Revisiting CARD-FISH for detection of Vibrio spp. in natural seawater

Vibrio is a bacterial genus widely distributed in natural aquatic systems. Some Vibrio species are pathogenic and can cause severe diseases in both marine organisms and humans. Previous studies revealed a link between the current climate change and increased incidence of the Vibrio-associated diseases recently causing sanitary, economic and/or ecological problems worldwide. The conventional culture-based methods (e.g. cultivation of TCBS agar) used to monitor the presence of Vibrio spp. in environmental samples are not always straightforward and can underestimate the number of cells, especially if the microbial population contains a fraction of ‘dormant’ cells (e.g. cells in Viable but not Culturable (VBNC) state). However, this problem can be overcome by using alternative culture-free approaches such as Catalyzed Reporter Deposition-Fluorescence In situ Hybridization (CARD-FISH). To optimize CARD-FISH for efficient and reliable detection of Vibrio spp. in environmental samples, we have used both computer-assisted and experimental approaches. Our results demonstrate that the use of the optimized protocol along with a very specific probe, ViB572a, can offer the high sensitivity and selectivity of CARD-FISH detection of marine vibrios in natural seawater.


Introduction
Global warming has lately emerged as a major concern for research and healthcare communities. The recent increase in the average temperature of earth resulted in melting the polar ice caps and rise of the sea level [1]. In turn, an increase in temperature of aquatic systems was shown to affect the balance and survival of organisms [2], thereby altering their diversity, structure and interaction in ecosystems [3]. Thompson et al. [4] suggested that the observed climate change has a direct impact on the current spread of invasive pathogenic species, including those of the Vibrio genus. As a result, there is an increase in the incidence of diseases caused by the use of water contaminated by these microorganisms [2,5,6], even in European countries where these diseases were not common [2]. Moreover, the increased incidence of vibriosis in non-endemic regions [6] has already led to great economic loss [6][7][8], thus causing public concerns around the globe.
The ubiquity and spread of Vibrio species indicate that these bacteria can easily adapt to the changing environment caused by the current climate change [9,10]. Although there are numerous studies addressing the impact of global warming on adaptation and survival of Vibrio spp. [5,6,9,11], still little is unknown about how these environmental changes can alter, directly or indirectly, the structure and abundance of Vibrio communities and their pathogenicity in various aquatic ecosystems [6].
The ever increasing presence of vibrios in aquatic systems as well as their appearance in areas that previously lacked the detectable levels of these microorganisms indicate the urgent need to monitor the abundance and distribution of Vibrio spp. worldwide. However, the direct detection of vibrios run into the same general problems associated with enumeration of bacteria in natural environments. To some extent, these problems are caused by the continuous revision of bacterial taxonomy [4,12], the inability to culture some environmental microorganisms and/or their entry into the Viable But Non Culturable (VBNC) state (nonculturable but metabolically active cells) [13][14][15], the use of culture media often formulated for clinical diagnostics [16,17], etc. To overcome the limitations of culture-based methods, different culture-independent techniques have been developed for detection and enumeration of microbial species [14,[18][19][20]. Among these techniques, Fluorescence In situ Hybridization (FISH) became one of the methods frequently used to detect and quantify microbial species in complex environmental samples [21,22]. The original FISH approach consists in hybridization of fluorescently-labeled oligonucleotide probes with 16S rRNA of microbial cells fixed on a solid support (e.g. membrane) followed by detection of fluorescent signals by epifluorescence microscopy [21,23,24]. However, the sensitivity of standard FISH protocol is not always enough to readily detect microorganisms that are grown in adverse environments (e.g. limitation of nutrients) and therefore have a reduced number of ribosomes [21]. To improve the sensitivity of conventional FISH, a variant of this technique, CARD-FISH (Catalyzed Reporter Deposition Fluorescence In situ Hybridization) was developed [25,26]. The use of CARD-FISH can increase the fluorescence emitted by cells up to 20 times in comparison to the standard FISH [27], thus making it possible to detect bacteria that could not be visualized by the conventional FISH technique [28]. CARD-FISH seems to be an ideal method for the study of marine microorganisms such as Vibrio spp., which usually causes detection problems [28]. Although the ability of this method to distinguish microorganisms depends on the specificity of the probe used, the adequate and specific probe design is often difficult to achieve. This is particularly challenging in the case of the Vibrio genus, known for its enormous diversity [4,29]. A number of Vibrio-specific probes reported previously [28,30-36 and others] show different specificity and have not been scrutinized for their ability to reliably detect and enumerate Vibrio species in complex environmental samples. The main objective of this study was to adapt the CARD-FISH technique for monitoring the presence of Vibrio spp. in seawater and demonstrate the efficiency of the optimized protocol and probes in detection of marine vibrios.

