Francisella tularensis is a highly infectious Gram-negative bacterium that causes tularemia, a zoonosis affecting humans and animals. The genus Francisella comprises several species including F. tularensis, F. philomiragia, F. noatunensis, F. hispaniensis and F. novicida (Kingry and Petersen, 2014; Sjodin et al., 2012). F. tularensis is divided into two main subspecies: Type A (F. tularensis subsp. tularensis [F. tularensis]) and Type B (F. tularensis subsp. holarctica [F. holarctica]). Type A is considerably more virulent than Type B and is commonly associated with severe disease in humans, whereas Type B typically causes milder illness. Pneumonic tularemia, the most dangerous form, is transmitted through aerosols or inhalation. Remarkably, inhaling as few as 25 colony-forming units (CFUs) may result in pulmonary infection, making F. tularensis one of the most infectious bacteria transmitted via the aerosol route (Harrell et al., 2024). Francisella is a significant pathogen associated with high morbidity and mortality rates. The most virulent strain of Francisella induces systemic forms of tularemia (pneumonic, typhoidal), with fatality rates reaching 60% (Yeni et al., 2021; Wawszczak et al., 2022). Due to its high infectivity and potential for airborne spread, F. tularensis has been developed and stockpiled as a biological warfare agent by several nations (Dennis et al., 2001).

F. novicida is classified as a separate species from F. tularensis, but due to the high genomic homology, it is sometimes classified as a subspecies of F. tularensis (Johansson et al., 2010). Because Francisella novicida (F. novicida) is considerably less virulent than F. tularensis, it is commonly used as a surrogate strain for virulent F. tularensis. In contrast to F. tularensis, F. novicida is an opportunistic pathogen that causes illness and even death in debilitated or immunocompromised patients but not in healthy individuals. Due to the rarity of F. novicida infections, effective detection is challenging (Brett et al., 2012; Kingry and Petersen, 2014). F. novicida shares many characteristics with Type A strains of F. tularensis, showing similarities in the genome sequence, intracellular life cycle, and mechanisms of infectivity, including rapid phagosomal escape followed by vigorous cytosolic replication. It also infects macrophages, causing disease in mice. However, due to its lower virulence in humans, it can be safely handled under Biosafety Level 2 laboratory conditions (Kingry and Petersen, 2014; Marecic et al., 2017). F. novicida strain U112 possesses a potent tool in the form of a specific transposon mutant, which has been used to investigate the intracellular life cycle of Francisella and elucidate its virulence mechanisms (Weiss et al., 2007; Gallagher et al., 2007).

Typically, nonpathogenic bacteria taken up by host cells are enclosed in vacuoles or phagosomes, which mature through the endocytic pathway and ultimately fuse with lysosomes, leading to bacterial degradation. Intracellular pathogens have evolved strategies to evade this process within phagocytic cells and resist phagosome–lysosome fusion (Ray et al., 2009). In particular, Francisella species exhibit a complex intracellular life cycle and notable environmental persistence (Rowe and Huntley, 2015). Their survival and replication rely heavily on escaping the initial phagosomal compartment and replicating within the host cell cytosol, making Francisella a model cytosolic pathogen. The ability to persist and multiply inside phagocytes and other host cells is fundamental to its pathogenicity. Disruptions in key mechanisms, such as receptor engagement, resistance to reactive oxygen species (ROS), phagosomal escape, cytosolic replication, and evasion of innate immune recognition, significantly impair F. tularensis survival and reduce its virulence in both in vitro and in vivo models (Celli and Zahrt, 2013).

wbtF is assigned as a potential NAD-dependent epimerase and UDP-glucose 4 epimerase (UGE) (Thomas et al., 2007). It shows sequence similarity with established UGEs, including wbpP from Pseudomonas aeruginosa (Prior et al., 2003). wbpP is necessary for producing UDP-N-acetyl-D-galacturonic acid, a precursor to the galactosaminuronic acid-derived components of the O-antigen (Miller et al., 2008). UGE, part of the short-chain dehydrogenase/reductase enzyme superfamily, catalyzes the final step in the interconversion of UDP-glucose and UDP-galactose during galactose metabolism in bacteria and mammals (Fushinobu, 2021). Thomas et al. (2007) identified wbtF genes as a virulence determinant of F. tularensis, but its function remains unknown.

In this study, we created a transposon mutant library of F. novicida and screened for infection using epithelial cells, HeLa cells expressing the FcγRII (HeLa-FcγRII cells) (Nakamura et al., 2024). We focused our investigation on wbtF and analyzed that the wbtF was associated with the intracellular replication of F. novicida.

Francisella is a Gram-negative bacterium, and its highly pathogenic species, F. tularensis, is a well-known cause of tularemia in humans and other mammals. F. tularensis outmaneuvers the host’s immune defenses by escaping from the phagosome—a cellular structure meant to destroy invaders. Instead of being degraded, the bacteria multiply inside the host cell’s cytoplasm, where they are better protected from immune responses (Pavlik et al., 2024). Scientists are still piecing together the mechanism of phagosomal escape, and a gene cluster called the Francisella pathogenicity island (FPI) has emerged as a key player. The FPI encodes component proteins of a unique type VI secretion system thought to be critical to the bacterium’s survival and virulence (Clemens et al., 2015; Rigard et al., 2016; Spidlova and Stulik, 2017).

