Discovery at the VICB
A Host Defense Protein Encourages Bacterial Cooperation
By: Carol A. Rouzer, VICB Communications
Published: July 8, 2016
The protein calprotectin fights bacterial infections by sequestering nutrient metals, but new evidence shows it also promotes co-infections by multiple pathogens.
Many species of bacteria respond to environmental stresses by aggregating on a surface to form a biofilm. Regardless of whether a biofilm contains a single species or many different species of bacteria, the microorganisms adapt physiologically to the growth conditions that are present in their region of the biofilm. This specialization promotes the survival of the entire community and enables pathogenic bacteria to evade host defense mechanisms. It also contributes to antibiotic resistance of some bacterial species. A key host defense mechanism against pathogenic bacteria is suppression of bacterial growth by the sequestration of nutrients - a process known as nutritional immunity. For example, binding of iron by transferrin or binding of zinc or manganese by calprotectin sequesters these key nutrient metal ions from invading bacteria. This led Vanderbilt Institute of Chemical Biology members Eric Skaar, Richard Caprioli, and Walter Chazin to hypothesize that nutrient deficiencies caused by nutritional immunity could play an important role in determining biofilm dynamics. They now present support for this hypothesis and show that calprotectin-mediated nutritional immunity modulates the ability of two species of bacteria, Pseudomonas aeruginosa and Staphylococcus aureus, to form co-cultures in infected human lung [C. A. Wakeman, J. L. Moore, et al., Nat. Commun, 7, 11951.]
The drip-flow reactor (DFR) is a simple apparatus for reliable biofilm formation in the laboratory (Figure 1a). A constant flow of nutrient-rich medium drips gently onto the upper end of a glass slide that rests in an incubation chamber at a 10o incline. The medium flows over the slide and is then collected as waste at the lower end. Bacteria (P. aeruginosa in these experiments) inoculated on the slide respond to these conditions by forming a biofilm. Visible inspection reveals differences in the bacteria growing in different regions of the biofilm. A pore at the site where the medium drip hits the slide leads into a channel that allows the medium to pass through the biofilm. Bacteria at the pore and along the channel are pink in color, whereas those closer to the edges are a light green due to production of different pigments by the bacteria, an indication of location-dependent specialization (Figure 1b). The medium flow establishes a nutrient gradient such that relative nutrient depletion increases as distance from the pore and central channel increases. Despite this gradient, the ratio of live to dead cells is fairly uniform across the biofilm.
FIGURE 1. (1) A drip flow reactor (DFR) consists of a slanted incubation chamber that contains a glass slide inoculated with bacteria. Fresh culture medium drips slowly onto the upper end of the slide and flows down the slide to the bottom where it is collected in a waste container. As the bacteria grow, they form a biofilm on the glass slide. (b) Picture of a biofilm obtained from a DFR. Highlighted are the pore that forms at the site where the medium dripped onto the slide, the central channel that formed around the medium as it flowed down the slide, and the edges of the biofilm that are relatively nutrient deficient. Figure reproduced under a Creative Commons Attribution 4.0 International License from C. A. Wakeman, J. L. Moore, et al., Nat. Commun, 7, 11951.
The investigators generated biofilms using a DFR and then used MALDI-IMS (matrix-assisted laser desorption/ionization-imaging mass spectrometry) to investigate the effects of location within the biofilm on bacterial protein expression. For this technique, they obtained thin sections from various regions of the biofilm and impregnated them with a light-absorbing matrix material. Then, they systematically focused a laser beam on different regions of the section to facilitate desorption and ionization of molecules from the surface. Capture of these molecules for mass spectral analysis yielded data about the molecular composition of each region of the section. The results (Figure 2) showed that protein expression varied depending on the region of the biofilm, with some proteins concentrated most around the nutrient pore and central channel and others more prevalent around the edges. These findings suggested that relative nutrient abundance likely affected expression of some proteins in the biofilm. Further analysis revealed the identities of the most consistently expressed proteins. In general, proteins associated with basic metabolic function tended to be uniformly expressed across the biofilm. Some proteins of unknown function were differentially expressed. Of particular interest, however, were a number of small ribosomal proteins that exhibited modulation of expression in response to Zn2+ concentration. The differential expression of these proteins across the biofilm suggested the possibility of metal ion-dependent regulation.
FIGURE 2. (a) Picture of a biofilm created in a DFR. The colored arrows indicate the flow of nutrients. Colors indicate the nutrient gradient from high (blue) to low (brown). The rectangles show the positions of sections of the biofilm that were analyzed by MALDI-IMS. (b and c) MALDI-IMS signals for ions with m/z 4,557 and 5,829 were found mostly in nutrient rich areas near the pore and central channel. (d and e) MALDI-IMS signals for ions with m/z 6.297 and 1,415 were found mostly in nutrient poor areas near the edges of the biofilm. (f) Color-coded composite of all four signals indicating their relative abundance at the four different regions of the biofilm. Figure reproduced under a Creative Commons Attribution 4.0 International License from C. A. Wakeman, J. L. Moore, et al., Nat. Commun, 7, 11951.
To better understand the role of nutrient metal availability on biofilm protein expression, the investigators conducted LA-ICP-IMS (laser ablation inductively coupled plasma imaging mass spectrometry) analyses on sections of the biofilm. Like MALDI-IMS, LA-ICP-IMS analyzes thin sections of a sample to determine the two-dimensional distribution of species of interest - in this case, metal ions. The results of this study demonstrated that magnesium ions were distributed evenly across the biofilm, whereas iron and calcium ions were concentrated more towards the edges, and manganese and zinc were found predominantly in the regions of the central canal and nutrient pore.
