Assessment of 16S rRNA Gene-Based Phylogenetic Diversity of Archaeal Communities in Halite-Crystal Salts Processed from Natural Saharan Saline Systems of Southern Tunisia

Simple Summary In the current investigation, we determine the archaeal diversity from halite-crystal salts sampled from Chott Djerid and Chott Douz, two saline Chotts in the Tunisian Sahara, using both culture-dependent and independent approaches citing DGGE (V3 regions), 16S rRNA-gene (full length gene ~ 1500 bp) based clone libraries and high-throughput sequencing technology of Illumina MiSeq platform (V3–V4 regions). The amalgamation of all results yielded a comprehensive view of the archaeal diversity represented by members of Halobacteria and Nanohaloarchaea classes. Abstract A thorough assessment of the phylogenetic diversity and community structure of halophilic archaea from three halite-crystal salts, processed from two separated saline systems of Southern Tunisia has been performed using culture dependent and independent methods targeting different regions of 16S rRNA gene sequences including DGGE, 16S rRNA clone libraries and Illumina Miseq sequencing. Two samples, CDR (red halite-crystal salts) and CDW (white halite-crystal salts), were collected from Chott-Eljerid and one sample CDZ (white halite-crystal salts) from Chott Douz. Fourteen isolates were identified as Halorubrum, Haloferax, Haloarcula, and Halogeometricum genera members. Culture-independent approach revealed a high diversity of archaeal members present in all samples, represented by the Euryarchaeal phylum and the dominance of the Halobacteria class. Nanohaloarchaea were also identified only in white halite samples based on metagenomic analysis. In fact, a total of 61 genera were identified with members of the Halorhabdus, Halonotius, Halorubrum, Haloarcula, and unclassified. Halobacteriaceae were shared among all samples. Unexpected diversity profiles between samples was observed where the red halite crust sample was considered as the most diverse one. The highest diversity was observed with Miseq approach, nevertheless, some genera were detected only with 16S rRNA clone libraries and cultured approaches.

Haloarchaea are known to promote the crystal halite formation [25]. In more detail, when NaCl begins to precipitate, Haloarchaea are embedded in situ within fluid inclusions and remain viable for thousands to hundreds of millions of years [26][27][28][29][30]. Therefore, Haloarchaea are considered as attractive models for astrobiologists due to their polyextremophilic nature with tolerance of saturating salinity, anaerobic conditions, high levels of ultraviolet and ionizing radiation, subzero temperatures, desiccation, and toxic ions [16,31,32].
During the last years research has been focused on the study of prokaryotic diversity and the structure of communities in different saline ecosystems from different geographical locations, based on the analyses of PCR-amplified 16S rRNA gene sequencing from environmental samples and using techniques such as DGGE (Denaturing Gradient Gel Electrophoresis), 16S rRNA libraries, SSCP (Single Strand Confirmation Polymorphism), Phy-loChip, and 16S rRNA amplicon sequencing [7,20,21,[41][42][43]. These approaches have been proved to be powerful tools to assess the microbial diversity since they enable the characterization of novel bacterial and archaeal lineages in such environments [6,7,9,21]. Previous studies on the microbiology of saline environments, indicated that Haloarchaea constitute the main dominant microbial group but most of them remain uncultured [21,44,45]. A differentiation of Haloarchaea within the different ecosystem studies was also observed [42,[45][46][47][48].
Southern Tunisia includes numerous ephemeral inland and costal Saharan salt lakes, known as Sebkhas and Chotts, lying south of the Atlas Mountains at the northern edge of the Sahara [1,2]. These salt ecosystems have been the subject of microbiological studies carried on saline water, sediments, and soils [3][4][5][6]. However, to the best of our knowledge, no studies have unraveled the phylogenetic diversity of archaeal communities of halitecrystal salts from Tunisian ecosystems based on culture dependent and culture independent approaches simultaneously.
The purpose of this study was to assess the archaeal diversity from halite-crystal salts sampled from Chott Djerid and Chott Douz; two saline Chotts in the Tunisian Sahara, using culture dependent and molecular approaches such as DGGE, 16S rRNA-gene based clone libraries and high-throughput sequencing technology of Illumina MiSeq platform. The present study will provide insights of the archaeal diversity from the Tunisian saline ecosystems and may be the first step towards the biotechnological exploitation of the Tunisian archaeal diversity.

