Monitoring of uncultured Dunaliella sp . in an Egyptian solar saltern field based on RuBisCO-encoding gene cbbL

Culture-independent molecular approach was used to explore and evaluate the diversity of Dunaliella species living at the salt field Malahat El-Max Alexandria, Egypt. Bulk genomic DNA was extracted directly from the collected salt water samples. Specific PCR primers and methodology were designed to amplify the gene cbbL, which encodes the large subunit of the enzyme ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO, EC 4.1.1.39) of only Dunaliella species, from the extracted microbial metagenome. The 700 bp-PCR amplicons were cloned and cbbL clone library was constructed and analyzed by sequencing. Rarefaction curve was saturated at sequence analyses of 23 clones, obtaining 19 phylotypes of Dunaliella cbbL, representing the total composition of Dunaliella in the collected sample. All recorded phylotypes had the known deduced amino acid cbbL motive sequence and catalytic sites. Fingerprint sequence, characterizing Dunaliella cbbL, was recorded. The cbbL phylotypes were grouped into two distinct phylogenetic clusters. The cluster 1, consisting of 18 current cbbL phylotypes was rooted with a cluster containing cbbLs of Dunaliella salina, Dunaliella bioculata, Dunaliella primolecta and Dunaliella tertiolecta. The single phylotype, uncultured Dunaliella ElMax.3, forming cluster 2, showed a unique phylogenetic lineage in the evolution of Dunaliella cbbL. This study introduced the first functional gene markers for exploring Dunaliella species in salt waters without culture.


INTRODUCTION
The genus Dunaliella, belonging to Chlorophyceae, has been represented by 27 known species, of which 23 are from salt water.Species of Dunaliella occur in freshwater, euryhaline habitats of all continents, oceans including the Dead Sea and even the salt lakes of the Antarctic (Tafreshi and Shariati, 2009).Thus, Dunaliella is a global genus of algae, thriving in habitats with wide ranges of salinity, pH, light intensity and temperature.The characterization of economically important algae in solar salterns has become a key focus for many research groups all over the world (Chen and Jiang, 2009).Dunaliella species that are living in these fields, have an economic importance due to its highly production of antioxidants (Oren, 2010).In Egypt, there are several coastal solar salterns, where Dunaliella is living, mainly on and near to the Mediterranean coast.Studies on Dunaliella in Egyptian aquatic habitats have been focused on characterization, environmental stresses and biotechnological applications of isolated species, including Dunaliella bardawil (Adam, 1997), Dunaliella salina (Abd El-Baky E-mail: hosameasa@yahoo.com.Tel: 0020506831378.Fax: 0020506831378.et al., 2004;Dardir and Abd El-Tawab, 2005;Shafik, 2008), Dunaliella parva (Said, 2009) and Dunaliella tertiolecta (Tammam et al., 2011).Molecular phylogeny techniques have been applied to the taxonomic study of Dunaliella from 1999 onwards.These studies have encompassed the 18S rRNA genes and the internal transcribed spacer regions, ITS and have been based on gene sequence comparisons as well as on restriction fragment length polymorphism studies (Gomez and Gonzalez, 2001;Assunção et al., 2012).Little correlation was found between the molecular data and the traditional morphological-physiological attributes to delineate species within the genus Dunaliella (González et al., 1999(González et al., , 2001)).
On the basis of 18S rRNA gene sequencing, Olmos et al. (2000) could differentiate between D. salina, D. parva and D. bardawil as species containing one, two and three introns, respectively, within the 18S rRNA gene.Also, Hejazi et al. (2010) reported a new 18S rRNA gene arrangement in term of intron localization and nucleotide sequence in a Dunaliella isolated from Iranian salt lakes.However, molecular phylogenetic studies of Dunaliella in Egyptian solar salterns have not yet been done; remaining the diversity of Dunaliella species in these fields are poorly described.The gene, cbbL that encodes the large subunit of the enzyme RuBisCO, the key of Calvin cycle has been used as an efficient molecular tool for studying the functional diversity of several phytoplankton (Tabita, 1999;Yoon et al., 2001;Ghosh and Love, 2011).The RuBisCO large subunit is distributed in almost all phytoplankton, through five different forms: form IA, IB, IC, ID and II (Watson and Tabita, 1997;John et al., 2007).The halophylic Dunaliella species, the pink micro-algae, are differentiated from other species in the class Chlorophyceae by harboring the RuBisCO form ID (Lin and Carpenter, 1997).This study developed a molecular approach for first monitoring of the diversity of Dunaliella based on PCR amplification, cloning and sequencing of the RuBisCO gene cbbL in metagenomic DNA extracted from the Saltern field, Malahat El-Max, Alexandria, Egypt.Statistical analysis was applied to determine the total composition of Dunaliella cbbL phylotypes and the phylogenetic tree was constructed showing the phylogenetic distribution of the recovered Dunaliella cbbL phylotypes.

