TY - JOUR
AU - Beech, Iwona
AB - Abstract Navy vessels consist of various metal alloys and biofilm accumulation at the metal surface is thought to play a role in influencing metal deterioration. To develop better strategies to monitor and control metallic biofilms, it is necessary to resolve the bacterial composition within the biofilm. This study aimed to determine if differences in electrochemical current could influence the composition of dominant bacteria in a metallic biofilm, and if so, determine the level of resolution using metagenomic amplicon sequencing. Current was generated by creating galvanic couples between cathodes made from stainless steel and anodes made from carbon steel, aluminum, or copper nickel and exposing them in the Delaware Bay. Stainless steel cathodes (SSCs) coupled to aluminum or carbon steel generated a higher mean current (0.39 mA) than that coupled to copper nickel (0.17 mA). Following 3 months of exposure, the bacterial composition of biofilms collected from the SSCs was determined and compared. Dominant bacterial taxa from the two higher current SSCs were different from that of the low-current SSC as determined by DGGE and verified by Illumina DNA-seq analysis. These results demonstrate that electrochemical current could influence the composition of dominant bacteria in metallic biofilms and that amplicon sequencing is sufficient to complement current methods used to study metallic biofilms in marine environments. Electronic supplementary material The online version of this article (doi:10.1007/s10295-016-1887-7) contains supplementary material, which is available to authorized users. Introduction US Navy ships are constructed from various metal alloys [20], each of which has a characteristic corrosion rate in seawater [32]. The most common alloys include stainless steel fittings, fasteners and propulsion systems; carbon steel seawater-compensated fuel ballast tanks; copper nickel sluice piping; and aluminum hulls. During design and fabrication, some metals are selected for corrosion resistance, while others are selected based on cost or weight restrictions [5, 22]. In the latter case, methods to modulate metal deterioration are typically employed. These include the intentional galvanic coupling of dissimilar metals, such as the cathodic protection of carbon steel by the addition of sacrificial zinc anodes [33, 43] or the application of corrosion-resistant coatings [46]. In addition to general corrosion, biofilm accumulation at the metal surface is thought to play a role in influencing metal deterioration [17] and both protective and destructive biofilm mechanisms have been proposed using defined cultures [47]. Biofilms consist of a community of microorganisms and their exopolymeric substances and current efforts to prevent bacterial attachment or delay biofilm growth include the application of antifouling coatings [8] or biocide treatments [39]. A recent review by Sultana et al. summarized a third option for biofilm eradication termed electrochemical biofilm control, whereby the addition of electrochemical current or potential could interfere with the attachment of bacteria to the metal surface [39]. The laboratory studies referenced therein utilized various metals including gold-covered glass and platinum, monocultures (either Pseudomonas, Vibrio, or Staphylococcus), and performed viable cell counts following 30 min to seven days of exposure. In a few cases, the mechanism proposed was the production of biocides at the surface of the metal from reactants in the system (i.e., reactive chlorine or oxygen species) [39]. Although more environmentally friendly than current biofilm control measures, conclusions based off of electrochemically generated biocide studies are limited due to high variability in biofilm removal on a case-by-case basis. Specifically, the efficacy of biofilm removal varies based on the bacterial monoculture under study, exposure time, and availability of reactants to form biocides in a given system [39]. Moreover, studies that include mixed-species biofilms are lacking. The application of high-throughput deep sequencing to identify microorganisms has shed light on the composition and complexity of uncultured microbial assemblies in diverse natural environments including coastal seawaters [25], soils [23], and glaciers [37]. Deep sequencing has also been used to identify core taxa in biofilms from manmade environments, including bioelectrochemical systems [45] and drinking-water networks [21]. Because electrochemical biofilm control is a promising alternative to traditional biofilm control measures, the major goals of the present study were to determine if 16S rRNA amplicon sequencing could resolve the bacterial composition of mixed-species biofilms on metal surfaces from a natural marine environment and if electrochemical current could influence the composition of dominant bacteria in a metallic biofilm. Three types of galvanic couples (stainless steel:carbon steel, stainless steel:aluminum, and stainless steel:copper nickel) representing two high- and one low-current treatment were submerged in the Delaware Bay and allowed to foul for 3 months. At the end of 3 months, biofilm material was removed from the SSCs and the bacterial composition was determined using Illumina DNA-seq analysis. Two significant findings resulted from this study. (1) Although Gammaproteobacteria dominated all three SSC biofilms, the dominant taxa at the family level differed between the high versus low-current SSC biofilms. (2) Taxonomic identification at the species level was not achieved due to the lack of represented members in databases used for sequence classification. However, resolution of the dominant unclassified Gammaproteobacteria was comparable to that of full-length 16S rRNA in this study. Results presented here suggest that the composition of the dominant bacteria at the cathode surface was current-dependent. It is proposed that the biofilm community profile reflects differences in the chemical environment at the metal surface. However, whether varying current values could serve as a tool in corrosion management, i.e., prevent the development of metal deteriorating and/or encourage the growth of protective marine biofilms, would require further investigation. Materials and methods Metal coupon description Three types of galvanic couples were constructed using four metal alloys: 316 stainless steel, 5083 aluminum, 70/30 copper nickel, and 1018 carbon steel. All metal coupons were 1/16″ × ½″ × 3″ with a 3/16″ diameter hole at one end and had glass-bead blasted finishes (Metal Samples Company, Munford, AL, USA). When galvanically coupled, stainless steel served as the cathode and aluminum, copper nickel, or carbon steel served as anodes (Fig. 1a). Fig. 1 Open in new tabDownload slide Experimental setup. a Generation of current by galvanically coupling stainless steel cathodes to aluminum, copper nickel, or carbon steel anodes. Twenty-one separate galvanic couples were set up for each of the three bimetallic pairs, stainless steel:aluminum, stainless steel:copper nickel, and stainless steel:carbon steel. Current measurements were taken weekly for 3 months. At the end of 3 months, the setup was dismantled and biofilm material from five stainless steel cathodes (SSCs) was removed and pooled to create one biofilm sample. There were a total of four SSC biofilm samples for each of the three types of bimetallic pairs. b Open circuit potential measurements in 21 unpaired metal coupons were taken weekly for 3 months: SS stainless steel, Al aluminum, CN copper nickel, CS carbon steel Galvanic couple setup and electrochemical measurements Nickel–chromium wire leads were connected to the coupons using a stainless steel spade terminal and nylon screw. The terminal was soldered to the lead, and the entire connection was coated with stop-off lacquer to prevent unintended galvanic connections. Metal coupons were washed successively with acetone, ethanol, and deionized water prior to immersion. Galvanic couples with a 1:1 surface area ratio were created by joining the leads from a single anode coupon (e.g., aluminum, copper nickel, or carbon steel) to a single cathode coupon (e.g., stainless steel) via a toggle switch, which could redirect the current to flow through a picoammeter for electrochemical measurements (Fig. 1). There were 21 separate galvanic couples set up for each of the three bimetallic pairs (stainless steel:carbon steel, stainless steel:aluminum, and stainless steel:copper nickel). To accumulate sufficient volumes of biofilm material for a