Site Description and Sampling
Samples studied in the present report were collected in February 2010 from arid saline ecosystems located in three different locations including: (i) Chott Djerid (also known Chott El Djerid), situated in South-West of Tunisia, and it is considered as the largest saline lake (5000 km 2 ) of the Sahara with a surface area of over 5000 km 2 . Chott Djerid is a seasonal lake, that is completely dry for most of the year. It represents an important source of NaCl production for Tunisia. Two representative halite-crystal salts (white and red pigmentation), were collected from two different sites named as BDV18.2 (CDR) (GPS: N 33 • 55 10" E 8 • 30 8" Al 16 m) and BDV19.2 (CDW) (N 33 • 55 55" E 8 • 28 19" Al 9 m) respectively ( Figure 1).
(ii) Chott Douz, located in the South of Tunisia at 125 km South-East of Chott Djerid (GPS N 33 • 26,845 E 9 • 1356 Al 64 m). White halite-crystal salts sample indicated as CDZ was used in this study ( Figure 1).
Generally, the pH of the two Chotts was quite close to the neutrality (~7.4). The surrounding environment was characterized by the extremely low relative humidity. The temperature was ranging from 4 • C in the night and 38 to 40 • C during the day. All samples were collected under aseptic conditions into 50-mL sterile falcon tubes and were kept at 4 • C until further processing.
based clone libraries and high-throughput sequencing technology of Illumina MiSeq platform. The present study will provide insights of the archaeal diversity from the Tunisian saline ecosystems and may be the first step towards the biotechnological exploitation of the Tunisian archaeal diversity.

Site Description and Sampling
Samples studied in the present report were collected in February 2010 from arid saline ecosystems located in three different locations including: (i) Chott Djerid (also known Chott El Djerid), situated in South-West of Tunisia, and it is considered as the largest saline lake (5000 km 2 ) of the Sahara with a surface area of over 5000 km 2 . Chott Djerid is a seasonal lake, that is completely dry for most of the year. It represents an important source of NaCl production for Tunisia. Two representative halite-crystal salts (white and red pigmentation), were collected from two different sites named as BDV18.2 (CDR) (GPS: N 33°55′10″ E 8°30′8″ Al 16 m) and BDV19.2 (CDW) (N 33°55′55″ E 8°28′19″ Al 9 m) respectively ( Figure 1).
Generally, the pH of the two Chotts was quite close to the neutrality (~7.4). The surrounding environment was characterized by the extremely low relative humidity. The temperature was ranging from 4 °C in the night and 38 to 40 °C during the day. All samples were collected under aseptic conditions into 50-mL sterile falcon tubes and were kept at 4 °C until further processing.

DNA Extraction from Environmental Samples and Archaeal Isolates
Total DNA from salt samples was extracted using the FastDNA spin kit for Soil (MoBio, Carlsbad, CA, USA) according to the manufacture procedure. Quantification of nucleic acid samples was performed with a Thermo Scientific NanoDrop ™ ND-2000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. Genomic DNA from isolates was extracted from log-phase cells, lysed in distilled water by phenol-chloroform method and subsequently precipitated with ethanol, as previously described [49].
Molecular identification was based on 16S rRNA gene amplification and sequencing using universal archaeal primers 4F (5 -TCCGGTTGATCCTGCRG-3 ) and 1492R (5 -GGTTACCTTGTTACGACTT3-) [50]. PCR reactions were performed in 50 µL reaction mixtures containing the following: PCR buffer (1X), MgCl 2 (1.5 mM), 0.25 mM of each dNTP, 0.5 µM of each primer, 1 µg of chromosomal DNA, and 1 U of Taq DNA polymerase (Fementas). The program used was: 95 • C for 5 min; 35 cycles of 94 • C 45 s, 64 • C 45 s, 72 • C 1 min, and a final extension step at 72 • C for 10 min. The presence of specific PCR product was visualized after electrophoresis on 1.5% (w/v) agarose gels under UV light. PCR products were then purified using the polyethylene glycol (PEG) precipitation protocol [51]. Sequencing was performed in both strands on an ABI3130 analyzer according to the manufacturer's instructions (Applied Biosystems, Waltham, MA, USA).