Sampling
Pink-color Dunaliella-containing surface waters, 5 L was collected from five stations, 1 L from each station, covering mostly the total area of the saltern field, Malahat El-Max, Alexandria, Egypt (Figure 1).Each liter of the collected water was filtered on a single cylindrical 0.2 µm filter membrane unit (type Sterivex-GS, Millipore Corp., USA).In order to remove the trapped pink carotenoid pigments, which may link with the extracted DNA and hinder PCR amplification, the filters were washed with 10 ml of 80% ethanol in sterile SET buffer (20% sucrose, 50 mM EDTA, 50 mM Tris-HCl, pH 7.6) several times.The inlet and outlet of the filters were capped, and the filters were stored at -30°C until processed.

RuBisCO gene cbbL
Bulk microbial DNA, metagenome, was extracted essentially within the Sterivex-GS filters housing according to the method of Somerville et al. (1989), with modifications.Cell lysis and proteolysis were carried out within the filter housing using 5M guanidine thiocyanate in SET buffer at 70°C for 20 min with shake.Crude, which contained high-molecular-weight nucleic acids, was then drawn off the filter and purified using high pure PCR template preparation kit, Roche, Catalog no.11796828001, Mannheim, Germany.The purified metagenomic DNAs, extracted from all samples, were combined and run on 0.9% agarose gel electrophoresis followed by staining with ethidium bromide and UV visualization.A new PCR primer set, Duna-537f (5ʹ-RTW AAA CCT AAA TTA GGT YTA TCW GGT-3ʹ) and Duna-1212r (5ʹ-AAT AGT ACC ACC ACC RAA TTG-3ʹ) were designed from multiple alignments of cbbL sequences of Dunaliella species, incorporated to date in the DNA databases, including D. salina strain CCAP 19/18 (acc.no.AY531529, GQ250046), D. acidophila (HQ142901), D. bioculata (AB127991), D. primolecta strain TS-3 (AB127992), D. tertiolecta (AY882012), D. bardawil strain TS-1 (AB127990) and D. parva strain UTEX 1983 (AJ001877).The primers corresponded to positions 511 to 537 (forward primer) and 1191 to 1212 (reverse primer) of the cbbL gene of Synechococcus sp.(Shinozaki et al., 1983).PCR reaction mixture (50 ml), contained 10x EX buffer II (Mg +2 plus), 0.2 M primer, 400 M dNTP each, 2.25U Takara EX-Taq Polymerase (Takara, Japan) and 5 to 30 ng DNA template.The PCR mixture composition was performed with an initial denaturation step of 3 min at 95°C.The reaction was continued with 30 cycles of 1 min at 95°C, 2 min at 46°C, and 3 min at 72°C, with a final extension of 10 min at 72°C.
PCR products had 3-A overhangs to facilitate TA-cloning into TOP10 Escherichia coli using a TOPO XL PCR-cloning kit according to the manufacturer's instructions (Catalog no.K4750-20, Invitrogen Life Technologies, Carlsbad, CA, USA).Only cells containing XL-TOPO vector with the insert were competent to grow with kanamycin.Colonies of these cells were screened directly by sequencing using vector primers T7 and an ABI 3730×l 96-capillary DNA analyzer (Applied Biosystems, Foster City, CA, USA).

