Bacteria respond to diverse growth-limiting stresses by producing a large s
et of general stress proteins. In Bacillus subtilis and related Gram-positi
ve pathogens, this response is governed by the sigma (B) transcription fact
or. To establish the range of cellular functions associated with the genera
l stress response, we compared the transcriptional profiles of wild and mut
ant strains under conditions that induce sigma (B) activity. Macroarrays re
presenting more than 3900 annotated reading frames of the B. subtilis genom
e were hybridized to P-33-labelled cDNA populations derived from (i) wild-t
ype and sigB mutant strains that had been subjected to ethanol stress; and
(ii) a strain in which sigma (B) expression was controlled by an inducible
promoter. On the basis of their significant sigma (B)-dependent expression
in three independent experiments, we identified 127 genes as prime candidat
es for members of the sigma (B) regulon. Of these genes, 30 were known prev
iously or inferred to be sigma (B) dependent by other means. To assist in t
he analysis of the 97 new genes, we constructed hidden Markov models (HMM)
that identified possible sigma (B) recognition sequences preceding 21 of th
em, To test the HMM and to provide an independent validation of the hybridi
zation experiments, we mapped the sigma (B)-dependent messages for seven re
presentative genes. For all seven, the 5' end of the message lay near typic
al sigma (B) recognition sequences, and these had been predicted correctly
by the HMM for five of the seven examples. Lastly, all 127 gene products we
re assigned to functional groups by considering their similarity to known p
roteins. Notably, products with a direct protective function were in the mi
nority. Instead, the general stress response increased relative message lev
els for known or predicted regulatory proteins, for transporters controllin
g solute influx and eff lux, including potential drug efflux pumps, and for
products implicated in carbon metabolism, envelope function and macromolec
ular turnover.