Two-Component System Cross-Talk in the Anthrax Pathogen
By: Carol Rouzer, VICB Communications
Published: April 4, 2014
A newly discovered two-component system (TCS) in B. anthracis that responds to cell envelope stress exhibits unusual cross-regulation with the heme sensor system TCS
Two-component systems (TCSs) provide bacteria with an important mechanism by which they sense and respond to their environment. Typically, a TCS comprises a membrane-bound receptor that, upon ligand binding, autophosphorylates at a histidine residue. This receptor histidine kinase (HK), then transfers the phosphate group to an aspartic acid residue on a response regulator (RR) protein. The phosphorylated RR binds to the promoter region of one or more target genes, regulating transcription of those genes. Most bacteria have 20 to 30 TCSs encoded in their genomes, although some may have up to ten times that many. In general, TCSs are noted for their high degree of specificity, which occurs at the level of the ligand-HK interaction, the interaction of the HK with the RR, and the interaction of the RR with target gene promoters. Indeed, there are few examples of cross-talk between these systems. Thus, the discovery of cross-regulation between two Bacillus anthracis TCSs by Vanderbilt Institute of Chemical Biology (VICB) members Eric Skaar and Gary Sulikowski and their laboratories reveals a potential for adaptive flexibility in bacteria that we have, thus far, not fully appreciated [L. A. Mike et al. (2014) PLoS Pathogens, 10, e1004044].
Known primarily as the cause of anthrax, B. anthracis (Figure 1) is a Gram positive, spore-forming pathogen that is widely distributed in soil. In a mammalian host, B. anthracis spores germinate, and the bacteria thrive. Their rapid multiplication in the bloodstream suggests the ability to efficiently adapt to a changing environment. One important adaptive mechanism is the use of heme as a source of nutrient iron. Heme is readily available through the lysis of red blood cells; however, at high concentrations, heme can also be toxic. Protection from heme toxicity is provided by the heme sensing system (HssRS) TCS. Binding of heme to HssS, the HK of the TCS, leads to activation of the RR (HssR), which induces expression of the target gene hrtAB. The product of this target gene is the heme-regulated transporter (HrtAB) heme efflux pump, which removes excess heme from the cell.
Figure 1. Photomicrograph of Bacillus anthracis bacteria. Image reproduced from the Centers for Disease Control Public Health Image Library, public domain.
Long interested in bacterial heme metabolism, the Skaar lab had previously conducted a high-throughput screen of the VICB compound library to identify molecules that modulate the hrtAB promoter (Phrt) in Staphylococcus aureus. A result of this screen and follow-up medicinal chemistry in the Sulikowski lab was the discovery of VU0120205 (‘205, Figure 2), an activator of the S. aureus Phrt. The Skaar lab used a plasmid (phrt) encoding the Phrt promoter fused to the xylE gene to test ‘205 for activity in B. anthracis. Activation of phrt expression results in production of the XylE catachol dioxygenase, which converts catechol to the intensely yellow hydroxymuconic semialdehyde, providing a convenient colorimetric assay for Phrt regulation. This approach revealed that ‘205 did not activate phrt expression in wild-type B. anthracis, due to a high background level of expression. However, ‘205 was highly effective in mutant bacteria (ΔhssRS), lacking functional genes for HssR and HssS. These surprising results suggested that ‘205 regulates Phrt expression by a mechanism that did not involve the HssRS TCS. In contrast, the xylE reporter plasmid, phrtDR- (Figure 3) failed to respond to ‘205. This plasmid carried mutations of four conserved nucleotides in the direct repeat (DR) region of Phrt that were known to be required for HssR-dependent gene activation. Thus, the investigators concluded that ‘205 must activate a TCS other than HssRS and that this putative TCS must also regulate Phrt through HssR’s DR binding site.
Figure 2. Structure of VU0120205. Image reproduced under the Creative Commons Attribution License from L. A. Mike et al. (2014) PLoS Pathogens, 10, e1004044.
Figure 3. Diagrammatic representation of the genomic loci of hitPQRS (i) and hssRS-hrtAB (ii). The inset shows the sequence of the direct repeat regions of the Phrt and Phit promoters. The four nucleotides required for HssR-dependent activation of Phrt are in bold. The underlined nucleotides are those that are different between the two promoters. Image reproduced under the Creative Commons Attribution License from L. A. Mike et al. (2014) PLoS Pathogens, 10, e1004044.
The search for a TCS that mediates the effects of ‘205 was facilitated by prior reports of a second promoter in the B. anthracis genome with Phrt-like structure. The DR sequence of this new promoter contained the same pattern of four nucleotides that characterize the Phrt DR, but differed from it by four nucleotides (Figure 3). It was found upstream of four other genes which appeared to encode a TCS and an ABC transporter. The Skaar lab investigators named the new TCS and the transporter the HssRS interfacing TCS (hitRS) and the HssRS interfacing transporter (hitPQ), respectively. They quickly noted that the RR of HitRS (HitR) shared 52.5% sequence identity with HssR, and the HK of HitRS (HitS) shared 40.7% identity with HssS. This high degree of structural similarity suggested the possibility of cross-talk between the two TCSs.
