Promoter and transcription analysis of penicillin-binding protein genes in Streptococcus gordonii

Abstract

An optimally cross-linked peptidoglycan requires both transglycosylation and transpeptidation, provided by class A and class B penicillin-binding proteins (PBPs). Streptococcus gordonii possesses three class A PBPs (PBPs 1A, 1B, and 2A) and two class B PBPs (PBPs 2B and 2X) that are important for penicillin resistance. High-level resistance (MIC, > or =2 microg/ml) requires mutations in class B PBPs. However, although unmutated, class A PBPs are critical to facilitate resistance development (M. Haenni and P. Moreillon, Antimicrob. Agents Chemother. 50:4053-4061, 2006). Thus, their overexpression might be important to sustain the drug. Here, we determined the promoter regions of the S. gordonii PBPs and compared them to those of other streptococci. The extended -10 box was highly conserved and complied with a sigma(A)-type promoter consensus sequence. In contrast, the -35 box was poorly conserved, leaving the possibility of differential PBP regulation. Gene expression in a penicillin-susceptible parent (MIC, 0.008 microg/ml) and a high-level-resistant mutant (MIC, 2 microg/ml) was monitored using luciferase fusions. In the absence of penicillin, all PBPs were constitutively expressed, but their expression was globally increased (1.5 to 2 times) in the resistant mutant. In the presence of penicillin, class A PBPs were specifically overexpressed both in the parent (PBP 2A) and in the resistant mutant (PBPs 1A and 2A). By increasing transglycosylation, class A PBPs could promote peptidoglycan stability when transpeptidase is inhibited by penicillin. Since penicillin-related induction of class A PBPs occurred in both susceptible and resistant cells, such a mutation-independent facilitating mechanism could be operative at each step of resistance development.

FIG. 1.

Determination of the transcriptional start sites of pbp1A, pbp1B, pbp2A, and pbp2X by 5′-RACE PCR. (A) Agarose gel electrophoresis of PCR-amplified cDNA tailed with BD SMART II oligonucleotide. DNA was stained with SYBR Safe and visualized under UV light. (B) A chromatogram from the sequencing of the 5′-RACE PCR products. The BD SMART II oligonucleotide tails (dashed arrow) and the nucleotides complementary to the transcript beginnings (solid arrow) are shown.

FIG. 2.

Determination of the transcriptional start of pbp2B by using specific fusions with the luciferase reporter gene. (A) Specific localization of the fusions. (B) Expression profiles of Ppbp2B_1 (⧫), Ppbp2B_2 (▴), and Ppbp2B_3 (▪) transcriptional fusions. Bacteria were grown in BHI medium, and at different time points, samples were withdrawn for the determination of the OD₆₂₀ and bioluminescence. Relative luciferase units (RLU) are plotted against the OD₆₂₀. Data from a representative experiment are shown as the means of triplicate values, with error bars indicating standard deviations.

FIG. 3.

Alignment of putative promoter sequences of S. gordonii and other streptococci. S. gordonii promoter sequences were deduced from 5′-RACE (pbp1A, pbp1B, pbp2A, pbp2X, and pgm), primer extension (luxS, arcA, cshA, and scaC), and transcriptional luc fusion (pbp2B) analyses (1, 2, 9, 16, 21). Residues identical to the consensus of the −35 and the extended −10 regions, indicated above the alignment, are highlighted with black boxes. Experimentally determined transcriptional start points are underlined. Sg, S. gordonii; Spn, S. pneumoniae; Sm, S. mitis; Spyo; S. pyogenes.

FIG. 4.

Expression profiles of the pbp genes and of the control arc gene transcriptional fusions in the susceptible wild-type S. gordonii. Cultures were grown under standard conditions (—) or in the presence of subinhibitory concentrations of penicillin G. Penicillin was added at an OD₆₂₀ of 0.12, and concentrations corresponded to 1/2× the MIC (- -) or 1/8× the MIC (- - -). The expression of arc (•), pbp1A (▵), pbp1B (⧫), pbp2A (▴), pbp2b (□), and pbp2x (▪) was monitored as described in the legend to Fig. 2B. The expression of all genes is shown in the same graph and expressed in relative luciferase units (RLU) (general pattern) for the sake of global comparison. In all other graphs, data are expressed as a percentage of the value for the nontreated control. The maximal expression of each gene is considered 100%. Data are then expressed as the percentage of the expression of the corresponding gene under standard conditions.

FIG. 5.

Expression profiles of the wild-type pbp genes and of the control arc gene transcriptional fusions in presence of suprainhibitory concentrations of penicillin G. Cultures were grown under standard conditions (—) or in the presence of penicillin at 2× the MIC (- - -) or 8× the MIC (- -). The expression of arc (•), pbp1A (▵), pbp1A (⧫) pbp2A (▴), pbp2b (□), and pbp2x (▪) was monitored as described in the legend to Fig. 2B. Data are expressed as a percentage of the nontreated control, as described in the legend to Fig. 4. RLU, relative luciferase units.

FIG. 6.

Expression profiles of the penicillin-resistant PR1_2evolved pbp gene transcriptional fusions. Cultures were grown under standard conditions (—) or in the presence of subinhibitory concentrations of penicillin G. Penicillin was added at an OD of 0.12, and concentrations corresponded to 1/2× the MIC (- -) or 1/8× the MIC (- - -). The expression of pbp1A (▵), pbp1B (⧫), pbp2A (▴), pbp2b (□), and pbp2x (▪) was monitored as described in the legend to Fig. 2B. Data are expressed as described in the legend to Fig. 4. RLU, relative luciferase units.