Selection of a specific probe for the detection and enumeration of Vibrio spp.
To retrieve the most updated articles (primarily published during the last 20 years) dealing with the use of CARD-FISH for detection of Vibrio spp., several databases (i.e. Science Direct, Springer, Google Scholar, NCBI and others) were systematically searched by using combinations of keywords including "Vibrio", "detection", "specific probe", "CARD-FISH" and "environmental".
Based on the information obtained, 12 oligonucleotide probes targeting specific sequences within Vibrio 16S rRNA genes were pre-selected. As the preference was given to those probes that could potentially hybridize with most of the species belonging to the genus Vibrio, only three probes were selected (i.e. GV841 [30][31][32]34,36], Vib-16S-1 [35] and VIB57a [31]), whereas the remaining 9 probes were discarded.
The coverage and specificity of the selected probes were further evaluated using the online tools such as TestProbe (https://www.arb-silva.de/search/testprobe/) [37], which is part of the SILVA rRNA database project (hereafter SILVA); as well as the computer Basic Local Alignment Search Tool (BLAST) (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Since in silico analysis (SILVA and BLAST) has shown that the Vib-16S-1 probe did not meet the above criteria, it was discarded too and only probes GV841 and VIB572a were further used in this study.

Bacterial strains and natural seawater samples
To test and optimize CARD-FISH detection of Vibrio spp., a number of Gram negative (including Vibrio species) and Gram positive bacteria along with one eukaryotic microorganism were used as positive and negative controls. Vibrio spp. included V. harveyi ATCC 14126 and 4 environmental isolates of V. harveyi, V. cyclitrophicus, V. kanaloae and V. tubiashii. The latter were isolated from seawater samples collected from Port of Armintza in the North of Spain (43º 26' 24" N; 2º 54' 24" W) by selection on Thiosulfate Citrate Bile salts Sucrose Agar plates (TCBS, Fluka). Gram negative bacteria that were oxidase-positive, were further characterized by using API 20E test (Biomerieux). The nature of Vibrio strains identified in phenotypical and biochemical assays was further verified by genotyping. Namely, multilocus sequence analysis (MLSA) was performed by analyzing the sequence of 8 genes including the 16S rRNA gene and 7 housekeeping genes (gapA, ftsZ, topA, mreB, gyrB, pyrH and recA). The corresponding fragments of these genes were PCR amplified and sequenced at the Genomics and Proteomics Core Facility-SGIKER (University of the Basque Country). The resulting sequences were subjected to BLAST search against the NCBI database to complete the assignment of each Vibrio isolate to a particular taxon as described in previous studies [38][39][40].
Apart from Vibrio spp., other Gram-negative (Escherichia coli ATCC 27325, Citrobacter freundii ATCC 13316, Pseudomonas aeruginosa ATCC 27853 and Serratia marcescens ATCC 274) and some Gram-positive (Bacillus subtilis ATCC 21770, Micrococcus luteus ATCC 10240 and Staphylococcus aureus ATCC 6538P) bacteria along with Saccharomyces cerevisiae ATCC 9763 were included as negative controls. The strains used were stored at -80ºC using the Pro Lab Diagnostics™ Microbank™ system. Recovery and growth conditions were adjusted to the specific microorganism. Namely, Vibrio species were grown in Marine Broth (MB, AppliChem) at 26ºC; whereas B. subtilis, M. luteus and S. marcescens in Tryptone Soy Broth (TSB, Oxoid Ltd.) at 30ºC; E. coli, C. freundii, P. aeruginosa and St. aureus in TSB at 37ºC and S. cerevisiae in Potato Dextrose Broth (Oxoid Ltd.) at 20ºC. Cells grown for 24 h were harvested by centrifugation (4,000 g, 20 min) and then suspended in sterile seawater (collected from the Armintza port sampling station and treated as explained below) to obtain cell suspensions A to G listed in Table 1. The final microbial density for these suspensions was 10 6 cells ml -1 . To perform the microorganism mixtures (Table 1; suspensions C, D, E and F), 25 ml of seawater were inoculated with equal volumes of each individual microorganism suspension. The final density of each of them was approximately 2 10 5 cells ml -1 . These suspensions were further used to test the efficiency and specificity of the probes selected for detection and quantification of Vibrio spp. To prepare sterile seawater for cell suspensions, it was consecutively filtrated through ester membranes (Merck Millipore) with a pore size of 8, 0.8, 0.45 and 0.22 µm, respectively, and then autoclaved (121ºC, 20 min).
