Table 3.

Polarity relief at phage t_R1 terminator in viable double and triple mutants^a

Genotype	Mean ± SE
	Without H-NSΔ64			With H-NSΔ64
	–tR1 (A)	+tR1 (B)	% B/A	–tR1 (A)	+tR1 (B)	% B/A
rho(Q32R)	5,728 ± 559	1,732 ± 339	30 ± 3	5,325 ± 314	832 ± 207	15 ± 3
rho(R102S, A243E)	2,275 ± 360	1,648 ± 161	74 ± 5	1,994 ± 462	738 ± 171	37 ± 0
rho(Q32R) nusA(R258C)	5,301 ± 451	2,818 ± 388	53 ± 5	4,682 ± 424	1,468 ± 232	32 ± 6
rho(Q32R) nusG(G146D)	4,156 ± 137	1,439 ± 214	34 ± 5	4,541 ± 664	832 ± 155	18 ± 1
rho(A243E) nusA(R258C)	NDb	ND	ND	4,465 ± 286	1,836 ± 425	40 ± 7
rho(R102S, A243E) nusA(R258C)	ND	ND	ND	2,144 ± 605	1,033 ± 169	52 ± 7
nusG(G146D) nusA(R258C)	ND	ND	ND	4,677 ± 546	1,370 ± 95	30 ± 4
rho(Q32R) nusG(G146D) nusA(R258C)	ND	ND	ND	3,779 ± 47	1,295 ± 78	34 ± 2

^aThe specific activities of β-galactosidase were determined for various pairs of strains without (-t_R1 [A]) or with (+t_R1 [A]) the phage t_R1 terminator upstream of the lacZ reporter gene. The B/A ratios (%) given in the table are a measure of transcriptional readthrough or the polarity relief at the t_R1 terminator. Where indicated, H-NSΔ64 expression was obtained by transformation of the strains with plasmid pLG-H-NSΔ64. The strain pairs (–t_R1 and +t_R1) used were as follows: rho(Q32R), GJ10798 and GJ10795; rho(R102S, A243E), GJ10802 and GJ10801; rho(Q32R) nusA(R258C), GJ10800 and GJ10797; rho(Q32R) nusG(G146D), GJ10799 and GJ10796; rho(A243E) nusA(R258C), GJ10814 and GJ10813; rho(R102S, A243E) nusA(R258C), GJ10794 and GJ10793; nusG(G146D) nusA(R258C), GJ10812 and GJ10813; and rho(Q32R) nusG(G146D) nusA(R258C), GJ10817 and GJ10818. Strains with the rho(R102S, A243E) mutation were grown in glucose-minimal A medium supplemented with Casamino Acids at 0.5%. All other strains were grown in LB medium.

^bND, not determined because these mutant combinations exhibited synthetic lethality in the absence of H-NSΔ64.