The Sulfate-Rich and Extreme Saline Sediment of the Ephemeral Tirez Lagoon: A Biotope for Acetoclastic Sulfate-Reducing Bacteria and Hydrogenotrophic Methanogenic Archaea

Our goal was to examine the composition of methanogenic archaea (MA) and sulfate-reducing (SRP) and sulfur-oxidizing (SOP) prokaryotes in the extreme athalassohaline and particularly sulfate-rich sediment of Tirez Lagoon (Spain). Thus, adenosine-5′-phosphosulfate (APS) reductase α (aprA) and methyl coenzyme M reductase α (mcrA) gene markers were amplified given that both enzymes are specific for SRP, SOP, and MA, respectively. Anaerobic populations sampled at different depths in flooded and dry seasons from the anoxic sediment were compared qualitatively via denaturing gradient gel electrophoresis (DGGE) fingerprint analysis. Phylogenetic analyses allowed the detection of SRP belonging to Desulfobacteraceae, Desulfohalobiaceae, and Peptococcaceae in ∂-proteobacteria and Firmicutes and SOP belonging to Chromatiales/Thiotrichales clade and Ectothiorhodospiraceae in γ-proteobacteria as well as MA belonging to methylotrophic species in Methanosarcinaceae and one hydrogenotrophic species in Methanomicrobiaceae. We also estimated amino acid composition, GC content, and preferential codon usage for the AprA and McrA sequences from halophiles, nonhalophiles, and Tirez phylotypes. Even though our results cannot be currently conclusive regarding the halotolerant strategies carried out by Tirez phylotypes, we discuss the possibility of a plausible “salt-in” signal in SRP and SOP as well as of a speculative complementary haloadaptation between salt-in and salt-out strategies in MA.

2) AB: Amino proportions according to Rhodes (2010Rhodes ( , 2010. 3) GC content percentage is calculated as: GC%= (G+C/G+C+A+T) * 100  (Tajima, 1989) compares the number of segregating sites per site with the nucleotide diversity (a site is considered segregating if, in a comparison of m sequences, there are two or more nucleotides at that site; nucleotide diversity is defined as the average number of nucleotide differences per site between two sequences). If all the alleles are selectively neutral, then the product 4Nv (where N is the effective population size and v is the mutation rate per site) can be estimated in two ways, and the difference in the estimate obtained provides an indication of non-neutral evolution. When D is equal or near to 0 then, the sequence(s) are neutrally evolving, whilst D > 0 indicates a positive selection and D < 0 indicates negative or purifying selection of the sequence(s). The analysis involved the same amino acid alignments used for the phylogenetic reconstruction (described below), and it was conducted in MEGA5 (Tamura et al., 2007).  Figure S1. Alignment of 100 amino acid sequences with the corresponding fragment to the AprA, N-terminal domain. AprA from 2 to 261 amino acid positions has been characterized in Archaeoglobus fulgidus in the reduced state (FAD red -APS, PDB ID: 1JNR) (Fritz et al., 2002) and in the oxidized state (FAD ox -APS, PDB ID: 2FJA) (Schiffer et al., 2006) as well as in Desulfovibrio gigas (PDB ID: 3GYX) (Chiang et al., 2009). This alignment 1 includes two of the nine functional active sites (marked with black background) of the Apr_alpha_N domain: Arg-R α265 and Trp-W α234 (Klein et al.) reported by Schiffer et al., (2006). Essential catalytic amino acids for cofactor and nucleotide binding sites are marked with grey backgrounds as reported from PDB and UniProt databases. Same species from different strain with interesting amino acid changes from basic (Lys, K) to polar (Gln, Q and Asn, N) are marked with blue backgrounds. 122 from 137 positions were used for the phylogenetic reconstruction.  .|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|....|..  Archaeoglobus fulgidus (1JNR-A) WQIMIHGESYKPIIAEAAKMAV  The iron-sulfur flavoenzyme adenosine-5'-phosphosulfate (APS) reductase (Apr) (EC: 1.8.99.2). Apr catalyzes a key reaction of the sulfur cycle by reversibly transforming APS to sulfite and AMP. The dissimilatory sulfate reduction operates under anaerobic conditions in Bacteria and Archaea (Schiffer et al., 2006). Apr consists of 22 heterotetrameres (Fritz et al., 2002). The -subunit (75 kDa) harbors the FAD cofactor. In the figure S2a, the Apr -subunit is shown (blue) and the -subunit (red). The evolutionary sequence region used in this work (see the alignment above) represents the N-terminal domain of the subunit (red rectangle), which is comprised of the [4Fe-4S] clusters and the FAD binding site (marked as ball-and-stick representations). The active site channel located between the central and capping domains of the -subunit is filled with water molecules depicted as green spheres (Schiffer et al., 2006).

(A)
Apr in the FADox-APS state. Apr binds in a curved conformation to the prebuilt channel. Invariant residues His-H 398 , Asn-N 74 and Arg-R 265 play a key role in phosphosulfate binding and catalysis. The adenine part is firmly clamped between Leu-L 278 and Arg-R 317 and the latter swings into the channel upon substrate binding (Schiffer et al., 2006).