molecular analysis, metagenomic DNA from five SSCs was pooled together to create one biofilm sample (Fig. 1a). Unpaired metal coupons were included in the setup to measure the open circuit potential (OCP) with a handheld digital multimeter and saturated calomel electrode (SCE) (Fig. 1b). Both galvanically coupled and unpaired metal coupons were hung from wooden racks and immersed in a flowing seawater tank (4 × 0.65 × 0.17 m) on July 10, 2013. Water was pumped directly from Roosevelt Inlet in the Delaware Bay into a settling tank, which gravity fed the flow-through tank. A baffle separated the water input pipe from the experimental coupons, which were in nearly quiescent water with approximately 0.15 cm/s flow. All metal coupons were immersed parallel to the flow with the screw initially 0.5 cm above the water line; however, the exact water level fluctuated slightly throughout the duration of the experiment. Galvanic couples were situated 5 cm apart, and unpaired metal coupons were a minimum of 7.5 cm apart. Current and OCP measurements were taken weekly for 3 months (Fig. 2; Online Resources 1 and 2). A one-way ANOVA with Tukey’s post-test for significance was performed on mean OCP values for uncoupled coupons and mean current values for galvanic couples using GraphPad Prism version 6.05 for Windows (GraphPad Software, San Diego, CA, USA, http://www.graphpad.com). The water temperature ranged from 17.3 to 25.4 °C and the salinity in the seawater tank ranged from 25.2 to 27.1 ppt for the duration of the experiment. Fig. 2 Open in new tabDownload slide a OCP measurements in unpaired metal coupons. The graph is a plot using the values from Online Resource 1. Left side each line represents a single stainless steel (orange), copper nickel (green), aluminum (blue), or carbon steel (purple) coupon. Right side each point represents the 3-month mean OCP value for a single coupon. Bars are mean ± SD for each group. All groups were significantly different (p < 0.0001) from one another with the exception of CS and Al (p = 0.7444). b Current measurements in galvanic couples. The graph is a plot using values from Online Resource 2. Left side each line represents a single galvanic couple, stainless steel:carbon steel (purple lines), stainless steel:copper nickel (green lines), or stainless steel:aluminum (blue lines). Right side each point represents the 3-month mean current value for a single galvanic couple. Bars represent mean ± SD for each group. The mean current for SS:CS and SS:Al were not significantly different (p = 0.8848). However, both were significantly different from SS:CN (p < 0.0001) All metal coupons were removed from the tank on October 3, 2013, and placed directly into 50-mL falcon tubes with 2 mL DNAzol reagent (Molecular Research Center, Inc., Cincinnati, OH, USA) to preserve the biofilm material. An additional 2 L of unfiltered seawater from the flowing seawater tank was also collected. These samples were placed into boxes and shipped on ice overnight to the University of Oklahoma laboratory for a molecular analysis. The galvanic couples and unfiltered seawater arrived at the destination, and all further analyses were performed using only these samples. Samples were stored at −20 °C until DNA extraction. DNA extraction from SSC biofilms and bulk water DNA was extracted from SSCs and seawater samples using PowerBiofilm DNA extraction kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) with modification. Two 1-L aliquots of seawater from the flowing seawater coupon-holding tank were filtered through two Nalgene Rapid-Flow bottle top filter units with a 0.2 μm pore size (VWR International, Radnor, PA, USA) to capture free-floating bacteria. The PES filters were removed from the units using a sterile scalpel, transferred to 50 mL falcon tubes, and preserved in 2 mL DNAzol (Molecular Research Center). One tube of PowerBiofilm Beads and one tube of Lysing Matrix E Beads (MP Biomedical, LLC., Santa Ana, CA, USA) were added to filters and to each metal coupon to facilitate removal of the biofilm material from the surface. For all subsequent steps, each sample (single PES filter or metal coupon) was treated as two extractions and buffer volumes were scaled up accordingly. A Vortex Genie 2 (MO BIO Laboratories, Inc.) with vortex adapter was used to bead-beat coupons on the maximum setting for 5 min. Spin-filters were centrifuged a final time for 2 min and placed under a vacuum for 5 min or until all traces of ethanol were removed. DNA was eluted from spin-filters in 100 μL volumes using BF7. The two 100-μL DNA extractions from each sample were pooled and concentrated using Amicon Ultra-0.5 mL 30K filtration devices (Millipore, Billerica, MA, USA) following the manufacturer’s instructions. Samples were loaded onto a single filter unit and centrifuged at 14,000×g for 5 min. The filter unit was transferred to a fresh collection tube and the concentrated DNA was recovered by centrifugation for 2 min at 1000×g. DNA concentrations were measured using a Qubit 2.0 fluorometer (Life Technologies) with the dsDNA BR Assay kit (Life Technologies) following the manufacturer’s instructions. There were four DNA samples for each of the three types of SSC biofilms and two DNA samples for seawater. DGGE A DGGE fingerprint comparing the banding pattern of dominant bacterial partial 16S rRNA sequences was performed as an initial assessment of bacterial diversity. For this analysis, two of the four DNA samples from each of the three types of SSC biofilms were included. The V3–V5 region was amplified from DNA samples using the primer set GM5F-GC clamp and 907R [35]. PCR amplification was performed using the same setup and cycling conditions as previously described [30]. 5–10 μL of each reaction were resolved on a 40–60% denaturing gradient gel at 65 V for 16 h at 60 °C, with 100% denaturant as 7 M Urea and 40% formamide [28]. The gel was stained for 15 min with SYBR Safe (LifeTechnologies) at a final concentration of 1:1000 in deionized water and imaged using the Gel Logic 112 Imaging System and Molecular Imaging Software v5 (Carestream, WoodBridge, CT, USA). Preparation of 16S rRNA amplicon libraries for Illumina DNA-seq To identify dominant bacterial taxa in each of the three SSC biofilms, partial 16S rRNA amplicon libraries were generated for each sample. The V4 region of the 16S rRNA gene was amplified using a modified universal primer set 519F and 806R [44], and PCR amplification, purification, and barcoding were performed as previously described [18]. Barcodes are listed in Online Resource 3. Purified barcoded libraries were pooled in equimolar amounts and submitted to the Oklahoma Medical Research Foundation’s Next-Generation DNA Sequencing Facility (omrf.org) for Illumina paired-end (2 × 250 bp) sequencing on a MiSeq instrument. Sequence file was submitted to the SRA under accession number 3185978. Illumina DNA-seq analysis Paired-end sequence reads were processed and analyzed using Trimmomatic 0.32 [6], UPARSE [12], and Qiime 1.8.0 [7]. Briefly, Illumina adapters were removed and low-quality bases were trimmed (Phred score <25 over a 10 base sliding window). Sequences shorter than 50 bases were dropped and surviving paired-end reads with a minimum overlap of 50 bases were stitched together using fastq-join of the EAUtils package [1, 2]. Adapter removal and sequence quality were verified using FastQC 0.11.2 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Stitched-reads were 280–320 bases in length. The UPARSE pipeline was used to filter chimeric sequences and cluster OTUs [12]. Briefly, the sequences were dereplicated using the derep_full length and sortbysize scripts. Sequences were clustered into OTUs based on 97% identity using the cluster_otus script and chimeras were detected and removed using the gold.fa reference database (http://drive5.com/uchime/gold.fa) [19] with the uchime_ref script. Taxonomy assignments, measures of alpha and beta diversity, and statistical analyses were performed using Qiime 1.8.0 [7]. Briefly, the Silva (release 108) database was used to assign taxonomic classifications to OTUs [31] at 90% confidence. All 16S rRNA libraries contained between 46,000 and 130,000 sequences (Table 1). An equivalent number of sequences (40,000) were randomly subsampled from each library for diversity measurements (Table 2) and hierarchical clustering using weighted UniFrac distances [26]. A one-way ANOVA with Tukey’s post-test for significance was performed on Shannon and inverse Simpson group means using GraphPad Prism. Illumina DNA-seq 16S rRNA gene sequence profiles for SSC and seawater