD USA. The pure cultures were inoculated and maintained in ATCC culture media recipes no. 2265 and no. 480 respectively. Both cultures were maintained in a shaker at 26 C. Nitrosomonas cultures were monitored every three days to maintain the pH between 7.2 and 7.8 with sterile 5% Na2CO3. Nitrobacter cultures were monitored weekly for nitrite accumulation to prevent inhibition conditions. The pure Nitrosomonas and Nitrobacter cultures served as controls to test the hybridization of the oligonucleotide probes. 2.6. FISH oligonucleotide probes The following rRNA-targeted oligonucleotide probes were used. (1) EUB338 a probe that anneals to the rRNA of all bacteria was used in this study as a positive control for hybridization efficiency; (2) NON338 a complementary probe to EUB338 that is incapable of annealing to the rRNA of bacteria was used in this study as a negative control; (3) NSO190 a probe specific to all known beta-subclass ammonia-oxidizing proteobacteria was used in this study to hybridize ammonia oxidizing bacteria (Nitrosomonas) and (4) NIT3 a probe specific to Nitrobacter spp. The oligonucleotide probes were synthesized labeled with fluorescent dyes and purified by MWG Biotech Inc. Huntsville AL USA. The labeling fluorophores used were indocarbocyanine dye CY5 indocarbocyanine dye CY3 indocarbocyanine dye CY3 and fluorescein isothiocyanate (FITC) for the oligonucleotide probes EUB338 NON338 NSO190 and NIT3 respectively. The probe sequences the specificity the attached fluorophores and the hybridization conditions are listed in Table 1. 2.7. FISH sample preparation Ten whole bead substrata samples with the attached biofilm were subjected to the in situ hybridization procedures of Amann et al. (1990) and Manz et al. (1992). Modifications to the procedure were introduced in order to hybridize the entire substratum with the attached biofilm as opposed to removed biofilm fragments attached to a microscope slide. A 30 gauge syringe (Fisher Scientific Ottawa ON Canada) was pushed into the polystyrene bead so that the sample (i.e. the bead substratum with the attached biofilm remaining intact) could be manipulated and controlled by the syringe staff attached to the bead thus eliminating the potential effects of tools coming into contact with the biofilm and thereby conserving the integrity of the biofilm around the exterior of the bead. The portion of the bead substratum influenced by the syringe was negligible with respect to the total surface area of the bead leaving greater than 95% of the bead surface area intact and available for analysis. The samples were fixed with 4% paraformaldehyde for 1 h at 4 C by submerging the bead into a 2 mLvial of paraformaldehyde. During this process care was taken to ensure that the samples did not contact the sides or bottom of the vial. The samples were then rinsed in PBS and placed in vials containing 50% 70% and 96% ethanol at 4 C for 3 min at each concentration. Hybridization with two probes was performed on each sample in subsequent steps by first hybridizing with the probe of higher stringency (Wagner et al. 1994). All hybridizations were performed by placing the samples within vials containing the hybridization buffer and the probe at 46 C for 90 min. The hybridization buffers were composed of 0.9 M NaCl 20 mM trishydroxymethylaminomethane– hydrochloric acid (Tris–HCl) 0.01% sodium dodecyl sulfate (SDS) and a probe-specific formamide percentage as indicated in Table 1. The probe concentration within the hybridization buffer for each of the four probes used in this study was 5 ng/mL. Hybridization was followed by rinsing the samples with a washing buffer and placing the beads within a vial containing the washing buffer at 48 C for 20 min. The washing buffers consisted of 20 mM Tris–HCl 5 mM ethylenediamine tetraacetic acid (EDTA) 0.01% SDS and a probespecific NaCl concentration as shown in Table 1. Following the washing step the bead samples were air dried rinsed with sterile distilled water and stained with 46-diamidino-2-phenylindole (DAPI 1 mg/ml) for 10 min in the dark. The samples were then rinsed with sterile distilled water air dried and mounted on glass bottom culture dishes (MatTek Corporation Ashland MA USA) in GelTol mounting medium (Fisher Scientific Ottawa ON Canada). The samples were stored at 4 C in the dark for up to a week until examined. 2.8. FISH image acquisition Hybridized samples of the bead substratum with the attached biofilm were analyzed using a Zeiss LSM 510 META confocal laser scanning microscope equipped with an argon laser (458 nm 477 nm 488 nm 514 nm) diode laser (405 nm) and two HeNe lasers (543 nm 633 nm). The HeNe laser (543 nm) and BP 560–615 filter was used to excite and emit the CY3 fluorophore; the HeNe laser (633 nm) and LP 650 filter was used to excite and emit the CY5 fluorophore; the argon laser (488 nm) and the BP 505–530 filter was used to excite and emit the FITC fluorophore; and the diode laser (405 nm) and the LP 420 filter was used to excite and emit fluorescence from the DAPI stain. A 63 oil immersion lens was used to view and capture 1.0 mm thick 207 207 mm optical sections. Five sections with vertical intervals of 2–6 mm between each section were acquired within the biofilm at randomly selected adjacent locations to produce a stack of images that can be combined to produce a visual 3D representation of the biofilm at each specific location and can ultimately be used to determine the average cell numbers per volume of biofilm. In this study a thickness of 1.0 mm was chosen for each section based on the value corresponding to the approximate mean cell diameter for Nitrosomonas and Nitrobacter populations (Sharma and Ahlert 1977). The intervals between the acquired sections were chosen depending