DGGE Fingerprinting
To assess the total archaeal community structures among the salt samples, DGGE fingerprinting approach was applied. A nested PCR technique was employed to amplify the variable V3 region of 16S rRNA gene sequences from archaea as previously described by Cindy et al. (2001) with slight modifications. In the first round of PCR two primers, PRA46F (5 -YTAAGCCATGCRAGT-3 ) and PREA1100R (5 -YGGGTCTCGCTCGTTRCC-3 ), were used in order to amplify the archaeal community [52]. PCR reactions were carried out in a final volume of 20 µL contained buffer 1 X (Fermentas), MgCl 2 1.5 mM, 0.25 mM of each dNTP, 0.25 µM of each primer, 1-5 ng of total DNA, and 1 U of Taq DNA polymerase (Fementas). The program used was: 95 • C for 5 min; 35 cycles of 94 • C 45 s, 43 • C 45 s, 72 • C 1 min, and a final extension step at 72 • C for 10 min. Five to 10 µl of the previously PCR product were used as a template for the second PCR under the same conditions using PARCH340F-GC (5 -CCCTAYGGGGYGCASCAG3 ) and PARCH 907R (5 -GWATTACCGCGGCKGCTG-3 ) as primers [53]. A 40 bp of GC clamp (CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGG) was added to primer PARCH340F to increase separation of DNA bands in DGGE analysis [54].
Profiles of the amplified 16S rRNA gene sequences were produced by DGGE as described by Muyzer et al. [54] with some modifications. One hundred ng of the PCR products were loaded onto 8% (m/v) polyacrylamide gel in 0.5X TAE buffer (20 mM Tris (pH 8.0), 10 mM acetic acid, 0.5 mM EDTA; Bio-Rad, Paris, France) with a gradient between 5 and 55% (100% denaturant contains 7 M urea and 40% (vol/vol) deionized formamide). The electrophoresis was carried out in 0.5X TAE buffer at 200 V for 5 h at a constant temperature of 60 • C. The DNA fragments were stained for 30 min in 0.5X TAE buffer with ethidium bromide and visualized under UV light. The gels were then washed in distilled water for 15 min. DNA bands from the polyacrylamide gels were excised and incubated in 80 µL of sterile distilled water overnight at 37 • C. Five to 10 µL portion of the eluted DNA was used as the template in a PCR re-amplification with the PARCH 340F (without GC clamp) and PARCH 907R. PCR products were verified on 2% (w/v) agarose gel, purified with QIAquick PCR Purification Kit (Qiagen, Hilden, Germany) and sequenced by PRIMM (Milano, Italy). Sequencing was performed in one sense with primer PARCH340F.

Full-Length 16S rRNA Archaeal Libraries
Amplification of nearly complete 16S rRNA gene sequences from total environmental DNA of salt samples was performed using universal primers for archaea: 4F (5 -TCCGGTTGATCCTGCRG-3 ) and 1492R (5 -GGTTACCTTGTTACGACTT3-') [50,55]. PCR reactions were carried out in a final volume of 50 µL containing: 1X buffer (Takara), 1. , and a final extension step at 72 • C for 10 min. PCR products verified on 1.2% (w/v) agarose gel electrophoresis and purified by PEG precipitation [56]. Purified DNA was cloned into pGEM-T, according to the manufacturer's protocol (Promega, Madison, WI, USA), and transformed into competent Escherichia coli DH5α cells. White colonies on ampicillin/X-gal plates were screened for inserts of the correct length by PCR with the pGEM-T compatible primers T7 and SP6. Inserts were sequenced using an ABI3130 analyzer according to the manufacturer's instructions (Applied Biosystems, Waltham, MA, USA).

Sequence Analysis and Phylogenetic Assessment of Archaeal 16SrRNA Gene Sequences of Isolates and Clone Libraries
The nearly complete 16S rRNA gene sequences were assembled using DNAstar software (DNASTAR Inc., Madison, WI, USA). Sequence similarity searches were performed using the "Seqmatch" (Ribosomal Database Project II (http://rdp.cme.msu.edu/seqmatch/ seqmatch_intro.jsp, accessed on 10 December 2020) against sequences from individual isolates (type strains) and environmental uncultured sequences. For each isolate and clone sequences the two nearest relatives were obtained for phylogenetic analyses. The one with the highest percentage of similarity was retained for homology.
For the 16S rRNA archaeal libraries clones were initially screened for possible chimeric structures using the program CHECK_CHIMERA of the Ribosomal Database Project (RDP-II) [57]. After sequencing, clustering analysis was carried out using the Mothur package at the 3% cutoff, and the nearest relative was assigned using the Ribosomal Database Project [57,58]. Species richness (alpha diversity) was evaluated based on the diversity indices values notably the Chao1 estimator, Shannon diversity index, and Good's coverage calculated with the Mothur software version 1.28.0 [58].
The alignment of sequences was carried out using the CLUSTALX 1.83 program. The distance matrix-based phylogenetic tree was created using the PAUP* 4b10 software package. The distances were calculated using the Jukes and Cantor method [59] and the Topology was inferred using the "neighbor-joining" method [60] based on a bootstrap analysis of 1000 trees. The maximum parsimony calculated phylogenetic tree using the PAUP phylogenetic package was also generated. Only sequences longer than 200 bp were used in tree construction.