libraries Sample library . Sample no. . No. sequences . % Proteobacteria . % Gammaproteobacteria . Seawater 1a 100,842b 47.22c 20.82d 2 76,242 48.56 22.95 0.39 mA (Al) 3 72,017 75.19 49.16 4 73,540 75.49 52.34 5 90,260 73.10 52.30 6 72,447 68.90 46.54 0.17 mA 7 65,519 82.68 63.29 8 111,111 82.15 65.00 9 75,512 80.74 64.49 10 95,617 80.19 61.50 0.39 mA (CS) 11 101,882 74.21 48.64 12 46,262 75.03 48.00 13 127,528 73.51 49.14 14 64,397 71.85 48.09 Sample library . Sample no. . No. sequences . % Proteobacteria . % Gammaproteobacteria . Seawater 1a 100,842b 47.22c 20.82d 2 76,242 48.56 22.95 0.39 mA (Al) 3 72,017 75.19 49.16 4 73,540 75.49 52.34 5 90,260 73.10 52.30 6 72,447 68.90 46.54 0.17 mA 7 65,519 82.68 63.29 8 111,111 82.15 65.00 9 75,512 80.74 64.49 10 95,617 80.19 61.50 0.39 mA (CS) 11 101,882 74.21 48.64 12 46,262 75.03 48.00 13 127,528 73.51 49.14 14 64,397 71.85 48.09 aSample numbers correspond to Fig. 4a bTotal number of high-quality chimera-free 16S rRNA sequences analyzed cProportion of high-quality chimera-free 16S rRNA sequences classified as Proteobacteria dProportion of high-quality chimera-free 16S rRNA sequences classified as Gammaproteobacteria Open in new tab Illumina DNA-seq 16S rRNA gene sequence profiles for SSC and seawater libraries Sample library . Sample no. . No. sequences . % Proteobacteria . % Gammaproteobacteria . Seawater 1a 100,842b 47.22c 20.82d 2 76,242 48.56 22.95 0.39 mA (Al) 3 72,017 75.19 49.16 4 73,540 75.49 52.34 5 90,260 73.10 52.30 6 72,447 68.90 46.54 0.17 mA 7 65,519 82.68 63.29 8 111,111 82.15 65.00 9 75,512 80.74 64.49 10 95,617 80.19 61.50 0.39 mA (CS) 11 101,882 74.21 48.64 12 46,262 75.03 48.00 13 127,528 73.51 49.14 14 64,397 71.85 48.09 Sample library . Sample no. . No. sequences . % Proteobacteria . % Gammaproteobacteria . Seawater 1a 100,842b 47.22c 20.82d 2 76,242 48.56 22.95 0.39 mA (Al) 3 72,017 75.19 49.16 4 73,540 75.49 52.34 5 90,260 73.10 52.30 6 72,447 68.90 46.54 0.17 mA 7 65,519 82.68 63.29 8 111,111 82.15 65.00 9 75,512 80.74 64.49 10 95,617 80.19 61.50 0.39 mA (CS) 11 101,882 74.21 48.64 12 46,262 75.03 48.00 13 127,528 73.51 49.14 14 64,397 71.85 48.09 aSample numbers correspond to Fig. 4a bTotal number of high-quality chimera-free 16S rRNA sequences analyzed cProportion of high-quality chimera-free 16S rRNA sequences classified as Proteobacteria dProportion of high-quality chimera-free 16S rRNA sequences classified as Gammaproteobacteria Open in new tab Alpha diversity measurements for each SSC and seawater librarya Sample library . Sample no. . % Library coverage . OTUs . Chao1 index (95% confidence intervals) . Shannon index . Simpson index (1/D) . Seawater 1b 0.994c 1137 1300 (1254,1365)d 7.33e 58.37f 2 0.994 1160 1377 (1316, 1460) 7.51 71.24 0.39 mA (Al) 3 0.993 966 1241 (1168, 1341) 6.29 25.59 4 0.993 1013 1227 (1169, 1306) 6.26 24.12 5 0.992 1235 1517 (1446, 1612) 6.67 25.46 6 0.994 986 1201 (1142, 1282) 6.33 24.72 0.17 mA 7 0.994 969 1165 (1110, 1242) 5.11 5.81 8 0.993 926 1227 (1146, 1337) 4.69 4.35 9 0.994 1002 1223 (1162, 1307) 4.94 4.72 10 0.993 1016 1240 (1181, 1321) 5.96 12.96 0.39 mA (CS) 11 0.993 1073 1311 (1249, 1394) 6.42 28.37 12 0.995 935 1081 (1038, 1142) 6.50 28.95 13 0.992 1052 1351 (1276, 1453) 6.31 25.27 14 0.994 936 1119 (1068, 1190) 6.11 20.72 Sample library . Sample no. . % Library coverage . OTUs . Chao1 index (95% confidence intervals) . Shannon index . Simpson index (1/D) . Seawater 1b 0.994c 1137 1300 (1254,1365)d 7.33e 58.37f 2 0.994 1160 1377 (1316, 1460) 7.51 71.24 0.39 mA (Al) 3 0.993 966 1241 (1168, 1341) 6.29 25.59 4 0.993 1013 1227 (1169, 1306) 6.26 24.12 5 0.992 1235 1517 (1446, 1612) 6.67 25.46 6 0.994 986 1201 (1142, 1282) 6.33 24.72 0.17 mA 7 0.994 969 1165 (1110, 1242) 5.11 5.81 8 0.993 926 1227 (1146, 1337) 4.69 4.35 9 0.994 1002 1223 (1162, 1307) 4.94 4.72 10 0.993 1016 1240 (1181, 1321) 5.96 12.96 0.39 mA (CS) 11 0.993 1073 1311 (1249, 1394) 6.42 28.37 12 0.995 935 1081 (1038, 1142) 6.50 28.95 13 0.992 1052 1351 (1276, 1453) 6.31 25.27 14 0.994 936 1119 (1068, 1190) 6.11 20.72 aBased on random subsampling of 40,000 sequence reads from each sample bSample number corresponds to Fig. 4a cEquation for calculating Goods estimator of coverage (C), where C=1-n1N $$C = 1 - \frac{{n_{1} }}{N}$$ ⁠. Value range between 0 and 1, where n 1 = number of OTUs sampled once (singletons) and N = total number of sequences in the sample [16] dEquation for calculating the Chao1 estimator of richness (S), where S=Sobs+n1(n1-1)2(n2-1) $$S = S_{\text{obs}} + \frac{{n_{1} (n_{1} - 1)}}{{2(n_{2} - 1)}}$$ ⁠. Minimum value for S is S obs, where S obs is the observed number of OTUs, n 1 are singletons (OTU with one sequence), and n 2 are doubletons (OTU with two sequences) eEquation for calculating the Shannon estimator of diversity (H), where H=-∑i-1SobsniNlnniN $$H = - \sum\nolimits_{i - 1}^{{S_{\text{obs}} }} {\frac{{n_{i} }}{N}{ \ln }\frac{{n_{i} }}{N}}$$ ⁠, where S obs, is the observed number of OTUs, n i is the number of sequences in OTU i; and N is the total number of sequences in the sample. An ANOVA performed on 1/D values between low current and both types of the high-current SSC libraries was significantly different (p < 0.01). Values were not significantly different between the two high-current SSC libraries fEquation for calculating the reciprocal of the Simpson index (1/D), where D=∑i=1Sobsni(ni-1)N(N-1) $$D = \frac{{\mathop \sum \nolimits_{i = 1}^{{S_{\text{obs}} }} n_{i} (n_{i} - 1)}}{N(N - 1)}$$ ⁠. Value of D ranges between 0 and 1, where S obs is the observed number of OTUs, n i is the number of sequences in OTU i , and N is the total number of sequences in the sample. An ANOVA performed on 1/D values between low current and both types of the high-current SSC libraries was significantly different (p < 0.001). Values were not significantly different between the two high-current SSC libraries Open in new tab Alpha diversity measurements for each SSC and seawater librarya Sample library . Sample no. . % Library coverage . OTUs . Chao1 index (95% confidence intervals) . Shannon index . Simpson index (1/D) . Seawater 1b 0.994c 1137 1300 (1254,1365)d 7.33e 58.37f 2 0.994 1160 1377 (1316, 1460) 7.51 71.24 0.39 mA (Al) 3 0.993 966 1241 (1168, 1341) 6.29 25.59 4 0.993 1013 1227 (1169, 1306) 6.26 24.12 5 0.992 1235 1517 (1446, 1612) 6.67 25.46 6 0.994 986 1201 (1142, 1282) 6.33 24.72 0.17 mA 7 0.994 969 1165 (1110, 1242) 5.11 5.81 8 0.993 926 1227 (1146, 1337) 4.69 4.35 9 0.994 1002 1223 (1162, 1307) 4.94 4.72 10 0.993 1016 1240 (1181, 1321) 5.96 12.96 0.39 mA (CS) 11 0.993 1073 1311 (1249, 1394) 6.42 28.37 12 0.995 935 1081 (1038, 1142) 6.50 28.95 13 0.992 1052 1351 (1276, 1453) 6.31 25.27 14 0.994 936 1119 (1068, 1190) 6.11 20.72 Sample library . Sample no. . % Library coverage . OTUs . Chao1 index (95% confidence intervals) . Shannon index . Simpson index (1/D) . Seawater 1b 0.994c 1137 1300 (1254,1365)d 7.33e 58.37f 2 0.994 1160 1377 (1316, 1460) 7.51 71.24 0.39 mA (Al) 3 0.993 966 1241 (1168, 1341) 6.29 25.59 4 0.993 1013 1227 (1169, 1306) 6.26 24.12 5 0.992 1235 1517 (1446, 1612) 6.67 25.46 6 0.994 986 1201 (1142, 1282) 6.33 24.72 0.17 mA 7 0.994 969 1165 (1110, 1242) 5.11 5.81 8 0.993 926 1227 (1146, 1337) 4.69 4.35 9 0.994 1002 1223 (1162, 1307) 4.94 4.72 10 0.993 1016 1240 (1181, 1321) 5.96 12.96 0.39 mA (CS) 11 0.993 1073 1311 (1249, 1394) 6.42 28.37 12 0.995 935 1081 (1038, 1142) 6.50 28.95 13 0.992 1052 1351 (1276, 1453) 6.31 25.27 14 0.994 936 1119 (1068, 1190) 6.11 20.72 aBased on random subsampling of 40,000 sequence reads from each sample bSample number corresponds to Fig. 4a cEquation for calculating Goods estimator of coverage (C), where C=1-n1N $$C = 1 - \frac{{n_{1} }}{N}$$ ⁠. Value range between 0 and 1, where n 1 = number of OTUs sampled once (singletons) and N = total number of sequences in the sample [16] dEquation for calculating the Chao1 estimator of richness (S), where S=Sobs+n1(n1-1)2(n2-1) $$S = S_{\text{obs}} + \frac{{n_{1} (n_{1} - 1)}}{{2(n_{2} - 1)}}$$ ⁠. Minimum value for S is S obs, where S obs is the observed number of OTUs, n 1 are singletons (OTU with one sequence), and n 2 are doubletons (OTU with two sequences) eEquation for calculating the Shannon estimator of diversity (H), where H=-∑i-1SobsniNlnniN $$H = - \sum\nolimits_{i - 1}^{{S_{\text{obs}} }} {\frac{{n_{i} }}{N}{ \ln }\frac{{n_{i} }}{N}}$$ ⁠, where S obs, is the observed number of OTUs, n i is the number of sequences in OTU i; and N is the total number of sequences in the sample. An ANOVA performed on 1/D values between low current and both types of the high-current SSC libraries was significantly