Besides testing CARD-FISH on pure cultures and their defined mixtures, detection of Vibrio spp. was also carried out with seawater samples collected from three sampling sites of the Monitoring Network of the Ecological Status of Transitional and Coastal Waters of the Autonomous Community of the Basque Country (AZTI Tecnalia). The collection sites are located in coastal (43º 23´ 1" N; 3º 4´ 50" W), open sea (43º 24´ 0" N; 2º 18´ 0" W) and estuarine (43º 22´ 15" N; 11º 47´ 20" W) areas. The seawater samples were kept refrigerated (4ºC) immediately after collection and the analysis was carried out within a 24 h period. Namely, the samples were processed to determine the total number of bacteria (Total Direct Count, TDC) according to Hobbie et al. [41], whereas vibrios were enumerated by using CARD-FISH (see below). The number of culturable chemoorganotrophic bacteria and the number of culturable presumptive Vibrio spp. were estimated by plating on Marine Agar (MA, AppliChem) and TCBS Agar, respectively. The plates were incubated at 26ºC for 48-96 h. If necessary, the cell suspensions were diluted with sterile saline solution (NaCl, 1.94% [w/v]) prior plating.
The quantification of total Vibrio spp. was performed using the CARD-FISH protocol described by Pernthaler et al. [26] with some modifications. 25 ml of cell suspensions and 200 ml of seawater samples were fixed with 25 ml or 100 ml of 4% [v/v] paraformaldehyde and kept refrigerated at 4ºC for 24 h. Then, each sample was filtered onto polycarbonate filters (type GTTP; 0.2 μm pore size; 47 mm diameter, Merck Millipore), and the filters were kept frozen at -20°C until further analyses.
To avoid cell loss during the permeabilization and hybridization, the filters were impregnated from both sides in 0.1% agarose solution at 35-40ºC. The endogenous peroxidases were deactivated with HCl (0.01 M, for 10 min). The filters were washed with sterile 1X Phosphate Buffered Saline (PBS, Sigma-Aldrich) and then with MilliQ water.
The permeabilization was carried out for 1 h at 37ºC in 2 ml of lysozyme solution (10 g l -1 ). This concentration of lysozyme solution double that recommended by Pernthaler et al. [26] and was selected based on the results of pilot experiments (high intensity of green fluorescence emitted). The filters were then washed with sterile MilliQ water and absolute ethanol and were afterwards, kept frozen at -20°C until further analyses. The samples containing Gram-positive bacteria were processed differently. Namely, following treatment with lysozyme, the filters were immersed in acromopeptidase solution (final concentration 60 U ml -1 ) with NaCl 0.01 M and Tris-HCl 0.01 M, pH 8.0 (for 30 min at 37ºC) and then washed twice with sterile MilliQ water and absolute ethanol.
The hybridization was carried out for 2 h in 900 µl of hybridization buffer containing 30% formamide (FM), at 46ºC with VIB572a probe [31] and at 49ºC with GV 841 probe [30]. The HPR-labeled probes were purchased from Biomers.net GmbH. They were diluted with sterile MilliQ water to a final concentration of 50 ng µl -1 , aliquoted into 30 μl aliquots and stored at -20ºC until further use. The specific hybridization conditions for each probe are indicated in Table 2. The HPR-labeled probes (50 ng µl -1 ) were used at dilutions 1:300 and 1:100 for filters containing cells from axenic cultures and for mixed cultures (or seawater), respectively. Following hybridization with VIB572a or GV 841 probes, the filters were washed with 640 µl of 5 M NaCl for 5 min at 48ºC or 42ºC, respectively, and then, with 1X PBS for 10-15 min at room temperature. The signal amplification and all subsequent steps were performed in dark. To amplify the signal, the filters were overlaid with solution containing 1 ml of Amplification Buffer, 10 μl of 100X H2O2 stock (30% [v/v]) and 4 μl of reconstituted tyramid labeled with Alexa Fluor 488 (0.5 mg ml -1 ) (Thermo Fisher Scientific) followed by incubation in dark at 42ºC for 15 min. Then, the filters were consecutively washed with 1X PBS for 10 min, running sterile distilled water (6 min), sterile MilliQ water and absolute ethanol.