Illumina MiSeq Sequencing, Data Analysis, and Diversity Estimates
As an additional technique, the archaeal diversity was performed by next generation sequencing the 16S rRNA gene, using the Illumina MiSeq platform. Polymerase chain reaction (PCR) was performed with KAPA HiFi HotStart ReadyMix PCR Kit (KAPA All PCR products were migrated on a 1.5% (w/v) agarose gel in 1X Tris-Acetate-EDTA buffer (40 mM Tris-acetate, 1 mM EDTA). The amplified 16S rRNA products were purified with a 20% PEG, 2.5 M NaCl solution. The concentration of purified DNA was measured with a Quawell Q5000 micro-volume UV-vis spectrophotometer.
The resulting PCR amplicons were first diluted up to 10 ng/µL and then used for a second-step PCR to include the indexes (barcodes) as well as the Illumina adaptors. The combinatorial use of index primers resulted in unique samples that were pooled and sequenced on one Illumina MiSeq run. Amplification reaction was performed in a final volume of 50 µL contained 1X of KAPA HiFi Fidelity Buffer, 0.3 mM dNTPs, 1 µM of the forward and the reverse indexing primers, 0.02 U of KAPA HiFi HotStart DNA Polymerase, 2 µL from the diluted PCR product (10 ng/µL) and 25.5 µL of sterile deionized water. The program of PCR amplifications consists of 3 min incubation at 95 • C followed by 8 cycles of 95 • C for 30 s, 55 • C for 30 s and 72 • C for 30 s, and a final elongation step at 72 • C for 5 min. The resulting amplicons were purified using Macherey-Nagel's NucleoMag ® NGS Clean-up and Size Selection kit according to the manufacturer's recommendations, and all amplicons were quantified with a Quawell Q5000 micro-volume UV-vis spectrophotometer and merged in equimolar ratios (8 nM). The paired-end (2 × 300 bp) sequencing was done by Macrogen (Seoul, Korea) on the Illumina MiSeq platform.
Analysis of sequenced reads was done according to the standard pipeline of DADA2 (Divisive Amplicon Denoising Algorithm2) following the related standard operating procedure pipeline [61]. In brief, sequences that contained one or more ambiguous bases (N), sequences having a homopolymer stretch longer than 7 bases, and sequences shorter than 200 were considered of poor quality and trimmed from the data set. Chimeras were removed as well. Then, high-quality reads were aligned against the SILVA alignment database version silva_nr_v128 was applied. Filtered alignments were selected to generate a pairwise distance matrix, followed by binning the sequences into operational taxonomic units (OTUs). The mothur software package was applied to compute alpha diversity including richness, evenness, diversity and Good's coverage indices. Indeed, rarefaction curves-based ranking were calculated using resampling-without-replacement approach [58].

Nucleotide Sequence Accession Numbers
All 16S rRNA sequences identified in this study were deposited in GenBank with the following accession numbers: MW565874-MW565887 for isolates; MW564198-MW564204 for DGGE; MW564054-MW564064 for CDZ sample; MW563869-MW563906 for CDW; and MW565874-MW565920 for CDR sample.

Isolation and Identification of Haloarchaeal Isolates
Our culture-dependent approach led to the isolation of 44 archaeal isolates that were grouped into 14 categories based on their morphology and color. All groups were then subjected to phenotypical and molecular characterization ( Figure 1 and Table 1). Almost all isolates produced red, pink or orange pigmentation on agar plates which may be related with the presence of carotenoids in their cell membranes or with the C50 compound bacterioruberin and its derivatives. These pigments are known to protect haloarchaeal cells from extremophilic conditions such as osmotic stress, damage from high ultraviolet light exposure and anoxic or anaerobic conditions [62,63]. They contribute also in the salt crystallization process [63,64]. Bacterioruberin pigment content in some saline ecosystems has been used to survey the haloarchaeal community in such ecosystems [65]. White colonies were also isolated from CDW halite-crystal salt sample, similar to those were described in another study regarding the isolation of halophilic archaea from salt samples [66]. All observed colonies were circular, convex, mucoid, or not, with entire edges (Table 1). Their cells were pleomorphic or rod-shaped and stained Gram-negative as expected. All isolates were extremely halophilic and required between 10-30% of NaCl for growth with an optimal at 25% NaCl (w/v) ( Table 1). Isolates also showed a large range of growth temperature from 4 • C to 45 • C with an optimal growth at 40 • C; and pH ranging from 6 to 9 with the an optimum at 7.4 (Table 1). Hence, based on phenotypic characteristics, isolates were presumably assigned to the Halobacteriacaea family [10,11,13,33]. In order, to confirm this assignment, the 16S rRNA genes of all isolates were sequenced and were then compared to sequences deposited in the RDP database including cultured and uncultured strains ( Table 1). The use of both distance-matrix and character-based (parsimony) methods enabled the construction of related phylogenetic trees. All isolates were placed within the family Halobacteriaceae and assigned to the following taxonomically described species with a similarity of 99%: Halorubrum chaoviator (AM048786) (n = 4), Halorubrum xinjiangense (AY510707) (n = 1), Haloferax mediterranei (D11107) (n = 1), Haloarcula vallismortis (EF645688) (n = 2), Halogeometricum borinquense (AF002984) (n = 1), uncultured Haloarcula sp. (EF153424) (n = 1), uncultured haloarchaeon (FN391270; GQ374937) (n = 2), and uncultured Halorubrum sp. (FN391188) (n = 2) (Table 1 and Figure 2).  The 16S rRNA gene sequences similarity percentages and phylogenetic affiliations are shown in Table 1, and all of them shared 99% similarity with their closest phylogenetic relative. Our results are in accordance with those reported the diversity of isolated halophilic archaea from salt samples-such as rock salt, halite, gypsum, ancient halite, and salt brineswhere isolates were mainly affiliated to the class Halobacteria (phylum Euryarchaeota) and particularly within the genera Haloarcula and Halorubrum, while representatives from other genera are less common [30,[67][68][69][70][71][72][73]. These extremely halophilic archaea are known to remain viable inside halite and salt brine for months and years [27,30,74].