Sequence and bioinformatics analyses of cbbL genes
Sequencing results were introduced to FASTA (http://fasta.ddbj.nig.ac.jp/top-e.html),to determine their similarity to known Dunaliella cbbL sequences, deposited in DNA database.Nucleotide sequences for cbbL open reading frames (ORFs) were submitted to Transeq, (http://www.ebi.ac.uk/emboss/transeq/), to obtain the inferred amino-acid sequences.The correct cbbL ORFs were identified from the presence of diagnostic cbbL CO 2 fixation motifs.Grouping of sequences into phylotypes was done based on genetic distances using the Mothur software package V.1.7.2 (Schloss et al., 2009), where cbbL sequences, having 100% deduced amino acid identities, were grouped into a single phylotype.The expected total composition of Dunaliella cbbL was determined by rarefaction analyses in the Mothur software package.The phylogenetic analysis was done based on deduced amino acid and nucleotide sequences, and included the current cbbL sequences and their Dunaliella cbbL homologues, with significant homologies, e-value < 0.001, from the DNA database bank.The process was performed by applying the neighbor-joining algorithm and drawing the trees using the MEGA 3.1 software (http://www.megasoftware.net/).The branching patterns of the constructed phylogenetic tree were confirmed by reconstruction of the phylogenies using two other methods of analysis, namely maximum parsimony and maximum-likelihood, contained within the Phylip package.
The sequences of recovered Dunaliella cbbL were deposited in DNA data base bank under accession numbers from AB781569 to AB781587.

Efficiency of primers for PCR amplification of Dunaliella cbbL
This work aimed to cover the molecular diversity of almost all Dunaliella species in the collected saline water sample.In order to achieve this target, several considerations were involved in the design of an efficient primer set for amplification of Dunaliella cbbL.The primers must be designed from cbbL region, which encodes CO 2 catalytic site of Dunaliella RuBisCO form ID. Hence, the codon region of the cbbL catalytic site amino acid group T(V) KPKLG (Newman and Gutteridge, 1993) represented specific sequence for the design of the primer Duna-537f.The primer Duna-1212r was designed from the codons that encode the catalytic amino acids QFGGGTI, characterizing the Dunaliella RuBisCO ID cbbL.These primer sequences were differed from other cbbL sequences, located in other autotrophic algae, in several nucleotides, sufficient to make each of the primers binds only with Dunaliella cbbL.This was clear in the positive PCR amplifications of Dunaliella cbbL, while negative amplifications were noticed for cbbL from other phytoplankton species (Figure 2).These primer sets provided cbbL product size, 700 bp, longer than that produced by Paul et al. (2000).This gave an extended view to screen the diversity of Dunaliella cbbL.
In order to reduce the possibility of bias in amplification of Dunaliella cbbL, the PCR was tested using a range from 26 to 30 cycles, and then the amplicons were combined for cloning (Suzuki and Giovannoni, 1996).This was clear in the rarefaction curves, for the recovered Dunaliella cbbL, which showed saturation after analysis of 23 clones, yielding 19 phylotypes (Supplementary   Figure 1).This result may reflect the efficiency of the designed primers and PCR methodology in covering the total composition of Dunaliella in the studied water sample.

The recovered cbbL phylotypes were belonged to uncultured Dunaliella species
All recovered Dunaliella cbbL phylotypes were from metagenomic DNAs and, consequently, the source Dunaliella species were uncultured and had no common synonyms.Hence, nomenclatures were given to these phylotypes, whereby each phylotype was the descriptor of uncultured Dunaliella followed by clone name, which included sampling site name and clone number, for example, uncultured Dunaliella ElMax.1 and so on.cbbL phylotypes have been recovered from uncultured autotrophic microbial communities, which have no obvious species identifications (Guo et al., 2013).