Using deletion mutants, the Skaar lab quickly determined that hitRS is required for activation of Phrt by ‘205. They also showed that ‘205 activates the promoter of hitPQ (Phit) through the action of HitRS. Phit activation required the presence of the four conserved nucleotides that were also required for activation of Phrt. In vitro studies using purified proteins confirmed HitS to be an HK that transfers a phosphate group to HitR. Furthermore, HssS was able to phosphorylate HitR, and HitS could phosphorylate HssR, verifying the potential for cross-talk between the two TCSs. Phosphate transfer activities, in order of decreasing efficiency were: HitS to HitR, HssS to HssR, HitS to HssR, HssS to HitR.
To test the hypothesis that cross-regulation occurs between HssRS and HitRS in vivo, the Skaar lab constructed B. anthracis mutants lacking the genes for HssS and HitR (ΔhssSΔhitR). They predicted that these bacteria should be nonresponsive to heme and could only respond to ‘205 through cross-talk between HitS and HssR. Treatment of the mutant bacteria with ‘205, however, resulted in no activation of Phrt as indicated by the xylE reporter assay. Similarly, they also constructed mutants lacking the genes for HitS and HssR (ΔhitSΔhssR). These bacteria should be nonresponsive to ‘205, and could only respond to heme through cross-talk between HssS and HitR. Indeed, exposure of the ΔhitSΔhssR strain to heme led to activation of Phit. These results confirmed cross-regulation between HssS and HitR, but not between HitS and HssR in vivo (Figure 4).
Figure 4. Diagrammatic representation of the cross-talk that occurs between HssRS and HitRS. (A) Heme activates HssS, which autophosphorylates and then transfers the phosphate group to HssR (B) or HitR (I). HssR and HitR can then activate both Phit (H and F, respectively) and Phrt (C and G, respectively). (D) ‘205 activates HitS, which autophosphorylates and then transfers the phosphate group to HitR (E). HitR can then activate both Phit (F) and Phrt (G). Phosphorylation of HssR by HitS (J) has been observed in vitro, but not in vivo, so the arrow there is shown as a dashed line. Image reproduced under the Creative Commons Attribution License from L. A. Mike et al. (2014) PLoS Pathogens, 10, e1004044.
To search for cross-regulation between HitR and Phrt, the investigators created mutant strains of B. anthracis lacking the gene for HssR (ΔhssR). They hypothesized that ‘205 could only activate Phrt in these bacteria through the action of HitR. Indeed, these bacteria responded to ‘205 exposure with activation of Phrt in the xylE reporter assay. In contrast, a ΔhssRΔhitR mutant failed to respond to ‘205, confirming the need for HitR in this response. Similarly, mutant bacteria lacking the gene for HitR (ΔhitR) were able to respond to heme with activation of Phit, confirming an interaction between HssR and Phit. The failure of the ΔhssRΔhitR mutant to respond demonstrated the need for HssR in this heme-dependent promoter activation. These results revealed cross-regulation between the HssRS and HitRS TCSs at the RR-to-target gene promoter interaction level (Figure 4).
The discovery of the HitRS TCS led the investigators to ponder the natural function of this system. In an attempt to gain insight into this question, they conducted a high-throughput screen of the Biolog Phenotypic Microarray Library for Microbial Cells. This library comprises twenty 96-well plates, each containing a compound known to affect microbial growth and behavior. They grew B. anthracis carrying the phit reporter plasmid in the presence of each of these compounds and then evaluated activation of the Phit promoter. The results yielded twelve Phit-activating compounds, of which four were subsequently confirmed to act through the HitRS TCS. These compounds included nordihydroguaiaretic acid (NDGA), an antioxidant; chlorpromozine, an antipsychotic drug; low concentration vancomycin, an antibiotic; and high concentration sodium phosphate. The investigators noted that chlorpromazine has been shown to inhibit cell wall synthesis in Bacillus megaterium, and low dose vancomycin disrupts bacterial cell membrane integrity. These considerations led them to hypothesize that HitRS responds to cell envelope stress. They obtained further support for this hypothesis by demonstrating that targocil, an inhibitor of cell wall teichoic acid synthesis in S. aureus, activates Phit through HitRS.
The studies of cross-regulation between HssRS and HitRS had all been carried out using the xylE reporter assay system. To verify that cross regulation also occurred at the level of the bacterial genome, the investigators exposed B. anthracis to heme, ‘205, NDGA, chlorpromazine, or targocil. In each case, they observed increased expression of both hrtA and hitP, demonstrating that these selective activators of either HssRS or HitRS were all capable of acting on both the Phrt and Phit promoters.
This elegant application of chemical biology to the study of bacterial signaling has revealed a previously uncharacterized TCS and demonstrated an unusual cross-regulation between two TCSs, one coordinating the response to heme, and the other the response to envelope stress. Further work will shed additional light into the functional role of HitRS and the interesting cross-regulation that occurs in the adaptation of B. anthracis to these two environmental stressors.