DGGE Fingerprint
As a preliminary step to get an overview in the differences in community profiles of archaeal population thriving in halite-crystal salts samples, DGGE analysis was carried out. A denaturing gradient 20-50% was performed ( Figure S1).
The resulting DGGE pattern highlighted different bands (n = 7) with diverse level of migration and intensity. Each band represents an archaeon-taxon and the presence of two similar migrating bands suggests that the same taxon was present in all samples. It is worth noting that almost all observed bands were encountered in the bottom of the gel reflecting their high % GC content, similar with the profile of Halobacteriaceae members, whose GC base ratios are exclusively high and range between 62-68% [75]. Close to these results is the survey of Archaea by Tamez and Lopes-Cortes [76]. CDR and CDW revealed a nearly similar DGGE fingerprint (three bands) and this similarity is rather expected due to the same sampling location (Chott Djerid). Moreover, CDZ revealed a completely different DGGE profile, characterized by the presence of two intense bands in the bottom of the gel, reflecting the low diversity of Archaea in that sample. Seven representative bands from different levels of migration were excised, re-amplified successfully, and finally sequenced. Data obtained indicated that archaeal communities in the three salt samples were affiliated to Euryarchaeota phylum (Table 2 and Figure 3). Most of the bands occurred in all samples were assigned to uncultured and unclassified_Halobacteriaceae (95.95-100% identity sequences) which is in accordance with previously published data on salt samples analyzed with DGGE technique [77]. Additionally, the other bands were assigned to cultivated Haloarchaea such as Natronomonas sp. with a 95% of identity observed only in CDZ, and Halorubrum sp identified in CDR with a 97.97% sequence similarity. Comparative analyses of the DGGE sequences and their closely relatives are illustrated in Figure 3. The neighbor-joining phylogenetic tree showed four main clusters corresponding to the phyla Euryarchaeota: three clusters, corresponding to order Halobacteriales, where the first could be assigned to Natrinema group encompassing the bands (2, 1, 4, and 5), the second represented by one band (5R) assigned to Halorubrum species and the rest assigned to unclassified uncultured Halobacteriacaea. It is worth noting that, limited research on the analysis of haloarchaeal community structure in halite crystals and salt samples, turned out that Halobacteriales members constituted the main group [53,78]. It is interesting to note that the distribution of archaeal members observed in our samples could be related to the adaptation of these microorganisms to some extreme physicochemical parameters such as salinity, pH, temperature, and oxygen [79][80][81].

Full Length 16S rRNA Gene Libraries
Archaeal community structure lying in different salt samples was further investigated by cloning and sequencing nearly the complete sequence of 16S rRNA genes from total DNA for the three samples (CDW, CDR and CDZ). All sequence reads (n = 90) were compared versus the reference database of known 16S rRNA genes obtained from RDP program. Operational taxonomic units were assigned based on a similarity distance threshold of 0.03 and their specific allocation was: 18, 12 and 2 OTUs in CDR, CDW, and CDZ respectively ( Table 3). The results showed that all clones were affiliated to Euryarchaeaota phylum with a diverse assemblage of classified and unclassified Halobacteriales members (Table 3).

Full Length 16S rRNA Gene Libraries
Archaeal community structure lying in different salt samples was further investigated by cloning and sequencing nearly the complete sequence of 16S rRNA genes from total DNA for the three samples (CDW, CDR and CDZ). All sequence reads (n = 90) were compared versus the reference database of known 16S rRNA genes obtained from RDP program. Operational taxonomic units were assigned based on a similarity distance threshold of 0.03 and their specific allocation was: 18, 12 and 2 OTUs in CDR, CDW, and CDZ respectively ( Table 3). The results showed that all clones were affiliated to Euryarchaeaota phylum with a diverse assemblage of classified and unclassified Halobacteriales members (Table 3).