upon the specific thickness of the biofilm at the location of interest; particularly the first of the 5 sections was acquired at the substratum/biofilm interface and the final section was acquired near the external limit of the biofilm (the biofilm/liquid interface when the bead sample was within the reactor). The whole-cell identification efficiency of the NSO190 and NIT3 specific probes were tested using pure cultures of N. europaea and Nitrobacter sp. Each sample of pure culture was hybridized with the NSO190 NIT3 EUB338 and NON338 in addition to being stained with DAPI. NSO190 EUB338 and DAPI allowed 99.3 2.5% 99.0 2.6% and 99.5 1.5% of the N. europaea cells to be identified respectively. Only 0.1 0.3% and 0.1 0.2% of the pure culture cells of N. europaea were falsely identified using NON338 and NIT3 respectively. NIT3 EUB338 and DAPI allowed 98.8 2.8% 98.3 3.4% and 99.0 2.6% of the Nitrobacter cells to be identified respectively. Only 0.1 0.3% and 0.5 1.2% of the Nitrobacter cells were falsely identified using NON338 and NSO190 respectively. In addition to inefficient hybridization the ability to quantify the bacterial community within biofilm using FISH with rRNA-targeted oligonucleotide probes can be compromised by (1) limited probe permeability of the target bacterial cells (2) low cellular rRNA content of the bacterial cells or (3) background fluorescence (Amann et al. 1995). In situ hybridization of the biofilm samples with the EUB338 NSO190 and NIT3 probes along with DAPI staining demonstrated that all the probes and the dye were able to penetrate to the complete depth of the biofilm and that the bacterial cells contained a sufficient rRNA content to be readily visualized. The accessibility of the cells at the substratum/biofilm interface to the DAPI stain and to the oligonucleotide probes was confirmed by staining cells with DAPI and hybridizing with the EUB338 probe. Specifically the fact that DAPI and EUB338 illuminated the same cells is an indication of good probe permeability. It was further confirmed that the cells located at the substratum/biofilm interface were able to be hybridized with NSO190 and NIT3 by comparing the total number of cells at the substratum/biofilm interface stained with DAPI with the total number of cells hybridized with NSO190 and NIT3. The small average thickness of biofilm (35 mm) associated with the samples used in this study is believed to be advantageous in terms of probe penetration through the film. Thus the study of the applicability of this protocol to thicker heterotrophic biofilms from municipal treatment plants is of interest for future studies in our laboratory. The intrinsic fluorescence of the bead surface and the biofilm (background fluorescence) can produce false positives or mask hybridized cells during analysis. Thus NON338 probe hybridization and hybridization without probes were used to measure the detector gain amplifier gain and amplifier offset settings of the confocal microscope that produced background fluorescence emission of the bead surface and/or the biofilm used in this study. The testing phase of this study was subsequently conducted below the detector gain amplifier gain and amplifier offset values that correlate to background fluorescence effects of the samples. 2.9. FISH image analysis The size and distribution of the bacterial colonies the percent coverage of the bacterial colonies with respect to the biofilm and the cell count of the ammonia oxidizing bacteria and the nitrite oxidizing bacteria were quantified from 200 digital CLSM images. The size and distribution of the ammonia oxidizing colonies and the nitrite oxidizing colonies were measured using the software package that accompanied the Table 1 – Probe sequences specificity attached fluorophores and hybridization conditions used for the rRNA oligonucleotide probes used in this study. Probe Probe sequence Specificity Fluorophore Hybridization conditions Reference Formamide concentration (%) NaCl concentration (mM) EUB338 GCT GCC TCC CGT AGG AGT Bacteria Cy5 20 0.225 Amann et al. (1990) NON338 ACT CCT ACG GGA GGC AGC Non-binding Cy3 20 0.225 Schramm et al. (1996) NSO190 CGA TCC CCT GCT TTT CTC C Ammoniaoxidizing bacteria of the beta-subclass of the Proteobacteria Cy3 55 0.020 Mobarry et al. (1996) NIT3 CCT GTG CTC CAT GCT CCG Nitrobacter spp. FITC 40 0.056 Wagner et al. (1994) confocal microscope (Zeiss LSM Image Browser Version 4.2). The percentage coverage and cell count of the ammonia oxidizing and nitrite oxidizing bacteria were determined using the same software used for the ESEM image analysis (Vision Assistant 7.1). The threshold analysis function of NI Vision Assistant was used to segment the illuminated cells of interest from the rest of the CLSM image. This procedure was followed for the ammonia oxidizing bacteria and subsequently for the nitrite oxidizing bacteria. The segmented pixels corresponding to the cells of interest were then quantified and expressed as the percentage coverage of the image area and/or enumerated and expressed as the number of cells per image. The threshold values used in this study were calibrated separately for each probe. The number of cells within a defined area of an image was first manually counted and secondly quantified in terms of pixels by adjusting the threshold value in order to generate the correct pixel number that corresponded to the counted number of cells. The agreement between the manually counted cells and the cells counted by the threshold analysis determined from an additional 20 analyzed images was 106.1 17.1% and 101.8 8.1% for ammonia oxidizing bacteria and nitrite oxidizing bacteria respectively.

Microscopic Technique
Confocal Microscopy, Laser Scanning Confocal Microscopy

Cell Type(s)
europaea