Previous studies have led to a model in which the promoter-specific recognition of prokaryotic transcription initiation factor, sigma (σ), is core dependent. Most σ functions were studied on the basis of this tenet. Here, we provide in vitro evidence demonstrating that the intact Bacillus subtilis primary sigma, σA, by itself, is able to interact specifically with promoter deoxyribonucleic acid (DNA), albeit with low sequence selectivity. The core-independent promoter-specific interaction of the σA is −10 specific. However, the promoter −10 specific interaction is unable to allow the σA to discern the optimal promoter spacing. To fulfill this goal, the σA requires assistance from core RNA polymerase (RNAP). The ability of σ, by itself, to interact specifically with promoter might introduce a critical new dimension of study in prokaryotic σ function.

Sigma factors (sigmas) are bacterial transcription factors that bind core RNA polymerase (RNAP) and direct transcription initiation at cognate promoter sites. However, most of their functions have been investigated in the context of RNAP. This has made the exact function of sigma, and the importance of core RNAP in modulating sigma function, ambiguous. Here we identify a Bacillus subtilis mutant sigma(A) that is independently capable of specific binding and melting of the promoter DNA. Interestingly, specific and independent promoter binding of sigma is sufficient for the temperature- and Mg(2+)-independent melting of promoter DNA around the transcription start site, in contrast to the temperature- and Mg(2+)-dependent melting by RNAP around the promoter -10 element. Thus core RNAP is able to negatively modulate the sigma-initiated melting of the transcription start site and, by sensing the changes in temperature and Mg(2+) concentration, to regulate the efficiency of promoter -10 melting.

Sigma (σ) factors are bacterial transcription initiation factors that direct transcription at cognate promoters. The promoters recognized by primary σ are composed of -10 and -35 consensus elements separated by a spacer of 17±1 bp for optimal activity. However, how the optimal promoter spacing is sensed by the primary σ remains unclear. In the present study, we examined this issue using a transcriptionally active Bacillus subtilis N-terminally truncated σA (SND100-σA). The results of the present study demonstrate that SND100-σA binds specifically to both the -10 and -35 elements of the trnS spacing variants, of which the spacer lengths range from 14 to 21 bp, indicating that simultaneous and specific recognition of promoter -10 and -35 elements is insufficient for primary σ to discern the optimal promoter spacing. Moreover, shortening in length of the flexible linker between the two promoter DNA-binding domains of σA also does not enable SND100-σA to sense the optimal promoter spacing. Efficient recognition of optimal promoter spacing by SND100-σA requires core RNAP (RNA polymerase) which reduces the flexibility of simultaneous and specific binding of SND100-σA to both promoter -10 and -35 elements. Thus the discrimination of optimal promoter spacing by σ is core-dependent.