To observe microorganisms, filters were stained with DAPI-Mix (1 μg ml -1 , Thermo Fisher Scientific) and examined with an Eclipse E-400 epifluorescence microscope (Nikon), equipped with filter blocks consisting of a B-2A filter (EX 450-490 excitation filter, DM 505 dichroic mirror and BA 520 barrier filter) and of a UV-2A filter (excitation filter EX 330-380, dichroic mirror DM 400 and barrier filter BA 420). The cells hybridized with the CARD-FISH probes and microorganisms stained with DAPI emitted green and blue fluorescence, respectively, detected by using B-2A and UV-2A filters. 50 fields were routinely used to enumerate each type microbial cells and calculate the number of bacteria ml -1 .

Statistical analysis
All the results obtained in this study were managed with Microsoft Excel. In the experiments performed to test the suitability of the probes selected for the CARD-FISH protocol tested here, cell densities were transformed into percentages. To determine the presence of the Vibrio genus in various seawater samples, the results obtained (cell densities) were transformed to decimal logarithms. In both cases, the experiments were carried out in triplicate, and the variation coefficients between replicates were found to be lower than 12%. Differences between means were determined by analysis of variance. Any variables with p ≤ 0.05 were assumed to be significant.

Results
According to the procedure described in Materials and Methods and results obtained by SILVA, three perspective CARD-FISH probes (i.e. Vib-16S-1, GV841 and VIB572a) were selected. Although the specificity of all 3 probes was similar towards Vibrio spp., the Vib-16S-1 probe, unlike GV841 and VIB572a, additionally showed the complete coverage and high specificity (99.3%) towards genomic regions of other species (e.g. Gayadomonas, Gallaecimonas and Candidatus Photodesmus and enterosymbionts 10 group). Furthermore, the data obtained by BLAST search indicated that the Vib-16S-1 probe could also be complementary to non-16S rRNA regions present in the genomes of some Vibrio species and non-related genera (e.g. Oceanisphoera, Avibacterium and Catenovulum) occasionally with complete coverage and identity. Therefore, given its broad cross-reactivity extended to non-Vibrio species revealed in silico (SILVA analysis and BLAST searches), the Vib-16S-1 probe was discarded and only GV841 and VIB572a probes were selected for further assays. The latter were performed with different microbial populations including both positive (group A and B) and negative (group C and D) controls. Table 3 summarizes the results of this analysis. As seen in Table 3, the efficiency of hybridization obtained with both probes (GV841 and VIB572a) towards V. harveyi strains in suspensions A and B (positive controls) was close to 100% (Figure 1). In other words, the numbers of cells counted using the UV-2A filter (i.e. the total number of cells detected after staining with DAPI) and those that had hybridized with the CARD-FISH probes were nearly identical. Furthermore, the intensity of green fluorescence emitted by the probe fluorophore (i.e. ALEXA Fluor 488) was very high. The specificity of the probes was verified using suspensions C and D (Table 1). For suspension C, composed of non-Vibrio bacteria (negative control), cells did not hybridized with GV841 or VIB572a and did not produce any fluorescence (data not shown). For suspension D (a mixture of non-Vibrio bacteria and V. harveyi ATCC 14126), 7.5% (± 0.1) and 21% (± 0.9) yielded fluorescence with probes GV841 and VIB572a, respectively (Table 3, Figure 2). The morphological characteristics of the fluorescent cells corresponded to those of V. harveyi ATCC 14126 and their percentage corresponded to that in the initial mix. As anticipated, epifluorescence microscopy using the UV-2A filter confirmed the presence of cells with different morphologies (coccoid, bacilar and coccobacilar) in C and D suspensions stained with DAPI. In addition, the CARD-FISH protocol was also tested on suspensions E and F (Table 1), composed of different Vibrio species. While the results obtained with suspension E, composed of two species previously tested (V. harveyi ATCC 14126 and environmental V. harveyi) and three new isolates (V. cyclitrophicus, V. kanaloae and V. tubiashii) demonstrated the high detection rate (nearly 100%) with VIB572a, the detection efficiency was unexpectedly low (18.2%) with GV841 (Table 3). Likewise, VIB572a have demonstrated high efficiency (nearly 100%) in detection of three environmental Vibrio isolates (V. cyclitrophicus, V. kanaloae and V. tubiashii) present in suspension F, whereas the result obtained with GV841 was entirely negative (Table 3, Figure 2).  Finally, the experiments carried out with suspension G containing a eukaryotic microorganism, Saccharomyces cerevisiae ATCC 9763, showed that both Vibrio-specific probes could not hybridize with eukaryotic nucleic acids. In all cases where hybridization was detected, the green fluorescence emitted by the probe fluorophore (i.e. ALEXA Fluor 488) was more intense using VIB572a than that obtained with GV841.