different (p < 0.01). Values were not significantly different between the two high-current SSC libraries fEquation for calculating the reciprocal of the Simpson index (1/D), where D=∑i=1Sobsni(ni-1)N(N-1) $$D = \frac{{\mathop \sum \nolimits_{i = 1}^{{S_{\text{obs}} }} n_{i} (n_{i} - 1)}}{N(N - 1)}$$ ⁠. Value of D ranges between 0 and 1, where S obs is the observed number of OTUs, n i is the number of sequences in OTU i , and N is the total number of sequences in the sample. An ANOVA performed on 1/D values between low current and both types of the high-current SSC libraries was significantly different (p < 0.001). Values were not significantly different between the two high-current SSC libraries Open in new tab Full-length 16S rRNA gene clone library analysis A clone library was generated using near full-length 16S rRNA sequences. The V1–V8 region was amplified using the primer set 27F and 1391R [10]. Briefly, PCR amplification was performed using 2 μL of DNA pooled from the four low-current SSC biofilm samples, 12.5 μL DreamTaq Green PCR Master Mix (2×) (ThermoScientific), 0.5 M Betaine, and primers at a final concentration of 250 nM each. Thermal cycling was carried out at 94 °C for 4 min followed by 35 cycles of 94 °C for 1 min, 46 °C for 1 min. and 72 °C for 2 min with a final annealing step of 72 °C for 10 min. The PCR reaction was analyzed by agarose gel electrophoresis, cloned into a pCR4-TOPO vector (LifeTechnologies), and transformed into One Shot Top10 Chemically competent cells (LifeTechnologies) following the manufacturer’s instructions. Positive clones were screened on Luria agar containing 50 μg/mL ampicillin and S-Gal (3,4-cyclohexenoesculetin-β-d-galacto-pyranoside) (Sigma-Aldrich, St. Louis, MO, USA). Ten clones were screened for the correct sized product in 25 μL reactions and 2 nM each M13F and M13R primers, and ten microliters of each reaction were treated with ExoSAP-IT (Affymetrix Inc., Canta Clara, CA, USA) following the manufacturer’s instructions and sequenced using an ABI 3730 Capillary Sequencer (at the Oklahoma Medical Research Foundation, Oklahoma City, OK, USA). Sequencher 4.7 (Gene Codes, Ann Arbor, MI, USA) was used to quality trim and assemble forward and reverse sequence reads. The closest matches were determined using NCBI-BLAST [3] and the nucleotide collection (nt/nt) in addition to the RDP Classifier [42] with the 16S rRNA training set 9 (>80% confidence threshold). The phylogenetic relationship of operational taxonomic unit 1 (OTU1) 16S rRNA sequence was inferred using three methods in MEGA6 [41]: Neighbor-Joining [34], Maximum Likelihood, and Maximum Parsimony. For all three methods, alignments were built using CLUSTALW. All positions with less than 95% site coverage were eliminated. There were a total of 1313 positions in the final dataset. Each tree was subjected to the bootstrap test using 1000 replications [14], and the evolutionary distances for the Neighbor-Joining Tree were computed using the Maximum Composite Likelihood method [40]. The rate variation among sites was modeled with a gamma distribution (shape parameter = 1). The Maximum Likelihood tree was based on the General Time Reversible model [29]. The initial tree for the heuristic search was obtained by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with the superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites [5 categories (+G, parameter = 0.2799)]. The length of the two most parsimonious trees derived using the Maximum Parsimony method was 2255, the consistency index was 0.3393, the retention index was 0.5274, and the composite index was 0.1946 (0.1790) for all sites and parsimony-informative sites (in parentheses). The MP tree was obtained using the Subtree-Pruning-Regrafting (SPR) algorithm [29] with search level 1 in which the initial trees were obtained by the random addition of sequences (10 replicates). Results and discussion OCP measurements in uncoupled metal coupons Weekly OCP measurements were recorded for the unpaired metal coupons and unpaired stainless steel ennobled within the first week (Fig. 2a, orange lines). The OCP values for stainless steel had a mean value of 220 ± 165 mV versus SCE. The high variance was likely due to changes in the passive layer on the metal surface. The mean OCP values for unpaired copper nickel, carbon steel, and aluminum were −162 ± 29, −732 ± 30 mV, and −705 ± 157 mV versus SCE, respectively (Fig. 2a), where carbon steel and aluminum were not significantly different. These observed values were similar to those listed for identical alloys on the galvanic series of metals in flowing seawater [4] and demonstrated that the four types of metal coupons existed at an OCP that was unique to that metal. Based on the galvanic series copper nickel, carbon steel, and aluminum should preferentially corrode when individually coupled to stainless steel, being more electronegative than the latter when immersed in seawater. And this was precisely what was observed for all galvanic couples during the weekly electrochemical measurements and setup dismantlement. Current measurements in galvanic couples Weekly current measurements were recorded for the three galvanic couples, where stainless steel served as the cathode for all three bimetallic pairs (Fig. 2b). As predicted from the similarity in OCP values, aluminum and carbon steel anodes produced similar galvanic currents when coupled to stainless steel (0.39 ± 0.17 mA, blue and purple lines). However, copper nickel created a circuit with a mean current (0.17 ± 0.08 mA, green lines) that was significantly lower (p < 0.0001). Thus, for the remainder of this communication the authors refer to the SSCs as having one of two treatments based on the anode to which it was coupled: high current 0.39 mA versus low current 0.17 mA. At the onset of the experiments, high fluctuations in current were observed in July and August of 2013. These fluctuations were between 0.25–0.95 mA for both high-current galvanic couples and 0.05–0.40 mA for the low-current galvanic couples. These variations were likely due to the initial instantaneous metal oxidation following immersion in seawater. Similar observations have been made for carbon steel in aerated seawater [24]. The authors observed the largest fluctuations in corrosion rates and corrosion potential in the first days of exposure [24]. The buildup of metal oxides and other corrosion products would have formed a passive layer on the metal surface. This then impedes mass transport to the interface and subsequently reduced the current values or corrosion rate. The formation of a biofilm on the metal surface would also cause a similar effect as described above. Biofilm formation is reported to begin immediately following immersion in liquids [9]. In some cases, the biofilm could be substantially developed into a complex heterogenous structure within 48 h of immersion [9]. Although the exact mechanisms by which electrons are transferred between microorganisms and solid metal substances is still a topic of intense debate, this is also a possibility. The fact that fluctuations were consistent among replicates between both high-current galvanic couples suggests that it was not due to faulty leads or sloughing of the precipitate (the carbon steel in particular built up a lot of rust and would slough randomly). There were no major fluctuations in salinity or temperature during that time. DGGE analysis of high- versus low-current SSC biofilms DNA extracted from high- and low-current SSC biofilms were analyzed by DGGE to determine if there were observable shifts in the dominant bacterial taxa. An initial comparison of bacterial diversity (i.e., 16S rRNA banding pattern) for two SSC biofilm samples from each bimetallic pair suggested that the three SSC biofilms were similar but not identical to each other (Fig. 3). There were several visible bands that ran to the same position in all three SSC biofilms (Fig. 3, arrows) suggesting that they may be a shared 16S rRNA sequence. Importantly, a couple of dominant bands appeared to be unique to the low-current SSC biofilms (Fig. 3, asterisks). The grouping of biofilm communities from both of the high-current SSCs reflects their similarity in current flow (Fig. 2b), suggesting that the ~0.2 mA difference in current between them and the low-current biofilm was enough to shift the dominant taxa on the cathode surface, and this shift was reflected in the DGGE analysis. Fig. 3 Open in new tabDownload slide DGGE analysis of bacterial 16S rRNA in SSC biofilm samples. The amplified V3–V5 region of 16S rRNA was resolved on a 40–60% denaturing gradient gel. The banding pattern from two SSC biofilm samples is shown for each treatment. 