Turning up Francisella pathogenesis

The ability of microbes to adapt in response to an environmental change, such as the infection of a warm-blooded host is an important mechanism in pathogenesis. One major mechanism by which microbes respond to an environmental change is through the modification or remodeling of their membranes. Lipid A, or endotoxin, the bioactive component of lipopolysaccharide (LPS) is the major component of the outer leaflet of the outer membrane of Gram-negative bacteria. Lipid A, the membrane anchor of LPS is a β-(1′-6)-linked glucosamine disaccharide backbone with amide linked fatty acids at the 2 and 2′ positions, ester linked fatty acids at the 3 and 3′ positions and terminal phosphate moieties at the 1 and 4′ positions. Early steps in lipid A synthesis are conserved across Gram-negative species (Raetz pathway) though species-specific structural modifications are also observed. Modification of lipid A in response to environmental stressors such as pH, cation concentration, oxygen saturation and osmolarity have been shown in a number of bacterial species to contribute to condition-specific membrane phenotypes. Temperature has also been shown to play an important role in lipid A structure supported by the presence of specific genes for growth at high temperatures in Escherichia coli (htrB/lpxM) and temperature dependent hypoacylation in Yersinia pestis. The ability of bacterial species to alter or modify their lipid A in response to specific environmental cues is important for pathogenesis.

In our recent manuscript, we demonstrated the importance of temperature as an environmental cue in outer membrane remodeling of the intracellular pathogen, Francisella novicida. Francisella has been identified in many environmental niches and a wide array of hosts and vectors including humans, lagomorphs, arthropods and amoebae. Throughout its complex and varied life cycle, Francisella adapts to a range of temperatures that alter the composition of its lipid A. Specifically, after growth at lower temperatures normally encountered in arthropods or water environments (~25°C), Francisella synthesizes a lipid A with predominately shorter fatty acids (16 carbons in length). At higher temperatures, like those encountered in warm-blooded hosts (37°C), lipid A is composed primarily of longer chain fatty acids (18 carbons in length). Recent work from our lab determined this shift in lipid A fatty acid composition was regulated at the genetic and enzymatic level utilizing two N-linked acyltransferases, LpxD1 and LpxD2 (Fig. 1).

Figure 1. Overview of lpxD control of pathogenesis. Francisella has a unique ability to adapt upon entering various environmental niches. Of particular interest is the activity of the acyltransferases LpxD1 and LpxD2. Upon entry of warm temperatures Francisella upregulates lpxD1 and this causes a larger lipid A structure, leading to decreased permeability as well as decreased sensitivity to killing by CAMPs. Conversely at cooler temperatures lpxD2 is upregulated, generating a smaller lipid A giving rise to increased permeability and sensitivity to CAMPs. Mutants of lpxD1 locked in the cooler state become avirulent while mutants of lpxD2 locked in the warm state remain virulent.

lpxD is conserved in most Gram-negative organisms and is present in an operon containing the essential genes fabZ, lpxB and lpxA. LpxD adds N-linked fatty acids to the glucosamine backbone at the “lipid X” stage present on the inner leaflet of the inner membrane and prior to MsbB transport to the outer leaflet of the inner membrane. Activity of LpxD has been shown to be required for bacterial viability. Interestingly, sequence comparisons of lpxD1 and lpxD2 from Francisella show low amino acid sequence identity (34%) suggesting distinct genetic origins, as opposed to gene duplication. lpxD1 is similar to the lpxDs found in enteric bacteria, whereas lpxD2 is more closely related to lpxDs from anaerobic bacteria, such as Desulfovibrio and Fusobacterium. The close relationship to the anaerobic species is striking, as Francisella is a strict aerobe. lpxD1 and lpxD2 are highly conserved across all Francisella subspecies (> 98%) indicating a critical role in remodeling the bacterial membrane. Also, the duplication of lpxD in the relatively small genome of Francisella (1,782 genes) highlights a potentially critical role for both genes in bacterial virulence.