(B)
Apr in the FAD-sulfite-AMP1 state. This state contains a FADsulfite adduct and AMP, the latter adopting a conformation similar to the AMP moiety of APS. This structure shows how the phosphate of AMP and the sulfite of the FAD-sulfite adduct (both negatively charged) are kept in van der Waals contact (Schiffer et al., 2006).
The binding affinity of APS for Apr is reflected by protein-substrate interactions. The sulfate O1 atom interacts with the ND2 atom of Asn-N 74 , its O2 atom with the NE2 atom of His-H 389 , and its O3 atom via a water molecule with Arg-R 265 and Trp-W 234 and via two water molecules with Glu-E 141 , Asp-D 361 and Asn-N 74 . Additionally, Arg-R 265 is hydrogen bonded to the O3A atom that links the sulfate and phosphate groups. The phosphate oxygen atoms are connected to Arg-R 265 via a salt bridge and to the peptide nitrogens of Val-V 273 and Gly-G 274 , located at the positively charged N-terminus of helix R6. The hydroxyl group O2 of ribose is linked to the side chain hydroxyl group of Tyr-Y 95 and that of O3 to His-H 446 and Tyr-Y 95 via a water molecule. The fixation of the adenine ring of APS is mainly accomplished by sandwiching it between Leu-L 278 and Arg-R 317 whereby the coplanar arrangement of the guanidinium group is optimal for - interactions. Additionally, one polar contact is formed between the N3 atom of the adenine ring and the protein mediated by a solvent molecule (Schiffer et al., 2006).  Figure S1. b) Relative amino acid composition of the AprA catalytic region. The frequency of the 20 amino acids present in the alignment (above) was plotted across phylotypes as a range of percentages: between 0-5% (blue), 5-10% (red) and 10-15% (green). Figure S1. c) Nucleotide codon composition for the aprA marker in gene sequences used in this study. Methyl-coenzyme M reductase (Mcr) (EC: 2.8.4.1) is a key enzyme in formation in methanogenic Archaea. It catalyzes the reduction of methyl-coenzyme M (methyl-CoM) with coenzyme B (Anderson et al.) to methane and the heterodisulfide of CoM and CoB (Ermler et al., 1997). This reaction proceeds under anaerobic conditions and requires a nickel-porphinoid prosthetic group, coenzyme F 430 , which is in the EPR-detectable Ni (I) oxidation state in the active enzyme. Mcr is a 300 kDa hexamer protein arranged as  subunits with two identical nickel porphinoid active sites, which form two long active site channels with F 430 embedded at the bottom (Grabarse et al., 2000). The evolutionary sequence region used in this work (see the alignment above) represents the C-terminal domain of the -subunit (red rectangle), which is comprised of an all-alpha multi-helical bundle (blue and turquoise helices).

(A)
The active site region of the Mcr ox1-silent structure. The binding positions of the coenzymes suggest the active site between the nickel of coenzyme F 430 and the sulfur atom of CoB. The active site is coated mostly by non-polar and aromatic residues. Five mutually contacting Phe and Tyr side chains are arranged as ring forming a tunnel (Ermler et al., 1997).

(B)
The active site region of the Mcr silent structure. Compared with the Mcr ox1silent structure, CoM has moved through the tunnel to form with CoB a heterodisulfide, the oxidation product of the reaction. The sulfonate moiety of CoM lost its interactions to the protein matrix and is coordinated to the Ni atom (Ermler et al., 1997).
The relative arrangement of the three coenzymes (CoM, CoB and CoF 430 ) suggests that the catalytic reaction takes place at the front side of CoF 430 in the channel between the nickel and the thiol groups of CoB. Each active site is lined up by an annular arrangement of Phe-F 330 , Tyr-Y 333 , Phe-F 443 , Phe-F 361 and Tyr-Y 367 flanked by further hydrophobic and aromatic residues. As can be seen in the alignment above, the Phe-F 330 , Tyr-Y 333 , Phe-F 443 amino acids are completely conserved in all Mcr sequences included in this work. Other studies on a catalytically inactive enzyme aerobically co-crystallized with CoM displayed a fully occupied CoM-binding site with no alternate conformations. The binding of CoM appears to induce specific conformational changes that suggest a molecular mechanism by which the enzyme ensures that methylcoenzyme M enters the substrate channel prior to CoB, as required by the activesite geometry (Ermler et al., 1997). Figure S2. b) Relative amino acid composition of the McrA catalytic region. The frequency of the 20 amino acids present in the alignment (above) was plotted across phylotypes as a range of percentages: between 0-5% (blue), 5-10% (red), 10-15% (green) and 15-20% (violet). Figure S2. c) Nucleotide codon composition for the mcrA marker in gene sequences used in this study. Figure S3. Correspondence analysis of Relative Synonymous Codon Usage (RSCU) for McrA sequences from halophiles, non-halophiles and Tirez phylotypes. The distribution of all codons (including the start and stop codons *) for every amino acid across the three datasets is shown on the X axis. The frequency of each codon (%) is represented with bars on the left Y axis. RSCU values for each codon across the three datasets are represented with differentiated dots on the right Y axis. In the absence of any codon usage bias, the RSCU value would be 1.00. A codon that is used less frequently than expected will have a value of less than 1.00 and vice versa for a codon that is used more frequently than expected. Figure S4. Diagram of the target sites for primer pairs within mcr. Modified from (Hales et al., 1996)