Heme and a five-amino-acid hemophore region form the bipartite stimulus triggering the has signaling cascade

Abstract

Bacterial cells sense the extracellular environment and adapt to that environment by activating gene regulation circuits, often by means of signaling molecules. The Serratia marcescens hemophore is a signaling molecule that acts as an extracellular heme-scavenging protein. The heme-loaded hemophore interacts with its cognate receptor (HasR), triggering transmembrane signaling and turning on transcription of hemophore-dependent heme uptake genes. We investigated the features of the holo-hemophore, the only HasR ligand known to act as an inducer. We used a hemophore mutant that does not deliver its heme and a HasR mutant that does not bind heme, and we showed that heme transfer from the hemophore to the receptor is necessary for induction. Using a hemophore mutant that does not bind heme and that blocks heme transport, we demonstrated that two molecules that do not interact (heme and the mutant hemophore) may nonetheless induce this system. These findings suggest that hemophore-mediated induction and heme transport involve different mechanisms. The hemophore region important for induction was precisely localized to amino acids 50 to 55, which lie in one of the two HasR-binding hemophore regions. This bipartite stimulus probably corresponds to a physiological process because heme is transferred to the receptor before apo-hemophore release. This bipartite regulation mechanism may allow the bacterium to adjust its heme transport mechanism to the perceived environmental heme concentration.

FIG. 1.

Location of single amino acid change (yellow) and insertion site of the pentapeptide (purple) in the three-dimensional structure of HasA.

FIG. 2.

Effect of HasA mutants, affected in heme binding or release, on hasR-lacZ expression. Strain POP3BF1(pAMrbshasISR) was grown in M63B1 glucose medium in the presence and absence of the heme-loaded form of wild-type or mutant hemophores at a concentration of 10⁻⁷ M. β-Galactosidase activity was assayed as described in Materials and Methods. The β-galactosidase activities are means of at least three experiments; the error bars indicate one standard deviation. WT, wild type.

FIG. 3.

Effect of heme transfer from the hemophore to the receptor on hasR-lacZ expression. Strain POP3BF1(pAMrbshasISR) and a derivative of this strain containing a mutant copy of hasR(H1-H2) were grown in M63B1 glucose medium in the presence and absence of wild-type heme-loaded HasA (holo-HasA) at a concentration of 10⁻⁷ M. β-Galactosidase activity was assayed as described in Materials and Methods. The β-galactosidase activities are means of at least three experiments; the error bars indicate one standard deviation. WT, wild type.

FIG. 4.

Role of the HasR-binding regions of HasA in hasR-lacZ induction. Strain POP3BF1(pAMrbshasISR) was grown in M63B1 glucose medium in the presence and absence of heme-loaded wild-type or variant hemophores having mutations in the binding regions of HasA (A) or all along HasA (B) at a concentration of 10⁻⁷ M. β-Galactosidase activity was assayed as described in Materials and Methods. The β-galactosidase activities are means of at least three experiments; the error bars indicate one standard deviation. WT, wild type.

FIG. 5.

Effect of pentapeptide insertions in HasA binding region 51-60 on hasR-lacZ expression. Strain POP3BF1(pAMrbshasISR) was grown in M63B1 glucose medium in the presence and absence of heme-loaded wild-type or mutant HasA hemophore at a concentration of 10⁻⁷ M. β-Galactosidase activity was assayed as described in Materials and Methods. The β-galactosidase activities are means of at least three experiments; the error bars indicate one standard deviation. WT, wild type.

FIG. 6.

Induction of hasR-lacZ by two noninteracting molecules, either heme and triple mutant HasA H32AY75AH83A, heme and quadruple mutant HasA H32AY75AH83AS51P, or heme and quadruple mutant HasA H32AY75AH83AG95V. Strain POP3BF1(pAMrbshasISR) was grown in M63B1 glucose medium in the presence and absence of heme-loaded HasA (holo-HasA) at a concentration of 10⁻⁷ M. β-Galactosidase activity was assayed as described in Materials and Methods. The β-galactosidase activities are means of at least three experiments; the error bars indicate one standard deviation. Bar A, wild-type holo-HasA at a concentration 10⁻⁷ M; bar B, heme at a concentration of 10⁻⁵ M; bar C, apo-HasA H32AY75AH83A at a concentration of 10⁻⁷M; bar D, apo-HasA H32AY75AH83A at a concentration of 10⁻⁷ M plus heme at a concentration of 10⁻⁷ M; bar E, apo-HasA H32AY75AH83AG95V at a concentration of 10⁻⁷ M; bar F, apo-HasA H32AY75AH83AG95V at a concentration of 10⁻⁷ M plus heme at a concentration of 10⁻⁷ M; bar G, apo-HasA H32AY75AH83AS51P at a concentration of 10⁻⁷ M; bar H, apo-HasA H32AY75AH83AS51P at a concentration of 10⁻⁷ M plus heme at a concentration of 10⁻⁷ M.