Besides testing the sensitivity and selectivity of CARD-FISH probes, this method was also used along with other techniques for detection and enumeration of microorganisms and Vibrio spp. in seawater samples collected in 3 different locations (coastal, open sea or estuarine areas) within the Biscay Bay at 11 to 14ºC according to AZTI Tecnalia. In these experiments, the cultivation on Marine Agar and TCBS agar were respectively used to determine the number of culturable chemoorganotrophic and putative Vibrio spp., whereas the total number of bacteria and Vibrio spp. were enumerated by epifluorescence microscopy using either acridine orange staining or CARD-FISH with VIB572a, respectively. The results of this analysis are summarized in Table 4. While the total number of bacteria in seawater samples was found to be between 1.05 10 5 and 4.92 10 6 cells ml -1 , the percentages of chemoorganotrophic bacteria were highly variable (<1-62.4%), with a number of chemoorganotrophic bacteria in the range of 7.5 10 1 to 3.25 10 5 CFU ml -1 depending on the place and time of sampling. The Vibrio counts determined by CARD-FISH were less than 1.3 10 3 Vibrio ml -1 , thus representing a very small fraction (less than 1%) of the entire bacterial population. The number of culturable putative Vibrio spp. ranged from 7 to 1.66 10 2 CFU ml -1 and its fraction was very dynamic (from less than 1% to nearly 81%) among the entire Vibrio spp. The higher percentages of culturable putative Vibrio spp. were obtained for estuarine water samples.
The vibrios density was low and not enough to be detected by CARD-FISH using GV841 probe. Occasionally, some cells were detected with this probe, but the green fluorescence emission was weak.

Discussion
In this study, CARD-FISH technique and probes cited in literature for Vibrio spp. detection and enumeration have been compared and tested. CARD-FISH technique corrects the problems attributed to the conventional FISH technique [23,42] that possesses insufficient sensitivity for detection of microorganisms in environmental samples. The latter often originate from nutrient-limited environments, in which microorganisms contain a low number of ribosomes and therefore have a limited number of sites for hybridization with the 16S rRNA-specific probes [21,43]. In present study, the protocol described by Pernthaler et al. [26] has been modified, increasing (up to twice) the amount of lysozyme used for the cell permeabilization [44]. This modification improves the accessibility of fluorochrome, and therefore provides a greater emission of fluorescence in turn, contributing to better detection of Vibrio spp.