0.17 mA for SSCs that were coupled to copper nickel, 0.39 mA (Al) for SSCs that were coupled to aluminum, and 0.39 mA (CS) for SSCs that were coupled to carbon steel. Arrows indicate dominant bands present in all three SSC biofilm samples. Arrows with stars indicate dominant bands unique to the low-current SSC biofilm Bacterial communities in proximal seawater and high- versus low-current SSC biofilms An Illumina DNA-seq analysis grouped 1,173,176 high-quality chimeric-free 16S rRNA sequences from all twelve SSC biofilm and the two seawater 16S rRNA amplicon libraries into 2492 operational taxonomic units (OTUs). The total number of sequences analyzed from each library is indicated in Table 1. Illumina libraries contained between 45,000 and 130,000 sequences and those that were classified to the phylum Gammaproteobacteria dominated all SSC biofilms (Table 1). Interestingly, the proportion of Gammaproteobacteria was noticeably higher in low-current SSC biofilm samples compared to the two high-current SSC biofilm samples, representing >60% versus ~50% of the total sequences, respectively. Notably, the structure of seawater communities from the Delaware Bay was very different from the SSC biofilms (Fig. 4; Table 3). These findings were consistent with the studies by Henne et al. illustrating that the dominant taxa in drinking water was significantly different from that in proximal biofilms [21]. Also consistent with previous studies was the composition and relative abundance of the dominant phyla in the bulk water [25, 27, 38], which included Alphaproteobacteria (20%), Gammaproteobacteria (22%), Bacteroidetes (18%), and Actinobacteria (18%). Together these data support numerous reported findings in bulk water systems, where the structure of free-floating bacterial populations does not necessarily reflect that of a biofilm community. Fig. 4 Open in new tabDownload slide Illumina DNA-seq analysis. a Relative abundance of bacterial 16S rRNA sequence reads in seawater and SSC biofilm libraries. The V4 region of the bacterial 16S rRNA gene was amplified and used for taxonomic classification of 16S rRNA in seawater replicates (1 and 2) and both high- and low-current SSC biofilm samples (3 through 14). Dominant taxa belonging to the Gammaproteobacteria, Alphaproteobacteria, and Bacteroidetes are shown in the legend. b UPGMA clustering of SSC biofilm and seawater communities using weighted unifrac algorithm. Left side a dendrogram generated by randomly subsampling 40,000 16S rRNA sequence reads from each library. The jackknifed tree illustrates preferential clustering with internal nodes colored red for high support (75–100%), yellow for moderate support (50–75%), green for 25–50% support, and blue for <25% support. Right side the OTU heatmap illustrates bacterial community composition based on the relative contribution of each OTU to the total OTUs present in the sample on a scale from dark blue (low representation) to red (high representation). Heatmap consists of OTUs with greater than 2000 sequence reads Most abundant OTUs (highest number of total 16S rRNA sequence reads across pooled libraries) OTU number . Sequence reads . Taxa (Phylum or class level for Proteobacteria) . Seawater 177,084a . 0.39 mA (Al) 308,264 . 0.17 mA 347,759 . 0.39 mA (CS) 340,069 . 1 151,809 Gamma Thiotrichales(o) Thiotrichaceae(f) unc(g);unc(s) 0.30 ± 0.01b 4.05 ± 7.08 38.91 ± 9.58 0.47 ± 0.06 4 69,288 Gamma Acidithiobacillale(o) Acidithiobacillaceae(f) Acidithiobacillus(g) unc(s) 0.17 ± 0.02 7.44 ± 2.78 2.05 ± 0.74 11.87 ± 2.26 2 61,411 Bacteroidetes Sphingobacteria(c) Sphingobacteriales(o) Flammeovirgaceae(f) Flexithrix(g);unc(s) 0.13 ± 0.01 8.65 ± 3.55 0.96 ± 1.18 9.23 ± 2.04 3 40,989 Gamma 0.08 ± 0.00 5.52 ± 5.49 1.25 ± 2.26 4.77 ± 2.85 49 40,683 Gamma Chromatiales(o) Chromatiaceae(f) Nitrosococcus(g) 0.09 ± 0.03 5.72 ± 2.06 1.79 ± 3.34 4.80 ± 1.25 5 37,831 Gamma 0.09 ± 0.02 5.16 ± 1.76 1.09 ± 1.26 5.45 ± 0.56 6 34,906 Gamma Chromatiales(o) Ectothiorhodospiraceae(f) Thioalkalispira(g) unc(s) 0.08 ± 0.01 5.47 ± 2.29 1.19 ± 2.11 4.01 ± 1.26 7 30,128 Alpha Rhodobacterales(o) Rhodobacteraceae(f) 0.07 ± 0.01 3.10 ± 0.69 1.74 ± 0.47 4.18 ± 0.24 8 21,679 Gamma Thiohalophilus(o) unc(f) 0.06 ± 0.01 0.35 ± 0.39 5.15 ± 4.09 0.48 ± 0.36 11 18,816 Gamma Oceanospirillales(o) Alcanivoracaceae(f) Kangiella(g) unc(s) 0.05 ± 0.00 2.74 ± 0.70 1.00 ± 1.02 2.12 ± 0.43 9 13,956 Alpha Parvularculales(o) Parvularculaceae(f) Parvularcula(g) unc(s) 0.04 ± 0.00 1.75 ± 1.03 0.37 ± 0.37 2.23 ± 0.27 15 13,072 Bacter-oidetes Flavobacteria(c) Flavobacteriales(o) Flavobacteriaceae(f) Psychroserpens(g) Psychroserpens(s) 0.10 ± 0.01 1.30 ± 0.49 1.27 ± 0.27 1.24 ± 0.34 10 12,716 Gamma 0.03 ± 0.00 1.06 ± 0.43 1.68 ± 0.31 1.05 ± 0.15 12 11,127 Gamma Thiotrichales(o) Piscirickettsiaceae(f) Thiomicrospira(g) unc(s) 0.02 ± 0.01 1.66 ± 1.20 0.18 ± 0.30 1.42 ± 0.62 14 10,124 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.00 ± 0.85 0.03 ± 0.00 0.02 ± 0.00 0.02 ± 0.01 19 10,116 Beta Nitrosomonadales(o) Nitrosomonadaceae(f) Nitrosomonas(g) 0.02 ± 0.00 0.93 ± 0.13 1.22 ± 0.36 0.88 ± 0.10 47 9429 Proteo-bacteria SPOTSOCT00m83(c) unc(o) 0.02 ± 0.00 0.90 ± 0.27 1.26 ± 0.55 0.76 ± 0.08 13 9235 Gamma 0.02 ± 0.00 1.44 ± 0.69 0.22 ± 0.36 1.28 ± 0.44 17 9233 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.43 ± 1.29 0.03 ± 0.01 0.03 ± 0.00 0.02 ± 0.00 18 9070 Gamma Xanthomonadales(o) 0.03 ± 0.00 0.86 ± 0.41 0.07 ± 0.06 2.06 ± 0.79 16 8889 Alpha Rhodobacterales(o) Rhodobacteraceae(f) Roseobacter(g) 0.06 ± 0.00 0.63 ± 0.14 1.22 ± 0.28 0.81 ± 0.16 39 8024 Alpha Rickettsiales(o) SAR116(f) 4.41 ± 0.33 0.03 ± 0.00 0.02 ± 0.00 0.03 ± 0.01 OTU number . Sequence reads . Taxa (Phylum or class level for Proteobacteria) . Seawater 177,084a . 0.39 mA (Al) 308,264 . 0.17 mA 347,759 . 0.39 mA (CS) 340,069 . 1 151,809 Gamma Thiotrichales(o) Thiotrichaceae(f) unc(g);unc(s) 0.30 ± 0.01b 4.05 ± 7.08 38.91 ± 9.58 0.47 ± 0.06 4 69,288 Gamma Acidithiobacillale(o) Acidithiobacillaceae(f) Acidithiobacillus(g) unc(s) 0.17 ± 0.02 7.44 ± 2.78 2.05 ± 0.74 11.87 ± 2.26 2 61,411 Bacteroidetes Sphingobacteria(c) Sphingobacteriales(o) Flammeovirgaceae(f) Flexithrix(g);unc(s) 0.13 ± 0.01 8.65 ± 3.55 0.96 ± 1.18 9.23 ± 2.04 3 40,989 Gamma 0.08 ± 0.00 5.52 ± 5.49 1.25 ± 2.26 4.77 ± 2.85 49 40,683 Gamma Chromatiales(o) Chromatiaceae(f) Nitrosococcus(g) 0.09 ± 0.03 5.72 ± 2.06 1.79 ± 3.34 4.80 ± 1.25 5 37,831 Gamma 0.09 ± 0.02 5.16 ± 1.76 1.09 ± 1.26 5.45 ± 0.56 6 34,906 Gamma Chromatiales(o) Ectothiorhodospiraceae(f) Thioalkalispira(g) unc(s) 0.08 ± 0.01 5.47 ± 2.29 1.19 ± 2.11 4.01 ± 1.26 7 30,128 Alpha Rhodobacterales(o) Rhodobacteraceae(f) 0.07 ± 0.01 3.10 ± 0.69 1.74 ± 0.47 4.18 ± 0.24 8 21,679 Gamma Thiohalophilus(o) unc(f) 0.06 ± 0.01 0.35 ± 0.39 5.15 ± 4.09 0.48 ± 0.36 11 18,816 Gamma Oceanospirillales(o) Alcanivoracaceae(f) Kangiella(g) unc(s) 0.05 ± 0.00 2.74 ± 0.70 1.00 ± 1.02 2.12 ± 0.43 9 13,956 Alpha Parvularculales(o) Parvularculaceae(f) Parvularcula(g) unc(s) 0.04 ± 0.00 1.75 ± 1.03 0.37 ± 0.37 2.23 ± 0.27 15 13,072 Bacter-oidetes Flavobacteria(c) Flavobacteriales(o) Flavobacteriaceae(f) Psychroserpens(g) Psychroserpens(s) 0.10 ± 0.01 1.30 ± 0.49 1.27 ± 0.27 1.24 ± 0.34 10 12,716 Gamma 0.03 ± 0.00 1.06 ± 0.43 1.68 ± 0.31 1.05 ± 0.15 12 11,127 Gamma Thiotrichales(o) Piscirickettsiaceae(f) Thiomicrospira(g) unc(s) 0.02 ± 0.01 1.66 ± 1.20 0.18 ± 0.30 1.42 ± 0.62 14 10,124 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.00 ± 0.85 0.03 ± 0.00 0.02 ± 0.00 0.02 ± 0.01 19 10,116 Beta Nitrosomonadales(o) Nitrosomonadaceae(f) Nitrosomonas(g) 0.02 ± 0.00 0.93 ± 0.13 1.22 ± 0.36 0.88 ± 0.10 47 9429 Proteo-bacteria SPOTSOCT00m83(c) unc(o) 0.02 ± 0.00 0.90 ± 0.27 1.26 ± 0.55 0.76 ± 0.08 13 9235 Gamma 0.02 ± 0.00 1.44 ± 0.69 0.22 ± 0.36 1.28 ± 0.44 17 9233 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.43 ± 1.29 0.03 ± 0.01 0.03 ± 0.00 0.02 ± 0.00 18 9070 Gamma Xanthomonadales(o) 0.03 ± 0.00 0.86 ± 0.41 0.07 ± 0.06 2.06 ± 0.79 16 8889 Alpha Rhodobacterales(o) Rhodobacteraceae(f) Roseobacter(g) 0.06 ± 0.00 0.63 ± 0.14 1.22 ± 0.28 0.81 ± 0.16 39 8024 Alpha Rickettsiales(o) SAR116(f) 4.41 ± 0.33 0.03 ± 0.00 0.02 ± 0.00 0.03 ± 0.01 The bold values represent the two most abundant OTUs in each pooled library aTotal number of 16S rRNA sequence reads for pooled seawater or SSC libraries bProportion of 16S rRNA sequence reads assigned the designated OTU classification. Mean ± SD Open in new tab Most abundant OTUs (highest number of total 16S rRNA sequence reads across pooled libraries) OTU number . Sequence reads . Taxa (Phylum or class level for Proteobacteria) . Seawater 177,084a . 