Table 1.

MD simulations

Simulation namea	Assembly	Charge state				Starting structureb	Run length
		D133	D163	D164	K300		µs
S1/1	dimer	—	—	0	+	xtal*	1.1
S1/2						S1/1@0.1 µs	0.2
S1/3						S1/1@0.1 µs	0.2
S2/1	dimer	—	—	—	+	xtal*	1.0
S2/2						xtal	1.0
S2/3						xtal	1.0
S2/1.1–1.5	monomer					S2/1	5 × 0.1
S2/2.1–-2.2						4AU5*	2 × 0.1
S3/1	dimer	—	0	0	+	S2/1*	0.1
S3/2						S2/1	1.0
S3/3						xtal*	0.1
S4/1	dimer	—	—	—	0	S2/1*	1.0
S4/2						S4/1@0.1 µs	1.0
S4/3						S4/1@0.1 µs	0.2
S5/1	dimer	0	—	0	+	xtal*	1.1
S5/2						S5/1@0.1 µs	0.2
S5/3						S5/1@0.1 µs	0.2
S6/1	dimer	0	—	—	+	S2/1*	0.1
S6/2						xtal*	0.1
S7	dimer	0	0	0	+	xtal*	0.1
S8	dimer	0	—	—	0	S2/1*	0.1

Asterisks denote simulations that were preceded by energy minimization and a 3-ns position restraint MD; simulations without an asterisk were repeats starting from the same initial system conformation as the starred one but with varied initial velocity distribution. The dimer simulation contained 112,700 atoms and the monomer contained 52,332. All simulations were performed with Gromacs 4.6.1 except simulations S2/1, S2/1.1–1.5, and S2/2.1–2.2, which were run in Gromacs 4.5.3. Judging from the RMSD and secondary structure analysis (not depicted), the simulations behaved in the same manner regardless of software version. In total, 10.5 µs of MD simulations was performed. xtal, crystal structure of wild-type NhaA, which was deposited in the Protein Data Bank under accession number 4AU5 after minor refinements; 4AU5, wild-type NhaA as deposited in the Protein Data Bank; S1/1, last frame of simulation S1/1, etc., or frame at 0.1 µs (“@0.1µs”).

^aSimulations are identified by the protonation states and resulting charge states of Asp133 (D133), Asp163 (D163), Asp164 (D164), and Lys300 (K300), using identifiers S1–S8. Repeat simulations are indicated with a serial number after the identifier.

^bStarting structure denotes the source for the initial input structure for the simulation.