Diversity and Richness Estimates
Alpha diversity of archaeal diversity was evaluated based on Shannon index, nonparametric richness (Chao) and Good's Coverage estimations (GC) ( Table 4). The results showed that based on the GC estimates the percentage of the total species were the order of 100%, 95%, and 76.31% for CDZ, CDW, and CDR respectively (Table 4), reflecting a sufficient sequencing depth for CDZ and CDR samples. In terms of archaeal richness, CDZ (25.2) had the highest chao value; followed by CDW (12.16) and CDZ (2.0) had the lowest values. Taxonomic Assignment, Quantification, and Distribution (i) In CDR sample 38 clones were obtained and were clustered in 18 OTUs. Almost all of them were phylogenetically assigned to uncultured Halobacteriales members (Table 3 and Figure 4). The clones were most closely related to 6 genera of extremely halophilic archaea including Halonotius-like sequences (OTU9, 10, and 11), Haloquadratum-like sequences (OTU1, 2, and 3), Halorubrum-like sequences (OTU5, 6, 7, and 8), Haloarcula-like sequences (OTU12 and 13), Halolamina (OTU4) and Halorubellus-like sequences (OTU14). The remaining OTUs (15 to 18) (15.78%) were phylogenetically unrelated to any previously cultivated taxa and are putative candidates for new genus and specieslevels members in Halobacteriaceae family ( Table 3). Members of the identified genera are consistent with that generally observed in many salt-saturated habitats like in several Tunisian hypersaline lakes [7,21,23], in Australian crystallizer ponds [82], in Santa Pola saltern in Spain [83] and in Keke Salt Lake [84]. Generally, such hypersaline habitats were known to hold high densities of extremely halophilic archaea characterized by a restricted number of genera-mainly Halotonuis, Halorubrum, and Haloarcula [9,85,86]. Haloarcula species are known to reside in the salt-saturated samples [98]. They are able to survive during the fluctuation of salinity and desiccation inside halite crystals. Actually, Haloarcula genus encompasses thirteen species [99]. In addition to the above observations, sequences affiliated to Halonotius, Haloquadratum, and Halorubrum genera dominated the clone libraries of CDR sample with a relative abundance of 26.3%, 21%, and 18.4% respectively. Ubiquity and abundance of Halonotuis genus members have been observed also in several hypersaline ecosystems such as saline lakes in Australia and China [84,87] and in solar salterns in Turkey and Spain [9,88]. Currently, two species are described, Halonotius pteroides, isolated from a crystallizer of an Australian saltern [43,89] and Halonotius aquaticus, isolated from a marine saltern in Spain [9]. On the other hand, the Haloquadratum genus, characterized by its peculiar square shape, was initially discovered in 1980 [90] and isolated in 2004 by Bolhuis and his colleague [91] and Bolhuis and his collaborators [92]. Similar to Halonotius, it was observed in many hypersaline habitats close to saturation such as salt crystals and saturated thalassic lakes [21,93,94]. Concerning the third observed genus, Halorubrum, proposed by McGenity [95], is considered as the largest Haloarchaeal genus with 42 valid species [40]. It is a well-known and omnipresent genus within the Halobacteria, whose members have been detected specially in salt and halite samples in several extremely saline environmentse.g., salterns, saline lakes, dead sea, salt-saturated crystallizer ponds, and ancient salt deposits [17,46,47,96]. Genetic diversity of, Halorubrum may due to high levels of both recombination and horizontal gene transfer [97].
The second subdominant group observed in CDR sample, was represented only by Haloracula genus members with a relative abundance of 10.52%. Similarly to Halorubrum, Haloarcula species are known to reside in the salt-saturated samples [98]. They are able to survive during the fluctuation of salinity and desiccation inside halite crystals. Actually, Haloarcula genus encompasses thirteen species [99].
The lowest abundant genera found were Halolamina and Halorubellus each one represented with 2.63% of the total clones. Generally, these two genera were most abundant in environments with fluctuating degrees of salinity [100,101]. The genus Halolamina contains only one described species isolated from an artificial marine saltern in China [100] Two species were described for Halorubellus genus, Halorubellus salinus, and Halorubellus litoreus [100,101].
(ii) In CDW sample, a total of 40 clones were obtained and phylogenetically clustered in 12 OTUs unevenly (Table 3 and Figure 5). All OTUs, were identified as uncultured Halobacteriales members (Table 3). Four genera, already observed in the CDR sample, were identified including Halonotius-like sequences (OTU6 and 7) with relative abundance of 30%, Haloquadratum-like sequences (OTU1, 2, and 3), Halorubrum-like sequences (OTU8 and 12) and Haloarcula-like sequences (OTU9), with a relative abundance of 25%, 22.5%, and 7.5% respectively ( Table 3). The remaining Halobacteriales clones (OTU10, 11, and 4) with a relative abundance of 12.5% were not affiliated to any described genera and were therefore form a new phylogenetic lineage (Table 3). Similar unclassified Halobacteriales members have been described in other extreme saline environments as well [21,44,102,103]. The distribution of the affiliation of clones' sequences of CDW is more or less similarly to CDR, except for the unclassified Halobacteriaceae which they were overrepresented in CDR (33%) than in CDW (12%).
(iii) CDZ sample was characterized by a restricted number of clones (n = 11) and whence a low number of OTUs (n = 2), and all of them were affiliated to uncultured Halorubrum related genus members (Table 3 and Figure 6).
Collectively, these results gave a general idea on the diversity of archaeal community members in three different halite-crystal salts samples based on almost complete 16S rRNA gene libraries. Globally, the community was dominated by clones' sequence affiliated to Halotonius-like sequences genera in the two samples CDR and CDW, and the members of Halorubrum genus were observed in the three halite-crystal salts samples. Our results are in accordance with other studies assessing the diversity of halophilic Archaea in saturated salt samples like in Korean (33% salt), Secovlje (25-30% salt) and Maras salterns (25-31% salt) showing that almost all sequences were grouped within members of Halorubrum genus [104]. maximum composite likelihood and the topology was inferred using the neighbor-joining method using MEGA 6 software. Numbers on the nodes present % bootstrap values based on 1000 replicates. Scale bar represents 0.05 substitutions per site. The 16S rRNA gene sequence of Sulfolobus metallicus was arbitrarily chosen as the outgroup to define the root of the tree. Accession number of sequences are shown in parenthesis.