0.39 mA (Al) 308,264 . 0.17 mA 347,759 . 0.39 mA (CS) 340,069 . 1 151,809 Gamma Thiotrichales(o) Thiotrichaceae(f) unc(g);unc(s) 0.30 ± 0.01b 4.05 ± 7.08 38.91 ± 9.58 0.47 ± 0.06 4 69,288 Gamma Acidithiobacillale(o) Acidithiobacillaceae(f) Acidithiobacillus(g) unc(s) 0.17 ± 0.02 7.44 ± 2.78 2.05 ± 0.74 11.87 ± 2.26 2 61,411 Bacteroidetes Sphingobacteria(c) Sphingobacteriales(o) Flammeovirgaceae(f) Flexithrix(g);unc(s) 0.13 ± 0.01 8.65 ± 3.55 0.96 ± 1.18 9.23 ± 2.04 3 40,989 Gamma 0.08 ± 0.00 5.52 ± 5.49 1.25 ± 2.26 4.77 ± 2.85 49 40,683 Gamma Chromatiales(o) Chromatiaceae(f) Nitrosococcus(g) 0.09 ± 0.03 5.72 ± 2.06 1.79 ± 3.34 4.80 ± 1.25 5 37,831 Gamma 0.09 ± 0.02 5.16 ± 1.76 1.09 ± 1.26 5.45 ± 0.56 6 34,906 Gamma Chromatiales(o) Ectothiorhodospiraceae(f) Thioalkalispira(g) unc(s) 0.08 ± 0.01 5.47 ± 2.29 1.19 ± 2.11 4.01 ± 1.26 7 30,128 Alpha Rhodobacterales(o) Rhodobacteraceae(f) 0.07 ± 0.01 3.10 ± 0.69 1.74 ± 0.47 4.18 ± 0.24 8 21,679 Gamma Thiohalophilus(o) unc(f) 0.06 ± 0.01 0.35 ± 0.39 5.15 ± 4.09 0.48 ± 0.36 11 18,816 Gamma Oceanospirillales(o) Alcanivoracaceae(f) Kangiella(g) unc(s) 0.05 ± 0.00 2.74 ± 0.70 1.00 ± 1.02 2.12 ± 0.43 9 13,956 Alpha Parvularculales(o) Parvularculaceae(f) Parvularcula(g) unc(s) 0.04 ± 0.00 1.75 ± 1.03 0.37 ± 0.37 2.23 ± 0.27 15 13,072 Bacter-oidetes Flavobacteria(c) Flavobacteriales(o) Flavobacteriaceae(f) Psychroserpens(g) Psychroserpens(s) 0.10 ± 0.01 1.30 ± 0.49 1.27 ± 0.27 1.24 ± 0.34 10 12,716 Gamma 0.03 ± 0.00 1.06 ± 0.43 1.68 ± 0.31 1.05 ± 0.15 12 11,127 Gamma Thiotrichales(o) Piscirickettsiaceae(f) Thiomicrospira(g) unc(s) 0.02 ± 0.01 1.66 ± 1.20 0.18 ± 0.30 1.42 ± 0.62 14 10,124 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.00 ± 0.85 0.03 ± 0.00 0.02 ± 0.00 0.02 ± 0.01 19 10,116 Beta Nitrosomonadales(o) Nitrosomonadaceae(f) Nitrosomonas(g) 0.02 ± 0.00 0.93 ± 0.13 1.22 ± 0.36 0.88 ± 0.10 47 9429 Proteo-bacteria SPOTSOCT00m83(c) unc(o) 0.02 ± 0.00 0.90 ± 0.27 1.26 ± 0.55 0.76 ± 0.08 13 9235 Gamma 0.02 ± 0.00 1.44 ± 0.69 0.22 ± 0.36 1.28 ± 0.44 17 9233 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.43 ± 1.29 0.03 ± 0.01 0.03 ± 0.00 0.02 ± 0.00 18 9070 Gamma Xanthomonadales(o) 0.03 ± 0.00 0.86 ± 0.41 0.07 ± 0.06 2.06 ± 0.79 16 8889 Alpha Rhodobacterales(o) Rhodobacteraceae(f) Roseobacter(g) 0.06 ± 0.00 0.63 ± 0.14 1.22 ± 0.28 0.81 ± 0.16 39 8024 Alpha Rickettsiales(o) SAR116(f) 4.41 ± 0.33 0.03 ± 0.00 0.02 ± 0.00 0.03 ± 0.01 OTU number . Sequence reads . Taxa (Phylum or class level for Proteobacteria) . Seawater 177,084a . 0.39 mA (Al) 308,264 . 0.17 mA 347,759 . 0.39 mA (CS) 340,069 . 1 151,809 Gamma Thiotrichales(o) Thiotrichaceae(f) unc(g);unc(s) 0.30 ± 0.01b 4.05 ± 7.08 38.91 ± 9.58 0.47 ± 0.06 4 69,288 Gamma Acidithiobacillale(o) Acidithiobacillaceae(f) Acidithiobacillus(g) unc(s) 0.17 ± 0.02 7.44 ± 2.78 2.05 ± 0.74 11.87 ± 2.26 2 61,411 Bacteroidetes Sphingobacteria(c) Sphingobacteriales(o) Flammeovirgaceae(f) Flexithrix(g);unc(s) 0.13 ± 0.01 8.65 ± 3.55 0.96 ± 1.18 9.23 ± 2.04 3 40,989 Gamma 0.08 ± 0.00 5.52 ± 5.49 1.25 ± 2.26 4.77 ± 2.85 49 40,683 Gamma Chromatiales(o) Chromatiaceae(f) Nitrosococcus(g) 0.09 ± 0.03 5.72 ± 2.06 1.79 ± 3.34 4.80 ± 1.25 5 37,831 Gamma 0.09 ± 0.02 5.16 ± 1.76 1.09 ± 1.26 5.45 ± 0.56 6 34,906 Gamma Chromatiales(o) Ectothiorhodospiraceae(f) Thioalkalispira(g) unc(s) 0.08 ± 0.01 5.47 ± 2.29 1.19 ± 2.11 4.01 ± 1.26 7 30,128 Alpha Rhodobacterales(o) Rhodobacteraceae(f) 0.07 ± 0.01 3.10 ± 0.69 1.74 ± 0.47 4.18 ± 0.24 8 21,679 Gamma Thiohalophilus(o) unc(f) 0.06 ± 0.01 0.35 ± 0.39 5.15 ± 4.09 0.48 ± 0.36 11 18,816 Gamma Oceanospirillales(o) Alcanivoracaceae(f) Kangiella(g) unc(s) 0.05 ± 0.00 2.74 ± 0.70 1.00 ± 1.02 2.12 ± 0.43 9 13,956 Alpha Parvularculales(o) Parvularculaceae(f) Parvularcula(g) unc(s) 0.04 ± 0.00 1.75 ± 1.03 0.37 ± 0.37 2.23 ± 0.27 15 13,072 Bacter-oidetes Flavobacteria(c) Flavobacteriales(o) Flavobacteriaceae(f) Psychroserpens(g) Psychroserpens(s) 0.10 ± 0.01 1.30 ± 0.49 1.27 ± 0.27 1.24 ± 0.34 10 12,716 Gamma 0.03 ± 0.00 1.06 ± 0.43 1.68 ± 0.31 1.05 ± 0.15 12 11,127 Gamma Thiotrichales(o) Piscirickettsiaceae(f) Thiomicrospira(g) unc(s) 0.02 ± 0.01 1.66 ± 1.20 0.18 ± 0.30 1.42 ± 0.62 14 10,124 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.00 ± 0.85 0.03 ± 0.00 0.02 ± 0.00 0.02 ± 0.01 19 10,116 Beta Nitrosomonadales(o) Nitrosomonadaceae(f) Nitrosomonas(g) 0.02 ± 0.00 0.93 ± 0.13 1.22 ± 0.36 0.88 ± 0.10 47 9429 Proteo-bacteria SPOTSOCT00m83(c) unc(o) 0.02 ± 0.00 0.90 ± 0.27 1.26 ± 0.55 0.76 ± 0.08 13 9235 Gamma 0.02 ± 0.00 1.44 ± 0.69 0.22 ± 0.36 1.28 ± 0.44 17 9233 Actino-bacteria Actinobacteria(c) Acidimicrobidae(o) Acidimicrobiales(f) Acidimicrobineae(g) OCS155(s) 5.43 ± 1.29 0.03 ± 0.01 0.03 ± 0.00 0.02 ± 0.00 18 9070 Gamma Xanthomonadales(o) 0.03 ± 0.00 0.86 ± 0.41 0.07 ± 0.06 2.06 ± 0.79 16 8889 Alpha Rhodobacterales(o) Rhodobacteraceae(f) Roseobacter(g) 0.06 ± 0.00 0.63 ± 0.14 1.22 ± 0.28 0.81 ± 0.16 39 8024 Alpha Rickettsiales(o) SAR116(f) 4.41 ± 0.33 0.03 ± 0.00 0.02 ± 0.00 0.03 ± 0.01 The bold values represent the two most abundant OTUs in each pooled library aTotal number of 16S rRNA sequence reads for pooled seawater or SSC libraries bProportion of 16S rRNA sequence reads assigned the designated OTU classification. Mean ± SD Open in new tab Although there was overlap between the three SSCs biofilms, the bacterial communities from the two high-current SSC libraries were similar to each other with respect to the relative abundance of bacterial taxa present, more so than either one was to the low-current SSC libraries (Fig. 4; Table 3). Specifically, a single OTU, representing an unclassified member of the family Thiotrichaceae (turquois bars), dominated all four low-current SSC biofilms (turquois bars) representing 25–50% of each library (Fig. 4a). In the two high-current SSC biofilms, this OTU represented less than 1% of the sequences for seven of the eight libraries (Fig. 4a). Membership of the Thiotrichaceae family includes the sulfur-oxidizing genera Beggiatoa [36] and Thiothrix, both of which contain species that form symbiotic relationships with other organisms [15]. However, this OTU could not be classified lower than the family level, thus little can be said about its selective colonization of low-current versus high-current stainless steel. Measures of alpha diversity are summarized in Table 2. All SSC biofilm libraries were similar to each other in richness (i.e., total number of OTUs present) as indicated by the overlap in Chao1 indices. However, the four low-current SSC libraries were less diverse (i.e., relative abundance of OTUs present) than both types of high-current SSC libraries based on Shannon (p < 0.01) and Simpson diversity indices (p < 0.001) (Table 2). These measurements and subsequent beta diversity measurements are based off of the random subsampling and comparison of 40,000 sequences from each library. OTU distribution and abundance across all libraries was directly compared using jackknifed UPGMA hierarchical clustering based on the weighted unifrac algorithm (Fig. 4b). A dendrogram supporting the hierarchical clustering of the seawater and the SSC biofilms was generated (Fig. 4b, left side) and a heatmap was used to illustrate OTU distribution and abundance across all libraries (Fig. 4b, right side). The two seawater libraries (1,2) were the first to separate demonstrating that specific bacterial taxa from the seawater preferentially colonized SSCs. Next, the low-current SSC libraries (7,8,9) separated. These data are consistent with the DGGE analysis illustrating that the dominant taxa present in low-current SSC biofilms were different than that in the higher current SSC biofilms. Libraries from the two types of high-current biofilms clustered together (3,4,6,11,12,13,14) and (Fig. 5), suggesting that they were similar to each other with respect to dominant OTU distribution and abundance. The more abundant OTUs (Fig. 4b, right side: red, orange, yellow, green, and light blue) in high-current SSC biofilm libraries were at background levels (dark blue) in the low-current SSC biofilm libraries demonstrating that the same taxa were in fact present across all biofilms, but at very different proportions in high versus low-current cathodes. Fig. 5 Open in new tabDownload slide PCoA summary plot of seawater and SSC biofilm libraries. UPGMA clustering based on the weighted unifrac algorithm. Seawater (red), 0.39 mA (Al) SSCs (blue), 0.39 mA (CS) SSCs (orange), and 0.17 mA SSCs (green). Dashed lines are included to emphasize clustering Sequences