Complementation of cold shock proteins by translation initiation factor IF1 in vivo

Abstract

The cold shock response in both Escherichia coli and Bacillus subtilis is induced by an abrupt downshift in growth temperature and leads to a dramatic increase in the production of a homologous class of small, often highly acidic cold shock proteins. This protein family is the prototype of the cold shock domain (CSD) that is conserved from bacteria to humans. For B. subtilis it has been shown that at least one of the three resident cold shock proteins (CspB to D) is essential under optimal growth conditions as well as during cold shock. Analysis of the B. subtilis cspB cspC double deletion mutant revealed that removal of these csp genes results in pleiotropic alteration of protein synthesis, cell lysis during the entry of stationary growth phase, and the inability to differentiate into endospores. We show here that heterologous expression of the translation initiation factor IF1 from E. coli in a B. subtilis cspB cspC double deletion strain is able to cure both the growth and the sporulation defects observed for this mutant, suggesting that IF1 and cold shock proteins have at least in part overlapping cellular function(s). Two of the possible explanation models are discussed.

FIG. 1

Comparison of the structures of cold shock protein CspB from B. subtilis (A, PDB accession no. 1CSP [37]) and IF1 from E. coli (B, structure 4 of PDB accession no. 1AH9 [44]). Ribbon models were generated with Swiss-PdbViewer version 3.7b2 (15), and images were rendered by using POV-Ray version 3.1g. (C) Superimposition of IF1 (blue) on CspB (yellow) was performed by using the “iterative magic fit” function of Swiss-PdbViewer.

FIG. 2

Growth comparison of B. subtilis strains JH642 (Δ, parental strain, no phleomycin added), MW_ΔBC-IF1^Ec(■, IF1 expressing cspB/cspC double-deletion strain, 5 μg of phleomycin per ml added), and 64BC (□, IF1 nonexpressing cspB/cspC double-deletion strain, no phleomycin added) grown in glucose minimal medium supplemented with 50 μg of tryptophane per ml, 50 μg of phenylalanine per ml, and 1 mM IPTG. Mean values and standard deviation of three independent experiments are shown.

FIG. 3

PCR verification of the constructed B. subtilis strain MW_ΔBC-IF1^Ec. The specific DNA regions analyzed are indicated within the figure (cspB, cspC, cspD, pDG148, P_spac-cspB, and P_spac-infA), and lanes are identified as follows: 1, PCR analysis of parental strain JH642 harboring all three csp genes (cspB to -D); 2, PCR analysis of control strain 64BCDbt harboring no chromosomal csp gene but a copy of cspB on the pDG148 derivative plasmid pDGcspB in trans; 3, PCR analysis of the cspB cspC double deletion mutant MW_ΔBC-IF1^Ec harboring the E. coli infA gene on the pDG148 derivative plasmid pMW_infA^Ec-1; M, HindIII-digested λ-DNA was used as a marker. For PCR analysis of wild-type cspB, cspC, and cspD gene loci, amplificate sizes of ca. 0.5 kb each were expected (see JH642 wild-type control), whereas a deletion in cspB, cspC, and cspD should yield fragments of ca. 3, 1.5, and 2 kb, respectively (see 64BCDbt deletion control). For the presence of a pDG148 derivative plasmid, a PCR amplificate of ca. 3 kb was expected, and for the presence of the cspB or infA gene located downstream of a P_spac promotor, PCR amplificates of approximately 0.3 kb were expected. PCR conditions are detailed in Table 2.