Archaeal Profiling Using MiSeq-Based High-Throughput 16S rRNA Sequencing Approach
A third method was used to improve our understanding on archaeal population present in the three halite-crystal salts based on MiSeq sequencing using the Illumina platform. A total of 95,144 16S rRNA reads for the three datasets was obtained. It is worth noting that only OTUs with a relative abundance ≥1% were considered in the analysis.

Diversity and Richness Estimates
An overview on the alpha diversity indices of the archaeal community is illustrated in Table 4. A good coverage rate was obtained at cut-off value of 3% for all samples ( Figure S2). Good's coverage with values >99% displays saturated sampling of all samples which indicate almost sampling completeness (Table 4). Indeed, the results indicated that CDR had the greatest archaeal diversity, followed by CDW and CDZ (Shannon in- Figure 6. Neighbor-joining tree showing relationships among CDZ clones' and their closest relatives retrieved from RDPII database. Evolutionary distances were calculated using the method of maximum composite likelihood and the topology was inferred using the neighbor-joining method using MEGA 6 software. Numbers on the nodes present % bootstrap values based on 1000 replicates. Scale bar represents 0.05 substitutions per site. The 16S rRNA gene sequence of Sulfolobus metallicus was arbitrarily chosen as the outgroup to define the root of the tree. Accession number of sequences are shown in parenthesis.

Archaeal Profiling Using MiSeq-Based High-Throughput 16S rRNA Sequencing Approach
A third method was used to improve our understanding on archaeal population present in the three halite-crystal salts based on MiSeq sequencing using the Illumina platform. A total of 95,144 16S rRNA reads for the three datasets was obtained. It is worth noting that only OTUs with a relative abundance ≥1% were considered in the analysis.

Diversity and Richness Estimates
An overview on the alpha diversity indices of the archaeal community is illustrated in Table 4. A good coverage rate was obtained at cut-off value of 3% for all samples ( Figure S2). Good's coverage with values >99% displays saturated sampling of all samples which indicate almost sampling completeness (Table 4). Indeed, the results indicated that CDR had the greatest archaeal diversity, followed by CDW and CDZ (Shannon indices of 3.55, 3.25, and 1.245, respectively). In terms of archaeal richness, CDR (134.7) had the greatest chao1 value, and CDZ (11) had the lowest one.