With a better understanding of the distribution of nutrient metals across the biofilm, the investigators sought a more detailed picture of protein expression. To obtain this, they dissected a biofilm, obtaining samples from the nutrient rich central regions and the nutrient poor edges. They homogenized these samples and subjected them to a comprehensive proteomics analysis. Overall, they found that the identities of the proteins expressed were similar throughout the biofilm, but their quantities varied. Of particular interest were the proteins encoded by the hcn, phz, and pqs genes, which are responsible for production of hydrogen cyanide, pyocyanin, and alkyl hydroxiquinolones (AQs) respectively. These are antibacterial molecules that P. aeruginosa produces to suppress the growth of competing bacteria. The proteomics analysis revealed that expression of the hcn, phz, and pqs gene products was reduced in the nutrient-poor regions of the biofilm. The antibacterial genes in P. aeruginosa are regulated by the quorum sensing system - a signaling pathway that enables bacteria to communicate with each other in response to environmental stress. However, expression of other quorum sensing-regulated proteins was not suppressed in nutrient-poor regions of the biofilm, suggesting that this was not the mechanism of regulation for the antibiotic genes. Similarly, the antibiotic genes are regulated in response to iron levels, but their pattern of expression was dissimilar to those of other iron-regulated proteins. This led the investigators to hypothesize that other metal ions might regulate antibacterial gene expression.
To test their hypothesis, the investigators added the Zn2+ and Mn2+-binding protein calprotectin to the medium in a DFR apparatus. MALDI-IMS analysis of the biofilms that developed under these conditions demonstrated that calprotectin treatment resulted in an alteration in expression levels of some proteins, including those that synthesize antibacterial factors. However, the MALDI-IMS data also indicated that calprotectin did not reach all regions of the biofilm. Therefore, the investigators carried out further studies using P. aeruginosa cultured in petri dishes. In these experiments, they used RNA-seq and quantitative RT-PCR to assess the level of expression of RNA coding for proteins of interest. They confirmed that calprotectin suppressed the expression of the hcn, phz, and pqs genes. Furthermore, MALDI-IMS demonstrated that levels of pyocyanin and two AQs produced by the bacteria progressively decreased in the presence of increasing concentrations of calprotectin.
The results suggested that sequestration of Zn2+ and/or Mn2+ by calprotectin suppresses the production of factors known to inhibit the growth of other bacteria. To determine if this effect alters the ability of P. aeruginosa to compete with S. aureus in co-culture, they placed single colonies of P. aeruginosa onto petri dishes containing a uniform lawn of S. aureus bacteria. As the P. aeruginosa cultures grew, a clear zone formed around them due to suppression of S. aureus growth. However, addition of increasing amounts of calprotectin blocked formation of the inhibitory zone in a concentration-dependent fashion (Figure 3). This did not occur if the investigators added a mutant form of calprotectin that could not bind metals. Furthermore, the effects of calprotectin were reversed by the addition of a 2-fold molar concentration of Zn2+, but not other metal ions, including Mn2+. These results confirmed that calprotectin's ability to block the synthesis of antimicrobial compounds depended on its sequestration of Zn2+ and that both of its two Zn2+ binding sites were involved.
FIGURE 3. When a culture of P. aeruginosa is inoculated onto a uniform lawn of S. aureus bacteria, an area of S. aureus growth inhibition forms around the P. aeruginosa culture. However, addition of calprotectin (CP) results in a concentration-dependent reduction in the size of the growth inhibitory zone. Figure reproduced under a Creative Commons Attribution 4.0 International License from C. A. Wakeman, J. L. Moore, et al., Nat. Commun, 7, 11951.
To determine if calprotectin affects the interaction between P. aeruginosa and S. aureus in vivo, the investigators infected wild-type mice with equal mixtures of both bacteria intranasally. These mice developed lung infections in which both bacteria were present in about equal concentrations. However, when they carried out the same experiment in mice genetically deficient in calprotectin, P. aeruginosa predominated in the infected lungs. To see if this relationship also occurred in humans, they examined an explant of human lung tissue from a cystic fibrosis patient. P. aeruginosa pneumonia is very common in these patients, and this bacterium was present in the explant. However, S. aureus was also present, and was found predominantly in regions where a strong inflammatory response had occurred. Calprotectin is a major protein in neutrophils, which are among the first cells to reach sites of inflammation. MALDI-IMS analysis confirmed the presence of calprotectin in the inflammatory sites of the infected cystic fibrosis lung explant.
The results support the investigators' hypothesis that mechanisms of nutritional immunity, such as metal sequestration by calprotectin, can influence the metabolic adaptation of bacteria in a biofilm. However, the data go beyond the original intent to show that the response of P. aeruginosa to calprotectin directly affects the ability of the bacteria to establish co-cultures with S. aureus, and likely other bacteria as well. The authors suggest that, P. aeruginosa may benefit from the presence of S. aureus when responding to stresses exerted by host defense mechanisms. The ways in which bacteria may synergize under those conditions are only now beginning to be explored. We look forward to the next discovery on the role of nutritional immunity in multibacterial biofilm formation and function.
View Nature Communications article: "The innate immune protein calprotectin promotes Pseudomonas aeruginosa and Staphylococcus aureus interaction"