Sulfate

The kinetics were investigated in sealed 250 mL glass bottles in which an equal amount of 200 mL of the previously autoclaved medium, without agar, and 50 mL of inoculum were added and put inside the anaerobic chamber. Then the bottles were continuously mixed in a mechanical shaker at 120 RPM at 38 °C. At certain time intervals, one bottle was selected, aliquots withdrawn and used for the chemical and biological analysis. The amount of Na₂SO₄ and FeSO₄·7H₂O of the modified Postgate medium were proportionately modified in the culture medium to reach the concentrations of 1790 mg/L; the amount of the other components of the culture medium was the same.
During the kinetic experiment samples were withdrawn at suitable time intervals and then after acidification the biomass concentrations are evaluated using the calibration curve in Fig. 1. The sulfate and sulfide solution content was measured by turbidimetric methods. Barium sulfate was added to the samples to precipitate the sulfate ions as barium sulfate and copper sulfate was added to precipitate the sulfide ions as copper sulfide. For the pH measurements a pH meter calibrated using buffer solutions of pHs of 4 and 7 was used. The redox potentials were measured using an ORP electrode with an internal Ag/AgCl reference electrode [1] , [2] .

Four batches of banana peel-supported sulfonic acid (BP-SO₃H) catalysts were prepared by varying the banana peel-to-sulfuric acid ratio, reaction time, and temperature. Catalysts prepared in the initial batch include dried banana peel powder (1 g) mixed thoroughly with conc. H₂SO_4. Banana peel powder: H₂SO₄ ratios (g L⁻¹) of 1:5, 1:10, 1:15, and 1:20 were used, whereas the reaction temperature varied between 80 and 120 °C while reaction time was monitored for 16, 18, 20, 22, and 24 h. In order to remove residual sulfate ions in the filtrate, 50–60 mL of deionized water was added to the mixture. It was washed several times with hot deionized water until no residual sulfate ions could be detected (a 6 mol L⁻¹ solution of BaCl₂ was used for testing). The resultant sulfonic acid functionalized banana peel (BP-SO₃H) was dried in an oven overnight. All the synthesized BP-SO₃H catalysts were assigned a code based on their synthesis procedure; BP-SO₃H-X-Y-Z, where X is the wt/volume ratio of banana peel/H₂SO₄, Y is the 'in 'situ' hydrothermal-sulfonation time, and Z stands for reaction temperature. Accordingly, the catalyst prepared using a 1:10 (wt/volume) banana peel/sulfuric acid ratio, hydrothermal sulfonation time of 16 h, and reaction temperature of 80 °C were designated as BP-SO₃H-10-16-80.

Molecular dynamics (MD)
simulations were carried out on a number of representative structures
for CM. They included two independent sets of simulations for apo
MtCM, starting either from the X-ray crystal structure of MtCM in
complex with malate (after removing malate) (PDB ID: 2VKL)^{21 (link)} or from the structure of the CM polypeptide in the apo
MtCM–MtDS complex (PDB ID: 2W19,^{21 (link)} chain D).
The malate complex was chosen over ligand-free MtCM (PDB ID: 2QBV)^{41 (link)} due to its higher resolution and better refinement statistics.
Both simulations gave essentially the same result; therefore, we will
not refer to the second data set any further. For the highly active
evolved MtCM variant (MtCM^V), we used the recent crystal
structure (PDB ID: 5MPV).^{12 (link)} The MtCM–ligand complex (MtCM^LC) was taken from PDB ID: 2W1A,^{21 (link)} excluding
the MtDS partner protein, where MtCM was co-crystallized with a transition
state analog (TSA) in its active site (Figure 1). Finally, the V55D variant was modeled
based on a partially refined experimental structure (Table S1). Residues that were not fully defined were added
to the models using (often weak) electron density maps as reference
in Coot.^{37 (link)} When no interpretable
density was visible, geometric restraints (and α-helical restraints
for residues in helix H1) were applied during model building, to ensure
stable starting geometries. The N-termini of all of the models were
set at Glu13, corresponding to the first defined residue in almost
all of the resolved structures available. Glu13 was capped with an
acetyl group to imply the continuation of the H1 helix. CM dimers
were generated by 2-fold crystallographic symmetry.
Missing
H-atoms were added to the model and the systems were solvated in a
periodic box filled with explicit water molecules, retaining neighboring
crystallographic waters, and keeping the protein at least 12 Å
from the box boundaries. The systems were neutralized through the
addition of Cl^– ions at a minimum distance of 7
Å from the protein and each other. Additional buffering moieties
like glycerol or sulfate ions found in the crystals were not considered.
MD simulations were run using the Gromacs 5.1.4 package^{42 (link),43 (link)} using the AMBER 12 force fields for the protein moieties^{44 (link),45} and the TIP3P model for water.^{46 (link)} The
ligand was modeled using the GAFF force field.^{47 (link)} The smooth particle mesh Ewald method was used to compute
long-range electrostatic interactions,^{48 (link)} while a cutoff of 11 Å was used to treat the Lennard–Jones
potential.
The systems were minimized using the steepest descent/conjugate
gradients algorithms for 500/1500 steps until the maximum force was
less than 1000 kJ mol^–1 nm^–1.
To equilibrate and heat the systems, first we ran 100 ps MD in the
NVT ensemble starting from a temperature of 10 K, using the canonical
velocity rescaling thermostat^{49 (link)} followed
by 100 ps in the NpT ensemble with a Parrinello–Rahman barostat^{50 (link)} targeting a final temperature of 310 K and a
pressure of 1 atm. After initial equilibration, 1 μs of MD simulation
was performed for each system. In all MD simulations, the time step
size was set to 2 fs.