Class and Genus Levels Distribution
At class level, all OTUs were assigned to the class Halobacteria for CDZ sample, while for CDW and CDR samples archaeal representatives span the classes Halobacteria with a relative abundance of 95.73% and 88.95% respectively, and Nanohaloarchaea with a rel-ative abundance of 4.62% and 11.04% for CDR and CDW respectively. The dominance of members of Halobacteria versus Nanohaloarchaea is in agreement with previous reports focused on the study of archaeal diversity in NaCl saturated samples like halite nodules, crystallizer ponds, crusts of halite and commercial salts, based on high-throughput sequencing methods [105][106][107][108][109]. It is interesting that Nanohaloarchaea was not reported in earlier studies as a component of archaeal community in hypersaline environment because of limitations of the approaches used before to evaluate the microbial diversity until the introduction of NGS technologies [110], citing the examples of the mismatches of primers used targeting the 16S rRNA gene copies with Nanohaloarchaea genes regions' or the lack of their sequences in databases used for similarity searches [111,112]. In fact, Nanohaloarchaea, is a group of nanosized halophilic archaea lineage first recuperated from the soda lake Magadi in the east-African Rift Valley [113] and then from several hypersaline ecosystems where some complete genomes of uncultivated "Candidatus Nanohaloarchaea" have been assembled from metagenomes analysis [114,115]. Nanohaloarchaea were first affiliated to Halobacteria class within the phylum Euryarchaeota based on 16S rRNA phylogeny and then based on the shotgun genomes assembling, a novel phylum "Nanohaloarchaeota" within a "DPANN" superphylum (Diapherotrites, Parvarchaeota, Aenigmarchaeota, Nanoarchaeota, and Nanohaloarchaea) was proposed [106,[116][117][118]. Nanohaloarchaea members are characterized by small size, a one-copy rRNA operon and low GC-content [106].
At genus level, 37 genera out of 51 currently described [40] were identified in all samples. A clear difference in genera distribution and abundance were observed (Figure 7). With a totally number of 140 OTUs, CDR is considered as the most diverse sample distributed on 43 genera with the abundance of Halorubrum genus (35.11%). For CDW sample, with 49 OTUs, 18 genera were identified, with two abundant genera Halonotius (17.83%) and Halorhabdus (16.28%). Lowest diversity was observed in CDZ sample representing only by 10 OTUs distributed on five genera with the abundance of Halogeometricum genus (64.7%). The abundance of such genera in the halite-salt crystal can reflect their physiological and ecological completeness and their survival mechanisms they deploy to cope with the high salt concentrations and anorexic conditions. By comparing the community profiles of CDR and CDW samples collected from the same location (Chott El Jeridi) and characterized by red and white colors respectively, a clear difference on genera composition was observed. CDR sample is characterized by the abundance of Halorubrum genus (35.11%), this genus is known by the high capacity in Some unclassified genera members (Halobacteriaceae Family) were observed in all samples as well. Further comparison showed that Halorhabdus, Halorubrum and Haloarcula genera constitute the main encountered genera within the datasets of the three samples. These results are in accordance with those carried out on several crystals and rock salts samples, where at least one of the archaeal genera has been found [17,43,119].
By comparing the community profiles of CDR and CDW samples collected from the same location (Chott El Jeridi) and characterized by red and white colors respectively, a clear difference on genera composition was observed. CDR sample is characterized by the abundance of Halorubrum genus (35.11%), this genus is known by the high capacity in carotenoids biosynthesis [120], which may explain the observed red pigmentation. However, for CDW sample two main abundant genera were observed which Halotonuis (17.83%) and Halorhabdus (16.28%) are known as light and non-pigmented genus. Indeed, CDZ sample, white halite sample collected from another chott Douz, showed a different profile, with the abundance of Halogeomtericum genus (64.70%).
It is noticeable that, the mainly observed genera, especially those affiliated to Nanohaloarchaea family, were only identified with Illumina MiSeq sequencing approach for CDR and CDW samples. Further genera were identified only with the full-length 16S rRNA archaeal libraries approach, the case of Halonotius, Halolamina, Haloquadratum, and Halorubellus for CDR sample (Table S1), and Halonotius for CDW sample (Table S2). Based on cultured dependent approach, only some isolates affiliated to Haloferax, Halorubrum, Halorcula, Halogeometricum, and uncultured haloarchaeal strains have been successfully isolated from all samples which may due to the still unknown growth requirements of the species [121]. The discrepancy between the haloarchaeal diversity analysis achieved by all approaches are in accordance with those reported by other authors [43,122].

Conclusions
In the present study, we assessed the haloarchaeal diversity in three halite-crystal salts (CDW, CDR, and CDZ) based on culture-independent approaches targeting different regions of 16S rRNA gene sequences including DGGE (V3 regions), 16S rRNA clones libraries (full length gene) and Illumina Miseq sequencing (V3-V4 regions), and classical culture-dependent approach. Results showed unexpected diversity profiles, a clear difference on archaeal profile distribution and abundance between samples based on each approach. The higher diversity was observed with NGS approach, nevertheless, some genera were detected only with 16S rRNA clones' libraries and cultured approaches. In fact, the amalgamation of all results yielded a comprehensive view of the archaeal diversity represented by members of Halobacteria and Nanohaloarchaea classes. The main core genera were affiliated to Halorhabdus, Halonotius, Halorubrum, Haloarcula, and unclassified members from Halobacteriacaea family. Hence, further studies based on whole-genome shotgun (WGS) method could enhance our understanding in the functionality and the mechanisms adopted by haloarchaea members to colonize halite-crystal samples.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/biology10050397/s1, Figure S1: Archaeal DGGE patterns of 16S rDNA gene sequences of the V3 variables regions' (567 bp) obtained from different salt samples. Lane 1: CDR; lane 2: CDW; lane 3: CDZ. Bands marked with a numbered asterisk were excised from gel, re-amplified and sequenced. The gradient of the urea and formamide ranged from 20% to 55%, Figure S2: Genus-level rarefaction curves of CDZ, CDR and CDW sample based on the 16S rDNA sequences analyzed, Figure S3: Venn diagrams showing the unique and shared OTUs (3% distance level) identified based on cultured, 16S rRNA libraries, DGGE and metataxonomic approaches, for CDR, CDW and CDZ respectively, Table S1: Unique and shared genera between CDW, CDR and CDZ samples based on cultured, 16S rRNA libraries, DGGE and metataxonomic approaches, Table S2: List of shared and unique genera identified in CDW sample based on both cultured and uncultured-dependant approaches