Complete characterization of new isolates of Neptunomonas phycophila leads to emend its description and opens possibilities of biotechnological applications

Abstract Five strains were isolated from gonad of Great scallop (Pecten maximus) broodstock in a Norwegian hatchery. The study of 16S rRNA gene sequences showed that these isolates belong to Neptunomonas phycophila, a bacterium originally isolated from a symbiont of the anemone Aiptasia tagetes from Puerto Rico. The gyrB and rpoB genes sequences confirmed the affiliation of the scallop isolates to this species. Phenotypic characterization was performed and some differences between the Norwegian isolates and the type strain of N. phycophila were detected, such as ranges of temperature, pH, and tolerance to salinity or the use of several substrates as sole carbon source which lead to an emended description of the species. The strain 3CM2.5 showed phosphatidylethanolamine and phosphatidylglycerol as the major polar lipids. The whole genomes of the scallop strain 3CM2.5 and type strain of the species CECT 8716T were obtained and the annotation of these genomes revealed the presence of genes involved in degradation of aromatic compounds in both strains. Results obtained not only widen the geographical and host ranges of N. phycophila, but also point out possible biotechnological applications for this bacterial species.

from broodstock gonad. These isolates, 3CM2.5, 3SH2.1, 3SM2.1, 2CH2.2, and 3CH2.4, were identified as N. phycophila by sequencing the 16S rRNA gene. In this study, phenotypic, genetic, and chemotaxonomic analyses were performed to fully characterize the Norwegian strains of N. phycophila which lead to an emended description of this species.

| Isolation and conservation of strains
All samples were obtained in January 2012 from broodstock gonad before and after spawning. Shellfish were opened aseptically by cutting the adductor muscle with a sterile scalpel. One gram of the gonad tissue was homogenized in 1 ml of artificial seawater (ASW). The mixture was serially diluted and 100 μl of each dilution were inoculated

| 16S rRNA and housekeeping genes sequencing
Genomic DNA of pure cultures was obtained using the 'Instagene' matrix (Bio-Rad), following the manufacturer's recommendations.
Amplification of 16S rRNA gene of the different strains was carried out using primers 27F (5′-AGAGTTTGATCCTGGCTCAG) and 1510R (5′-GGTTACCTTGTTACGACTT) as previously described by Lane (1991). For the reference strains, sequences of 16S rRNA gene were retrieved from GenBank/EMBL. Sequence data analysis was performed with the DNASTAR Lasergene SEQMAN program. Sequences of the isolates were subjected to a BLAST search against EzTaxon-extended database (Kim et al., 2012).
The phylogenetic trees based on 16S rRNA gene sequences and concatenated sequences of 16S rRNA and housekeeping genes were constructed with neighbor-joining (NJ) and maximum-likelihood (ML) algorithms, using the program MEGA version 6.0. (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013). Distance matrices were calculated using Kimura's two-parameter correction and stability of the groupings was estimated by bootstrap analysis (1000 pseudoreplicates).
All media were supplemented with 1% NaCl when required. Catalase activity was determined based on bubble production in H 2 O 2 . The oxidase activity was determined by oxidation of 1% (w/v) N,N,N′,N′tetramethylρ-phenylenediamine. Gelatinase and lipase activities and hydrolysis of aesculin and starch were analyzed on MA plates supplemented with 0.4%, 1%, 0.1%, and 0.4% (w/v) of substrates, respectively. Use of different compounds as a sole carbon source was tested on basal medium agar (BMA) supplemented with different substrates according to Baumann and Baumann (1981). A total of 48 enriched media, positive control (BMA and 5 g yeast extract) and negative control (BMA) were inoculated with 10 μl of each isolate suspension via a Denley Multipoint-Inoculator (Denley Ltd) and incubated at 23°C for 14 days. Enriched media were examined daily using positive and negative control plates. Growth at various temperatures (4-44°C) was determined on MA plates. The optimum pH range for growth was examined in MB using appropriate biological buffers for adjusting the pH to 3, 5, 7, 9, 10, and 11. Tolerance to different saline concentrations was tested on basal medium (4 g/L neopeptone and 1 g/L yeast extract solidified with 1.5% (w/v) American bacteriological agar [Pronadisa]) supplemented with NaCl in a range from 0% to 10% and in sea salts solution (SIGMA) at concentrations between 0% and 15%, measuring cellular growth by turbidity (OD 600 ). Additional biochemical tests for Norwegian strains were carried out using the miniaturized system API ZYM (bioMerieux) according to the manufacturer's instructions but using a medium containing artificial sea salts suspension (3.5% w/v). All these test and system were incubated at room temperature for 7 days.