were pooled together and the top 20 most abundant OTUs in are listed in Table 3. The two dominant OTUs present in low-current SSC biofilms belong to Gammaproteobacteria and include unclassified Thiotrichaceae and unclassified Thiohalophilus at 39 and 5%, respectively. Neither OTU was dominant in the two higher current biofilms. Rather, the genera Acidithiobacillus, also Gammaproteobacteria and Flexithrix (Phylum Bacteroidetes), were dominant, both at approximately 7–12%. Together, these data demonstrate that different bacteria populate the biofilms of stainless steel dependent upon the applied current (i.e., anode to which it was galvanically coupled). Finally, to attempt a deeper classification of the dominant Gammaproteobacteria (OTU1), a small clone library analysis was performed on a nearly full-length 16S rRNA gene segment using metagenome DNA pooled from the four low-current SSCs. Five of the ten clone sequences matched 280/283 bp of the dominant OTU1 sequence and had more than 99% identity to each other (total of 6 mismatches/1361 bp). The phylogenetic relationship of the consensus of the five clones was assessed using three treeing methods. All clustered the same taxa (Fig. 5) and all showed the greatest similarity to sequences from two uncultured bacteria, accession numbers JN977291 and AB530210. Both samples were obtained from marine sediments, JN977291 from Jiaozhou Bay, China, and AB530210 from Tokyo Bay, Japan. This dominant Gammaproteobacterium remains therefore as an “unclassified Thiotrichaceae” (Fig. 6). Fig. 6 Open in new tabDownload slide The phylogenetic relationship of 16S rRNA nucleotide sequence of OTU1 (unclassified Thiotrichaceae) to those of selected members of the families Thiotrichaceae, Piscirickettsiaceae, and Francisellaceae (Gammaproteobacteria, Order Thiotrichales) is shown as the optimal neighbor-joining tree. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The percentage of replicate trees in which the associated taxa clustered together at greater than 60% in the bootstrap test (1000 replicates) is shown next to the branches [14]. The same taxa were clustered together at greater than 60% in the bootstrap test (1000 replicates) when trees were constructed using the Maximum Likelihood and Maximum Parsimony methods. Evolutionary analyses were conducted in MEGA6 [41]. Bar 0.02 substitutions per site Conclusion Differences between the high and low-current SSC biofilm communities were observed. Although classification of dominant taxa was not achieved at the species levels for most OTUs, there was enough difference between the approximately 300 bp16S rRNA sequences to conclude that the dominant taxa were in fact different. The authors amplified almost full-length 16S rRNA representing OTU1, but higher resolution was still not achieved because a close representative was not present in the databases (SILVA or GENBANK). This is not uncommon for environmental microbiologists, as many microorganisms in the environment have yet to be cultured, classified, and added to curated databases. Thus, the approximately 300 bp 16S rRNA used for Illumina DNA-seq was comparable to full-length 16S rRNA sequence in taxa resolution in this study. Whether or not the shorter Illumina reads would provide the same level of resolution as the longer Sanger reads in other environmental settings would require investigation. DNA-seq studies, similar to the one presented here, could compliment current methods used to study the influence of marine biofilms on cathodic metals [11, 13]. For example, a recent report by Eashwar et al. concluded that sunlight and biofilms had opposite affects on cathodic kinetics in galvanic couples [11]. However, an analysis of bacteria in the biofilm was not explored. It would have been interesting to compare the bacterial community profiles of biofilms obtained from the cathodes that were set at the different sunlight exposure intensities to determine if there was a correlation between the dominant taxa in the biofilm and cathodic aggressiveness. In the study presented here, three types of cathodic biofilm communities were shown to be very different in structure (relative abundance of dominant bacterial taxa) from that of the planktonic seawater. It is proposed that the SSC biofilm community is influenced by the electrochemical current (i.e., to which anode coupon the stainless steel was galvanically coupled), although the extent of this variation has yet to be examined. Unintentional galvanic cells would not only accelerate corrosion of the anode, but may also influence the biofilm composition of that substrate. This leads to an added level of complexity when studying marine microbiology and ecology. More work is needed to determine if this pattern of sensitivity to electrochemical current applies to other metal alloys, such as the carbon steel, copper nickel, and aluminum anodes and to what extent. Acknowledgements This work was funded by Carderock Division under the In-House Laboratory Independent Research (ILIR) program, Program Element 0601152N, managed by the NSWC Carderock Division Director of Research for the Office of Naval Research under Work Request N0001414WX20148. The authors would like to thank Joseph Scudlark and Horn Ewart from the University of Delaware College of Earth, Ocean, and Environment for their assistance setting up the flowing seawater tank. Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. Ethical approval This work did not involve the use of human participants or animals. References 1. Aronesty E (2011) ea-utils: command-line tools for processing biological sequencing data. https://github.com/ExpressionAnalysis/ea-utils. Accessed 9 Nov 2016 2. Aronesty E Comparison of sequencing utility programs TOBIOIJ 2013 7 1 8 10.2174/1875036201307010001 Google Scholar Crossref Search ADS WorldCat 3. Altschul SF , Gish W, Miller W, Myers EW, Lipman DJ Basic local alignment search tool J Mol Biol 1990 215 403 410 10.1016/S0022-2836(05)80360-2 Google Scholar Crossref Search ADS PubMed WorldCat 4. ASTM G82-98. Reapproved 2014. Standard guide for development and use of a galvanic series for predicting galvanic corrosion performance. West Conshohocken (PA): ASTM International (US) 5. Axe D (2011) Builder blames navy as brand-new warship disintegrates. In: Conde Nast Digital, 23 June 2011. Web. 26 March 2014. Available via Wired.com. http://www.wired.com/dangerroom/2011/06/shipbuilder-blames-navy-as-brand-new-warship-disintegrates. Accessed 30 March 2014 6. Bolger AM , Lohse M, Usadel B Trimmomatic: a flexible trimmer for Illumina sequence data Bioinformatics 2014 30 15 2114 2120 10.1093/bioinformatics/btu170 4103590 Google Scholar Crossref Search ADS PubMed WorldCat 7. Caporaso JG , Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JIet al. QIIME allows analysis of high-throughput community sequencing data Nat Methods 2010 7 5 335 336 10.1038/nmeth.f.303 3156573 Google Scholar Crossref Search ADS PubMed WorldCat 8. Carteau D , Vallee-Rehel K, Linossier I, Quiniou F, Davy R, Compere C, Delbury M, Fay F Development of environmentally friendly antifouling paints using biodegradable polymer and lower toxic substances Prog Org Coat 2014 77 2 485 493 10.1016/j.porgcoat.2013.11.012 Google Scholar Crossref Search ADS WorldCat 9. Dexter SC (2005) Biological effects. In: Baboian R (ed) Corrosion tests and standards: applications and interpretation, 2nd edn, pp 509–522 10. Duncan KE , Gieg LM, Parisi VA, Tanner RS, Tringe SG, Bristow J, Suflita JM Biocorrosive thermophilic microbial communities in Alaskan North Slope Environ Sci Technol 2009 43 7977 7984 10.1021/es9013932 Google Scholar Crossref Search ADS PubMed WorldCat 11. Eashwar M , Lakshman KA, Sreedhar G, Kennedy J, Surech BR Stainless steel in coastal seawater: sunlight counteracts biologically enhanced cathodic kinetics Biofouling 2014 30 8 929 939 10.1080/08927014.2014.952632 Google Scholar Crossref Search ADS PubMed WorldCat 12. Edgars RC UPARSE: highly accurate OTU sequences from microbial amplicon reads Nat Methods 2013 10 10 996 998 10.1038/nmeth.2604 Google Scholar Crossref Search ADS PubMed WorldCat 13. Erable B , Lacroix R, Etcheverry L, Feron D, Delia ML, Bergel A Marine floating microbial fuel cell involving aerobic biofilm on stainless steel cathodes Bioresour Technol 2013 142 510 516 10.1016/j.biortech.2013.05.063 Google Scholar Crossref Search ADS PubMed WorldCat 14. Felsenstein J Confidence limits on phylogenies: an approach using the bootstrap Evolution 1985 39 783 791 10.2307/2408678 Google Scholar Crossref Search ADS PubMed WorldCat 15. Flot JF , Worheide G, Dattagupta S Unsuspected diversity of Niphargus amphipods in the chemoautotrophic cave ecosystem of Frasassi, central Italy BMC Evol Biol 2010 10 1 171 10.1186/1471-2148-10-171 2896373 Google Scholar Crossref Search ADS PubMed WorldCat 16. Good IJ The population frequencies of species and the estimation of population parameters Biometrika 1953 40 237 264 10.1093/biomet/40.3-4.237 Google Scholar Crossref Search ADS WorldCat 17. Gu J, Mitchell R (2013) Biodeterioration. In: Rosenberg E et al (eds) The prokaryotes—applied bacteriology and biotechnology, Springer, Berlin. doi:10.1007/978-3-642-31331-8_31 18. Gutarowska B , Celikkol-Aydin S, Bonifay V, Otlewska A, Aydin E, Oldham A, Brauer J, Duncan K, Adamiak J, Sunner J, Beech I Metabolomic and high-throughput sequencing analysis—modern approach for the assessment of biodeterioration of materials from historic buildings Front Microbiol 2015 6 979 10.3389/fmicb.2015.00979 4586457 Google Scholar Crossref Search ADS PubMed WorldCat 19. Haas BJ , Gevers D, Earl AM, Feldgarden M, Ward DV, Giannoukos G, Ciulla D, Tabbaa D, Highlander SK, Sodergren Eet al. Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons Genome Res 2011 21 3 494 504 10.1101/gr.112730.110 3044863 Google Scholar Crossref Search ADS PubMed WorldCat 20. Hasson DF, Crowe CR (eds) (1988) Materials for marine systems and structures. Treatise on Material Science and Technology, vol 28. Academic Press, New York 21. Henne K , Kahlisch L, Brettar I, Hofle MG Analysis of structure and composition of bacterial core communities in mature drinking water biofilms and bulk water of a citywide network in Germany Appl Environ Microbiol 2012 78 10 3530 3538 10.1128/AEM.06373-11 3346358 Google Scholar Crossref Search ADS PubMed WorldCat 22. Kable: Naval-Technology.com c2016. Littoral Combat Ship (LCS), United States of America. http://www.naval-technology.com/projects/littoral/. Accessed 12 Jan 2016 23. Lazzaro A , Hilfiker D, Zeyer J Structures of microbial communities in Alpine soils: seasonal and elevational effects Front Microbiol 2015 6 1330 10.3389/fmicb.2015.01330 4660872 Google Scholar Crossref Search ADS PubMed WorldCat 24. Lee JS , Ray RI, Lemieux EJ, Falster AU, Little BJ Evaluation of carbon steel corrosion under stagnant water conditions Biofouling 2004 20 4/5 237 247 10.1080/08927010400013274 Google Scholar PubMed OpenURL Placeholder Text WorldCat 25. Lee JS , Ray RI, Little BJ, Duncan KE, Oldham AL, Davidova IA, Suflita JM Sulphide production and corrosion in seawaters during exposure to FAME diesel Biofouling 2012 28 5 465 478 10.1080/08927014.2012.687723 Google Scholar Crossref Search ADS PubMed WorldCat 26. Lozupone CA , Hamady M, Kelley ST, Knight R Quantitative and qualitative beta diversity measures lead to different insights into factors that structure microbial communities Appl Environ Microbiol 2007 73 5 1576 1585 10.1128/AEM.01996-06 1828774 Google Scholar Crossref Search ADS PubMed WorldCat 27. Michelou VK , Caoirasi JG, Knight R, Palumbi SR The ecology of microbial communities associated with Macrocyctis pyrifera PLoS One 2013 8 6 e67480 10.1371/journal.pone.0067480 3686729 Google Scholar Crossref Search ADS PubMed WorldCat 28. Muyzer G , deWaal EC, Uitterlinden AG Profiling of complex microbial populations by denaturant gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA Appl Environ Microbiol 1993 59 3 695 700 202176 Google Scholar Crossref Search ADS PubMed WorldCat 29. Nei M , Kumar S Molecular evolution and phylogenetics 2000 New York Oxford University Press Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 30. Oldham AL , Drilling HS, Stamps BW, Stevenson BS, Duncan KE Automated DNA extraction platforms offer solutions to challenges of assessing microbial biofouling in oil production facilities AMB Express 2012 2 1 60 10.1186/2191-0855-2-60 3539857 Google Scholar Crossref Search ADS PubMed WorldCat 31. Pruesse E , Quast C, Knittel K, Fuchs BM, Ludwig WG, Peplies J, Glockner FO SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB Nucleic Acids Res 2007 35 21 7188 7196 10.1093/nar/gkm864 2175337 Google Scholar Crossref Search ADS PubMed WorldCat 32. Reinhart FM, Jenkins JF (1972) Technical Note N-1213. Corrosion of materials in surface seawater after 12 and 18 months of exposure. Naval Civil Engineering Laboratory, Port Hueneme 33. Rousseau C , Baraud F, LeleyterL Gil O Cathodic protection by zinc sacrificial anodes: impact on marine sediment metallic contamination J Hazard Mater 2009 167 953 958 10.1016/j.jhazmat.2009.01.083 Google Scholar Crossref Search ADS PubMed WorldCat 34. Saitou N , Nei M The neighbor-joining method: a new method for reconstructing phylogenetic trees Mol Biol Evol 1987 4 406 425 Google Scholar PubMed OpenURL Placeholder Text WorldCat 35. Santegoeds CM , Damgaard LR, Hesselink G, Zopfi J, Lens P, Muyzer G, deBeer D Distribution of sulfate-reducing and methanogenic bacteria in anaerobic aggregates determined by microsensor and molecular analyses Appl Environ Microbiol 1999 65 10 4618 4629 91616 Google Scholar Crossref Search ADS PubMed WorldCat 36. Schmidt TM , Arieli B, Cohen Y, Padan E, Strohl WR Sulfur metabolism in Beggiatoa alba J Bacteriol 1987 169 12 5466 5472 10.1128/jb.169.12.5466-5472.1987 213973 Google Scholar Crossref Search ADS PubMed WorldCat 37. Sheik CS , Stevenson EI, Den Uyl PA, Arendt CA, Aciego SM, Dick GJ Microbial communities of the Lemon Creek glacier show subtle structural variation yet stable phylogenetic composition over space and time Front Microbiol 2015 6 495 10.3389/fmicb.2015.00495 4438255 Google Scholar Crossref Search ADS PubMed WorldCat 38. Suflita JM , Aktas DF, Oldham AL, Perez-Ibarra BM, Duncan KE Molecular tools to track bacteria responsible for fuel deterioration and microbiologically influenced corrosion Biofouling 2012 28 9 1003 1010 10.1080/08927014.2012.723695 Google Scholar Crossref Search ADS PubMed WorldCat 39. Sultana AT , Babaauta JT, Beyenal H Electrochemical biofilm control: a review Biofouling 2015 31 9–10 745 758 10.1080/08927014.2015.1105222 4668948 Google Scholar PubMed OpenURL Placeholder Text WorldCat 40. Tamura K , Nei M, Kumar S Prospects for inferring very large phylogenies by using the neighbor-joining method Proc Nat Acad Sci (USA) 2004 101 11030 11035 10.1073/pnas.0404206101 Google Scholar Crossref Search ADS WorldCat 41. Tamura K , Stecher G, Peterson D, Filipski A, Kumar S MEGA6: molecular evolutionary genetics analysis version 6.0 Mol Biol Evol 2013 30 2725 2729 10.1093/molbev/mst197 3840312 Google Scholar Crossref Search ADS PubMed WorldCat 42. Wang Q , Garrity GM, Tiedje JM, Cole JR Naïve Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy Appl Environ Microbiol 2007 73 16 5261 5267 10.1128/AEM.00062-07 1950982 Google Scholar Crossref Search ADS PubMed WorldCat 43. Wagner P , Little B, Hart K, Ray R, Thomas D, Trzaskoma-Paulette P, Lucas K Environmental fate of sacrificial zinc anodes and influence of a biofilm Int Biodeterior Biodegrad 1996 37 3–4 151 157 10.1016/S0964-8305(96)00013-3 Google Scholar OpenURL Placeholder Text WorldCat Crossref 44. Wuchter C , Banning E, Mincer TJ, Drenzek NJ, Coolen MJ Microbial diversity and methanogenic activity of Antrim Shale formation waters from recently fractured wells Front Microbiol 2013 4 367 10.3389/fmicb.2013.00367 3853793 Google Scholar Crossref Search ADS PubMed WorldCat 45. Xiao Y , Zheng Y, Wu S, Zhang E, Chen Z, Liang P, Huang X, Yang Z, Ng I, Chen B, Zhao F Pyrosequencing reveals a core community of anodic bacterial biofilms in bioelectrochemical systems from China Front Microbiol 2015 6 1410 10.3389/fmicb.2015.01410 4679932 Google Scholar Crossref Search ADS PubMed WorldCat 46. Yue J , Cao Y Corrosion prevention by applied coatings on aluminum alloys in corrosive environments Int J Electrochem Sci 2015 10 5222 5237 Google Scholar OpenURL Placeholder Text WorldCat 47. Zuo R Biofilms: strategies for metal corrosion inhibition employing microorganisms Appl Microbiol Biotechnol 2007 76 6 1245 1253 10.1007/s00253-007-1130-6 Google Scholar Crossref Search ADS PubMed WorldCat © Society for Industrial Microbiology 2017 This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © Society for Industrial Microbiology 2017
TI - Molecular methods resolve the bacterial composition of natural marine biofilms on galvanically coupled stainless steel cathodes
JF - Journal of Industrial Microbiology and Biotechnology
DO - 10.1007/s10295-016-1887-7
DA - 2017-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/molecular-methods-resolve-the-bacterial-composition-of-natural-marine-cbulmEDTC8
SP - 167
EP - 180
VL - 44
IS - 2
DP - DeepDyve
ER -