Listeria infection modulates mitochondrial dynamics

Abstract

Mitochondria are highly dynamic organelles that are central to several cellular processes, the most prominent being energy production. Several reports have shown that pathogens target mitochondria in various ways to interfere with apoptosis, but to our knowledge only one study has specifically addressed the effects of infection on mitochondrial dynamics. We focused on this aspect during infection with the intracellular pathogen L. monocytogenes and could recently show that this bacterium profoundly alters mitochondrial dynamics, causing transient fragmentation of the mitochondrial network. This mitochondrial fragmentation occurs early during infection and is specific to pathogenic L. monocytogenes, as it is not observed with other intracellular pathogens. The relevance of mitochondrial dynamics for L. monocytogenes infection is highlighted by the finding that siRNA-mediated inhibition of mitochondrial fusion or fission decreases or increases the efficiency of L. monocytogenes infection, respectively. The main bacterial factor responsible for mitochondrial network disruption was identified as the secreted pore-forming toxin listeriolysin O, which also appeared to impair mitochondrial function. Our work suggests that in order to establish an efficient infection, L. monocytogenes interferes with cellular physiology at early timepoints by transient disruption of mitochondrial dynamics and function.

Mitochondria are crucial organelles. They provide most of the cellular ATP, and are also increasingly recognized as signaling platforms in biological processes as diverse as apoptosis^1,2 and innate immune signaling.³ Mitochondria hence represent sites where signaling and energy-producing activity are integrated.⁴ In addition, under different physiological conditions, changes in mitochondrial activity influence mitochondrial morphology and vice versa.⁵ Mitochondrial fission is essential for proper distribution of mitochondria (and thereby of energy supply) within the cell,^6,7 for the elimination of nonfunctional mitochondria,⁸ for mitochondrial inheritance during mitosis⁹ and for embryonic development.^10–12

Due to their key role in cell survival, mitochondria represent attractive targets for pathogens. Interfering with mitochondria can aim at preserving the bacterial replication niche by inhibiting death of the host cell, or at inducing death of immune cells that counteract pathogen dissemination.¹³ Several pathogens, including both viruses and bacteria, have been shown to target mitochondria in order to interfere with the host cell apoptotic machinery (reviewed in ref. 14 and 15). Apoptotic cell death eventually entails fragmentation of the mitochondrial network,¹⁶ and bacteria that induce apoptosis can thus lead to mitochondrial network fragmentation. On the other hand, evidence is accumulating that induction of mitochondrial fragmentation is not sufficient to cause cell death.^17,18

Only few data are available on the effects of intracellular pathogens on host cell mitochondrial dynamics and function under conditions where apoptosis is not induced. This prompted us to specifically analyze the effect on host cell mitochondria upon infection with an invasive bacterial pathogen, Listeria monocytogenes. In our study, we show for the first time that infection with L. monocytogenes alters the morphology of host cell mitochondria, inducing fast and drastic fragmentation of the mitochondrial network. We have identified the secreted bacterial toxin listeriolysin O (LLO) as the main factor causing mitochondrial fragmentation via the induction of calcium fluxes at early time-points of infection. LLO therefore differs from bacterial effectors that are directly targeted to mitochondria and insert into the outer membrane or effectors that are imported into the mitochondrial matrix (reviewed in ref. 15 and 19). The rapid fission kinetics observed in the case of Listeria infection leads us to postulate that local production of small amounts of LLO before or during bacterial entry is sufficient to induce such fragmentation. In this context, LLO affects also bystander cells, therefore acting as a classical diffusible toxin. Interestingly, previous studies have shown that not only inactivation of extracellular LLO with a monoclonal antibody (A4–8) renders wild type Listeria avirulent,²⁰ but also that a non-secreted, surface-associated version of LLO causes a decrease in in vivo infectivity and virulence despite a normal in vitro phenotype.²¹ These data support the notion that several effects of LLO occur well before bacterial entry and depend on its ability to form pores on the plasma membrane.

While we could not detect rapid mitochondrial fragmentation upon infection with several other intracellular bacterial pathogens, e.g., Shigella and Salmonella, recombinant pore-forming toxins of the same family as LLO appeared to have a similar effect as LLO on mitochondria. Some of these pore-forming toxins, e.g., S. pneumoniae pneumolysin (PLY) and S. pyogenes streptolysin O (SLO) have been shown to activate the apoptotic cascade by damaging mitochondria.^22,23 However, as we showed, Listeria induces mitochondrial fragmentation without classical apoptosis, suggesting that LLO action differs from that of pore-forming toxins produced by extracellular bacteria. Interestingly, the only other study explicitly analyzing mitochondrial dynamics upon infection showed that Helicobacter pylori causes cell death-associated mitochondrial fragmentation.²⁴

The molecular mechanism that mediates the LLO-induced mitochondrial fission is still unclear. Due to the rapid kinetics of the process, we speculate that the fission machinery is activated by a calcium-dependent mechanism. We currently cannot exclude that in addition the fusion machinery is blocked due to the LLO-induced decrease in mitochondrial membrane potential.⁵

Our work links mitochondrial dynamics with bacterial infection using the intracellular pathogen Listeria as a model. Two elegant reports have recently linked mitochondrial dynamics to innate immunity in the context of viral infection: mitochondrial morphology, and in particular mitofusins were shown to regulate mitochondrial antiviral signaling protein (MAVS).^25,26 It is possible that bacterial pathogens also take advantage of the connection between innate immunity and mitochondria and Listeria is certainly a prime candidate to address this hypothesis. Interestingly, MAVS signaling was recently shown to depend on the energetic status of mitochondria.²⁷ This result confirms the intuitive notion that cellular bioenergetics is critical for correct functioning of defense mechanisms. Indeed, mitochondrially diseased cells are more susceptible to Legionella infection.²⁸ Our work supports the existence of a connection between bioenergetics and bacterial infection. Concomitant to mitochondrial fragmentation, Listeria infection induces a decrease in the mitochondrial membrane potential and in respiration, which possibly contributes to the bioenergetic crisis that is also reflected by a significant drop in intracellular ATP levels (Fig. 1). We speculate that these processes are beneficial for the initial establishment of infection, particularly because evidence is accumulating that beyond providing energy, metabolites play important signaling functions.²⁹ Detailed analyses are now required to shed light onto which pathways mediate the link between host cell bioenergetics and Listeria invasion.

Figure 1.

Model depicting the effect of Listeria infection on host cells. Early during Listeria infection, secreted LLO inserts into the plasma membrane of host cells (which can also be bystander cells), causing fragmentation of the mitochondrial network along with a decrease in mitochondrial membrane potential and intracellular ATP levels. Cellular signaling pathways that remain to be identified may sense changes in mitochondrial dynamics and bioenergetics to impact on infection efficiency.

Addendum to: Stavru F, Bouillaud F, Sartori A, Ricquier D, Cossart P. Listeria monocytogenes transiently alters mitochondrial dynamics during infection. Proc Natl Acad Sci USA. 2011;108:3612–3617. doi: 10.1073/pnas.1100126.