| Chemotaxonomy
Whole-cell fatty acid analysis was concurrently performed on cells of five scallop isolates, N. naphthovorans CECT 7132 T and N. phycophila CECT 8716 T grown for 3 days at 25°C on MA (Difco). Extraction and analysis of whole-cell fatty acids was carried out using the MIDI system in accordance with the protocols and instrument specifications recommended by the manufacturer (Sasser, 1990) using phospholipid fatty acids (PFLAD1) and environmental aerobes (TSBA) databases. Extraction and analysis of respiratory quinones and polar lipids of the strain 3CM2.5 were carried out by the identification service of the deutsche dammlung von mikroorganismen und zellkulturen (DSMZ) using thin-layer chromatography following the methodology described by Tindall (1990a,b).

| Genome sequencing and annotation, DNA-DNA relatedness, average nucleotide identity and G+C content
In addition, DNA of isolates 3CM2.5 and N. phycophila CECT 8716 T were extracted with a High Pure PCR Template Preparation Kit (Roche) and sequenced at FISABIO Sequencing and Bioinformatics Service (Valencia, Spain) using Illumina MiSeq technology. Genome assembly was performed using SPAdes 3.6.2 (Nurk et al., 2013) and QUAST (Gurevich, Saveliev, Vyahhi, & Tesler, 2013). These whole-genomes shotgun projects have been deposited at DDBJ/ ENA/GenBank under the accession numbers of MRCJ00000000 (3CM2.5) and MRCI00000000 (N. phycophila CECT 8716 T ). The versions described in this paper are versions MRCJ01000000 and MRCI01000000, respectively.
A sequenced-based comparison analysis was performed using the RAST annotation server and Blast Ring Image Generator (BRIG; Alikhan, Petty, Ben Zakour, & Beatson, 2011) was used to obtain a genomic map showing similarity percentages. The upper and lower identity thresholds were set at 90% and 70%, respectively.

| Polyphasic characterization of the isolates
According to the analysis of 16S rRNA gene sequences, the five isolates from great scallop showed a high similarity of 100% with the type Phenotypically, Norwegian isolates exhibited convexed and glistening colonies with a white coloration and were approximately 0.8-1.5 mm in diameter and circular with opaque entire margins. These strains displayed homologous biochemical and morphological characteristics; being rod-shaped, Gram-negative, oxidase and catalase positive, and exhibiting oxidative carbohydrate metabolism. All isolates, with the exception of 3SM2.1 and 3CM2.5, were nonmotile, unlike N. phycophila-type strain CECT 8716 T which is defined as a motile bacterium. Additionally, these isolates tested negative for gelatin, starch and aesculin hydrolysis, lipase and urease activity, indol and nitrate reduction. The range of growth at different temperature, NaCl concentration or pH observed for the scallop strains differed from those given for N. phycophila CECT 8716 T . Temperature of growth for Norwegian isolates ranged between 4 and 37°C, with an optimum of 23°C. In addition, our isolates showed variable results of tolerance to saline concentrations. Unlike type strain of N. phycophila, scallop isolates were able to grow between 3% and 8% of NaCl except 2CH2.2 and 3CH2.4, which only grew to 6% of NaCl. Also, no isolate was able to grow with only 0.5% of NaCl. The pH growth range was 5-10 with the exception of 3SH2.3, which presented a pH range of 5-8, but contrarily to CECT 8716 T they don't grow at pH 11. Glucose was not fermented by scallop isolates, unlike N. phycophila CECT 8716 T which was able to weakly ferment several carbohydrates (Table 1).
Variable results were observed in the utilization of trans-aconitic acid, tyrosine, citruline and putrescine. Type strain of N. phycophila CECT 8716 T showed different ability in the use of carbon sources being able to utilize d-ribose, arabinose, d-galactose, trehalose, d-mannose, cellobiose, lactose, and l-histidine but not gluconic acid (Table 1).
As for the type strain CECT 8716 T , the principal respiratory quinone was Q8.

| Emended description of Neptunomonas phycophila Frommlet 2015
The description of the N. phycophila is based on that given previously The major polar lipids of species are phosphatidylethanolamine and phosphatidylglycerol in addition to unknown lipid, phospholipid, aminolipid, and glycolipid. Annotation of these genomes revealed the same subsystems in both isolates but with some important differences ( Use as Carbon source of
only in strain 3CM2.5. Another important difference among these genomes was in the presence of genes related to stress response. Isolate 3CM2.5 showed two genes associated to flavohemoglobin (fHb) and 40 sequences corresponding to bacterial hemoglobins which were absent in type strain CECT 8716 T . The coded proteins are involved in responses to nitric oxide (NO) and nitrosative stress.