The recovered phylotypes had the known cbbL deduced amino acid catalytic sites, beside specific sequence characterizing Dunaliella cbbL
Alignments of 702 bp-cbbL phylotypes, which yielded 234 deduced amino acids, were done with those recovered from known Dunaliella species (Figure 3).The positions of catalytic sites were corresponded to the cbbL of Synechococcus sp.(Newman and Gutteridge, 1993).All the current Dunaliella cbbL phylotypes possessed the characteristic RuBisCO motif sequence, GXDFXKXDE.Each of the amino acids GDFKE in the motif sequence was represented by two codons, for example, the amino acid G was represented by the codon "GGT" in the phylotype uncultured Dunaliella ElMax.3 and the codon "GGA" in other recovered phylotypes (Figure 3).The recovered cbbL phylotypes contained the known cbbL catalytic deduced amino acid sequences.The nucleotide positions from 714 to 741, of cbbL of Synechococcus sp. were found to contain the finger print sequence, 5- GCTGAAGGAATGCTTCAACGTGCACAA-3, of the Dunaliella cbbL (Figure 3).Consequently, the amino acids AGLQQ (Figure 3), encoded by this finger print nucleotide sequence, were considered as Dunaliella RuBisCO key amino acids, which differentiate it from those of in other phytoplankton.cbbL finger print sequences have been recorded for several autotrophic bacterial endosymbionts and used as nucleotide probes for in situ localization of the corresponding cbbL gene in the host tissues (Elsaied et al., 2002(Elsaied et al., , 2006;;Kimura et al., 2003).
On the other hand, red-like RuBisCO cbbL marker sequence has been used as a TaqMan probe for detection and quantification of soil autotrophic bacteria (Selesi et al., 2007).Hence, this current Dunaliella cbbL finger print sequence may be used as a biomarker for detection of uncultured Dunaliella by probing and hybridization techniques.

The recovered Dunaliella cbbL phylotypes were phylogenetically unique
Phylogenetic tree of cbbL was constructed in order to demonstrate the classification of the recovered Dunaliella cbbL (Figure 4).The nineteen cbbL phylotypes were distributed in two unique phylogenetic clusters.The cluster 1 harbored 18 cbbL phylotypes, which showed an average value of 95% ND (nucleotide identity), that is, 97% aa (amino acid identity) with each others.On the other hand, the cbbL phylotypes, constituting cluster 1, were distinct from other known Dunaliella species located in the same Dunaliella phylogenetic branch by showing homologies of 89 to 90% ND (92 to 93% aa ) (Figure 4).The uncultured Dunaliella ElMax.3, forming the cluster 2, had a unique cbbL phylogenetic lineage.It showed average identity of 87% ND (89% aa ) with other current Dunaliella

Figure 1 .
Figure 1.A map showing the sites of sampling marked by black circles.

Figure 3 .
Figure 3. Multiple alignments, within the amplification size, 700 bp, of recorded uncultured Dunaliella ElMax.1 to 19, and cbbLs from other known phytoplankton recorded in database.nucleotide alignments are only in the sites of RuBisCO cbbL motive sequence, marked by shading, and catalytic amino acids, bold face.Alignments in the box represent the finger print sequence of Dunaliella cbbL.Species used in the alignments had accession numbers listed in Figure 4.

Figure 4 .
Figure 4.A tree shows the phylogenetic relationship between the current uncultured Dunaliella clones, ElMax.1 to 19, marked with bold face, and other cbbL sequences from Dunaliella species available in database, beside cbbL sequences from other phytoplankton, as out groups.The consensus tree was constructed by neighbour-joining analysis and confirmed by maximum-parsimony and maximum-likelihood methods.Bootstrap values of >50% are indicated at the nodes.The bar represents 0.1 changes per amino acid or nucleotide.