Regulation of both lpxD1 and lpxD2 is tightly controlled in Francisella. Microarray data showed that lpxD1 was upregulated at 37°C, in contrast to lpxD2 which was upregulated at 25°C. Not only was the expression of these genes temperature regulated, but their abilities to catalyze acylation was also linked to temperature. The LpxD1 enzyme catalyzed the addition of longer chain acyl groups at 37°C, while the LpxD2 enzyme added shorter chain acyl groups at 25°C. Genetic deletion of these genes allowed further insight into their role in pathogenesis, leading to distinct phenotypes. Deletion of lpxD1 resulted in a smaller lipid A structure similar to that seen after growth of wild-type Francisella at lower temperatures, whereas deletion of lpxD2 “locked” the lipid A into a larger configuration similar to wild-type samples grown at warm temperatures.

These mutants, having different lipid A phenotypes and lacking the ability to alter their membrane composition, were tested in a variety of assays to determine their role in virulence. Francisella is lethal to mice at extremely low doses (LD₁₀₀ ~10 CFU) regardless of infection route. To assess if the overall lethality of the mutants was affected, mice were infected with increasing amounts of each mutant, as well as WT Francisella. The ΔlpxD2 mutant maintained full virulence at amounts similar to that of the WT strain. Interestingly, the ΔlpxD1 mutant, which is locked in the smaller, cold phenotype, showed a total lack of virulence even at the highest doses tested (5 × 10⁶ CFU, 500,000 × LD₁₀₀). As the ΔlpxD1 mutant was completely attenuated after a single primary infection dose, we sought to determine if the ΔlpxD1 mutant could induce protective immunity against a lethal wild-type challenge. Our results showed that protection was dependent on the primary infection dose. When mice were vaccinated using a prime-boost model they were completely protected, regardless of the immunizing dose.

To determine if specific membrane alterations were responsible for attenuation in the lpxD mutant strains, a variety of lipid A/membrane functions were examined. As is seen in lipid A isolated from WT strains of Francisella, neither mutant produced a lipid A that resulted in NFκB-mediated proinflammatory responses through TLR4, suggesting that the attenuation was not due to increased recognition by the host innate immune sensors. Another function of lipid A is to resist killing by host peptides by modulating both the charge and permeability of the outer membrane. The ΔlpxD2 mutant, with the larger lipid A was shown to be more resistant to cationic antimicrobial peptides (CAMPs) presumably due to decreased permeability in the outer membrane when compared with WT Francisella. Decreased permeability was confirmed by measuring the uptake of ethidium bromide into cells. Conversely, the ΔlpxD1 mutant showed increased permeability and sensitivity to CAMPs.

Temperature regulation of virulence factors is common in many bacterial species. Generally, temperature regulates the production of classical virulence factors such as toxins in C. difficile, or master regulators of virulence, such as prfA in L. monocytogenes. Here we have shown that a single, temperature regulated, non-classical virulence gene is important for pathogenesis of Francisella. The inability to alter its lipid A and thus remodel its membrane upon infection of warm-blooded hosts is an important mechanism in the overall pathogenesis of Francisella. Though the loss of lpxD2 showed no effect on murine pathogenesis, it may be required for growth in the environment or arthropod vectors. These questions are currently being pursued in alternative models.

The importance of maintaining proper membrane function and organization upon adaptation to varied environmental conditions is essential. In Francisella, the tight control of lipid A biosynthetic enzymes occurring at multiple levels; transcriptional and enzymatic mark a critical role in pathogenesis. Controlled regulation of the lpxD1 gene potentially works as a “virulence switch” and alters not only the membrane permeability and resistance to host defenses but plays a critical role in turning on or off pathogenesis. The consequences of this seemingly simple acyl-chain alteration, addition of two carbons to lipid A remain to be explored. Additionally modulation of lipid A structure has a profound impact on the protein constituency in the outer membrane, suggesting a role for specific outer-membrane proteins to be present or absent during pathogenesis. This new cadre of membrane proteins may have far-reaching effects on signaling both from extracellular stimuli and other partners in the membrane. These initial observations in Francisella invite examination of the temperature controls placed on lipid A in other bacterial species that are found at multiple temperatures and how these may affect virulence.

Li Y, Powell DA, Shaffer SA, Rasko DA, Pelletier MR, Leszyk JD, et al. LPS remodeling is an evolved survival strategy for bacteria. Proc Natl Acad Sci U S A. 2012;109:8716–21. doi: 10.1073/pnas.1202908109.