Since the accuracy of detection depends on the specificity of the probe towards the target cells, we initially pre-selected three probes, GV841 [30], Vib-16S-1 [35] and VIB572a [31] and evaluated their specificity using in silico tools (SILVA and BLAST search). Although according to the results obtained with SILVA, the specificity of all 3 probes was similar with respect to Vibrio spp., further examination of Vib-16S-1 by BLAST showed that it has wider coverage and specificity and should efficiently hybridize with 16S rRNA of some non-Vibrio genera. Namely, unlike GV841 and VIB572a, Vib-16S-1 showed a complete coverage of and perfect complementarities to the target 16S rRNA regions present Oceanisphoera, Avibacterium and Catenovulum, and therefore, this probe was discarded due to its lower specificity for the Vibrio genus. Likewise, the BLAST search results obtained for the GV841 probe [30] indicated that it was capable of hybridizing not only with the target regions of Vibrio spp. but also with those present in other genera of the Vibrionaceae family (Candidatus Photodesmus, Aliivibrio and Catenococcus). An example of Candidatus Photodesmus species includes the symbiotic bacterium of Anomalops katoptron [45]. As to Catenococcus genus, a single species, C. thiocycli isolated from a hydrothermal marine sulfuric environment in New Guinea has been described [46]. Considering the habitats of Candidatus Photodesmus and Catenococcus species, these genera are likely absent in regular water samples and therefore should not interfere with detection of Vibrio spp. using GV841. In contrast, the genera Aliivibrio and Vibrio share the same habitats and their 16S rRNAs have a high sequence identity (ca. 95.5%), which makes it difficult to distinguish these genera with GV841 [47]. Moreover, further analysis of GV841 revealed that it is not capable of hybridizing with all the species of the Vibrio genus. Apparently, the coverage of the target region was incomplete [31] and not specific enough to detect all Vibrio species that may be present in seawater [38]. Indeed, Oberbeckmann et al. [34] previously pointed out that this probe failed to detect V. cholerae and V. mimicus [48].
Concerning the VIB572a probe [31], in silico analysis revealed that this probe should provide a comprehensive coverage and high specificity in detection of different Vibrio strains, including V. cholerae and V. mimicus. Although, Huggett et al. [31] also suggested its ability to hybridize with 16S rRNA of Photobacterium and Listonella; our in silico data ruled out this scenario. In silico analysis has also revealed that, similar to GV841, VIB572a can potentially hybridize with 16S rRNA of other genera such as Photodesmus, Aliivibrio or Catenococcus.
Although in silico results suggest that both GV841 and VIB572a probes are potentially suitable for detection of nearly all Vibrio spp., the results obtained with different bacterial suspensions showed that, in contrast to VIB572a, GV841 was not capable of hybridizing with all the species of the Vibrio genus tested in this study. This different hybridization capacity could be attributed to an imperfect binding of the GV841 probe to the target site. Indeed, previous studies demonstrated that, some areas in 16S rRNA are less accessible than others; and their poor accessibility can strongly impede hybridization [49,50]. According to the existing classification, class I and class VI regions are the most (81-100%) and least (0-5%) accessible ones, respectively. Thus, GV841 apparently hybridizes to the class IV and V regions with an approximate accessibility of 6-40%, and therefore this probe is less suitable for the detection of Vibrio spp. In contrast, the hybridization sites of VIB572a [31] are within class I and class II areas with an accessibility of 61-100%, which makes this probe nearly ideal for CARD-FISH detection of Vibrio spp. Similar results have been obtained in studies dealing with the accessibility of hybridization sites within 16S rRNA [51] and 23S rRNA in E. coli [52], thus indicating that they can be extrapolated to other microorganisms.
Further analysis of seawater samples led to several observations. In agreement with the results obtained in other studies [14,41,53] including those carried within the Bay of Biscay [54][55][56], we found that the number of total bacteria was significantly higher than that of culturable chemoorganotrophic bacteria. Although the total number of vibrios enumerated with CARD-FISH was dependent on the probe used, the VIB572a probe provided the most consistent results, provided that the percentage of vibrios among the total bacteria was similar to that obtained for their culturable fractions (i.e. vibrios vs chemoorganotrophic bacteria). Moreover, the CARD-FISH protocol we tested here is robust yielding no difference in the intensity of the fluorescence signals emitted by vibrios from oligotrophic environments and by Vibrio spp. grown under laboratory conditions. Therefore, our findings demonstrate that the CARD-FISH technique that employs the VIB572a probe [31] could offer a nice tool to detect Vibrio spp. in seawater samples, even when vibrios are present at low concentrations.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Table S1. Oligonucleotide probes targeting 16S rRNA from bacteria of the Vibrio genus (collected in the scientific literature of the last 20 years). They are shown for each of the probes: the sequence, the population with which they hybridize and references, Table S2. Coverage and specificity, expressed as percentages, for the probes GV841, Vib-16S-1 and VIB572a (SILVA database). Those taxa whose values exceed 75% for both characteristics are shown, Table S3. Coverage (QC, query cover, %), identity (Ident, %), matches (Match) and E value (being the probability of finding the sequence -among all organisms -within this database), for the probes GV841, Vib-16S-1 and VIB572a (BLAST database). Those taxa whose values exceed 75% in coverage and identity for any of the three probes analyzed are shown.