| DISCUSSION
The genus Neptunomonas and its type species N. naphthovorans were described in 1999 to accommodate two strains isolated from creosote-contaminated PAH degradation abilities (Hedlund et al., 1999). Between 2008 and 2014 five new species were added to the genus, all of them associated to marine environments or aquatic organisms, such as ark clams. The geographical range of these species not only includes Asiatic countries like China, Japan, or Korea, but also the Antarctica. During these years, the genus was subjected to two emended descriptions as new species were being described Yang et al., 2014).
Neptunomonas phycophila was recently described on the basis of a unique bacterial strain isolated during a study of the microbiota associated with in vitro cultures of the dinoflagellate Symbiodinium sp. (Frommlet et al., 2015). To our knowledge, no further reference exists in the literature on this bacterial species. The isolates included in the present study, obtained from a Great scallop hatchery in Norway, were undoubtedly assigned to N. phycophila on the basis of their 16S rRNA gene sequence similarity, indicating that this bacterial species presents a wider host and geographical distribution, from the Caribbean to the North sea.
The scallop isolates and N. phycophila CECT 8716 T presented several differential characteristics as the temperature, pH, and salinity tolerance among others. The type strain of N. phycophila CECT 8716 T was described as able to grow at 40°C with an optimum of 30°C, whereas the five Norwegian isolates showed a lower range of temperature to growth, with an optimum of 23°C. One of the bacterial mechanisms of Values are percentages of the total fatty acids; fatty acids that make up <1% of the total are indicated by tr. For unsaturated fatty acids, the position of the double bond is located by counting from the methyl (ω) end of the carbon chain. cis-Isomer is indicated by the suffixes c. Summed features are groups of two fatty acids that cannot be separated by GLC with the MIDI system. Summed feature 3 contains C 16:1 ω7c/C 16:1 ω6c and Summed feature 8 contains C 18:1 ω7c and/or C 18:1 ω6c. All data obtained in the present study. ND, not detected. adaptation to changes in temperature is the modification of the fatty acid composition of their membranes. However, no differences were found between the two isolates in the analysis of the genes involved in cell membrane formation or in fatty acid biosynthesis. On the other hand, in this study, all isolates were grown under the same conditions of nutrients and temperature for the comparison of fatty acid composition. Thus, the differences in the profiles obtained were minimal, and could not be related to the variations in ranges of growth temperature among strains. The differences in the optimum growth temperatures exhibited by the strains might be due to an evolutionary adaptation of these bacteria to the distinct climate conditions of their habitat (Chattopadhyay, 2006) or, on the other hand, to the phenotypic plasticity of strains able to produce diverse phenotype characteristics in different environments (Sikorski, Brambilla, Kroppenstedt, & Tindall, 2008). Therefore, this variation should be in-depth analyzed to determine if it is due to a transient change or an adaptation.
All the differences observed in the temperature, pH, salinity, utilization of carbon sources or enzymatic profiles made necessary an F I G U R E 3 Graphical map of the BLASTN comparison of two strains of Neptunomonas phycophila genomes. From center to outside: GC content (ring 1), GC skew (ring 2), N. phycophila 3CM2.5 (ring 3), stress response genes (ring 4), phages genes (ring 5), CRISPR (ring 6), N. phycophila CECT 8716 T (ring 7). emended description of the species, which was originally performed on the basis of the analysis of a single strain.
As mentioned above, the type species of the genus Neptunomonas, N. naphthovorans, is related to the degradation of PAH. N. phycophila was not originally described as a degrading bacteria, and none of its representatives were obtained from contaminated waters, but our results demonstrated that genes involved in degradation of aromatic compounds are present in the genomes of its type strain CECT 8716 T as well as the scallop isolate 3CM2.5. Among others, genes implicated in catechol and chloroaromatic degradation pathways were detected in both genomes.
Another important difference between strains 3CM2.5 and CECT 8716 T was found in genes of the stress response in each of the strains.
The 3CM2.5 isolate harbor two fHb and 40 hemoglobin sequences that are absent in the CECT 8716 T strain. These fHb are involved in the inactivation of NO by NO dioxygenase activity. NO is a molecule that modulates numerous aspects of animal physiology, including defense mechanisms in response to attack by microorganisms for which NO is toxic. However, in pathogenic bacteria the synthesis of fHbs that inactivate NO counteracts the defense mechanisms of infected cells.
In addition, sequences from bacteriophages and CRISPRs were detected only in strain 3CM2.5. Phages are the most abundant organisms in the biosphere and are widely represented in marine environments.
The presence of these sequences in the 3CM2.5 strain might provide a kind of immune system against new infections (Koonin & Wolf, 2015).
Also, the presence of several phage sequences can lead to increased genetic diversity and adaptation to different environments of bacterial populations (Rodríguez-Valera et al., 2009), as well as conferring to the host certain advantages such as expression of resistance genes to antibiotics or even infective capacity (Davies, Winstanley, Fothergill, & James, 2016;Wagner & Waldor, 2002).
In conclusion, the results obtained in this study not only demonstrate an increased range of hosts and geographic distribution for N. phycophila, but also lead to update the description of this species.
Our study also demonstrated that strains belonging to N. phycophila, as other representatives of the genus Neptunomonas, posses the genes necessary to degrade PAH compounds, opening an important field of research to properly characterize the degradative pathways in this bacterial species and to determine its potential biotechnological applications.