We constructed a transposon mutant library of F. novicida to isolate mutants that were less cytotoxic. We screened the transposon mutants using a new infection model of Francisella, HeLa cells expressing mouse FcγRII (HeLa-FcγRII). HeLa-FcγRII cells are a cell line that expresses the FcγRII receptor, enabling efficient uptake of IgG-opsonized bacteria (Nakamura et al., 2024). Although epithelial cells, like HeLa, are readily amenable to genetic modification, infection with Francisella is rarely effective and significantly lower than the infection efficiency of macrophage-like cells. Francisella infection of epithelial cell lines, such as like HeLa, A549, and HEK293, has been performed at a high MOI (100–10,000) and requires a relatively long time (approximately 8–24 h) (Lo et al., 2013; Amer et al., 2010; Bradburne et al., 2013). The ability of Francisella to infect macrophages largely depends on whether the bacteria are opsonized. The use of unopsonized bacteria leads to rapid disruption of the phagosome. In contrast, opsonizing Francisella with fresh serum facilitates directing the bacteria to various phagocytic receptors, resulting in delayed phagosomal escape (Geier and Celli, 2011). Notably, when HeLa cells engineered to express FcγRII were infected with antibody-coated bacteria, the infection efficiency was approximately 100 times higher compared with normal HeLa cells (Nakamura et al., 2024).

Among the 2,232 transposons mutants library, 13 showed less cytotoxicity and were successfully sequenced. Two mutants involved the same gene: FTN_0496 (slt). FTN_1686 (figE), FTN_0496 (slt), FTN_1322 (IglC), FTN_1321 (IglD), FTN_1309 (pdpA), FTN_1310 (pdpB), FTN_1317 (IglI), FTN_1313 (IglF), and FTN_0096 were previously reported as pathogenic factors of Francisella (Llewellyn et al., 2011; Ozanic et al., 2015; Brodmann et al., 2017; Pérard et al., 2018; Nakamura et al., 2019; Dean et al., 2020). FTN_0756 (fopA) and FTN_1641 (ampG) were previously reported as immunosuppressive factors involved in dampening host immune responses (Nakamura et al., 2022). However, this was the first report of FTN_1425 (wbtF). Therefore, we focused on wbtF. Although wbtF has been reported as a non-essential gene that can be disrupted (Thomas et al., 2007), we were unable remove the gene using a suicide vector (Nakamura et al., 2019) in this study. Therefore, we attempted to directly investigate wbtF gene from a transposon mutant library of F. novicida.

wbtF is one of the O-antigen biosynthetic gene clusters of F. tularensis (Clemens et al., 2012). The O-antigen capsule in Francisella comprises a polysaccharide that is structurally identical to the O-antigen found in lipopolysaccharide (LPS). This LPS O-antigen is a critical pathogenic factor. O-antigen deficiency results in increased sensitivity to serum and a loss of virulence in Type A and B strains of F. tularensis (Catanzaro and Inzana, 2020). Thomas et al. (2007) created knock-out mutants targeting the O-antigen gene cluster in F. novicida and F. tularensis SchuS4, leading to the complete loss of O-antigen production. Notably, the SchuS4 mutant lacking the O-antigen was fully attenuated in mice.

Francisella species are highly pathogenic, largely due to their ability to replicate within the cytosol of phagocytic cells, such as macrophages and dendritic cells. After entering these cells through phagocytosis, the bacteria are initially enclosed within a membrane-bound compartment called the Francisella-containing phagosome. However, they rapidly disrupt the phagosomal membrane and escape into the host cell cytosol, where they replicate. The bacteria adapt their metabolism to the cytosolic environment by relying on gluconeogenesis and amino acid catabolism as carbon and energy sources. Thus, phagosomal escape is critical for Francisella’s ability to replicate inside host cells, leading to disease in vivo. Phagosomal escape also triggers the host’s antimicrobial and innate immune defenses. In many cases, once the infected cell becomes overloaded with bacteria and dies, the bacteria exit and potentially spread to neighboring cells (Degabriel et al., 2023; Chong and Celli, 2010).

In this research, intracellular replication of the F. novicida WT strain in HeLa-FcγRII cells and THP-1 cells increased at 12–48 h after infection. This correlated with previous findings of increased intracellular replication of Francisella WT in HeLa-FcγRII and THP-1 cells (Nakamura et al., 2021, 2024). Compared with the WT strain, the wbtF mutant underwent intracellular replication for up to 12 h after infection in HeLa-FcγRII and THP-1 cells, but then decreased at 24 and 48 h after infection. The rate of intracellular replication of the wbtF mutant was significantly lower than that of the WT strain (P <0.01). Thomas et al. (2007) demonstrated that the O-antigen of F. tularensis was essential for the survival and replication inside host cells.

The acidic organelles were stained with LysoTracker. The wbtF mutant bacteria showed colocalization with LysoTracker in the early stages of infection (2–6 h) of HeLa-FcγRII cells (60%–71%) and THP-1 cells (72%–74%). Furthermore, the wbtF mutant also colocalized with the lysosome marker LAMP-1 at 2, 6, 12, and 48 h postinfection in 78%–88% of HeLa-FcγRII cells and 71%–82% of THP-1 cells.

F. tularensis virulence is closely associated with its ability to survive inside phagocytic cells and escape into the cytosol. Upon host cell entry, the bacterium is initially enclosed in a phagosome that typically matures into a phagolysosome. However, F. tularensis disrupts this maturation by blocking fusion between the phagosome and lysosome. Then, it breaks down the phagosomal membrane and escapes into the cytoplasm, where it replicates. The molecular mechanism underlying the phagosomal escape of Francisella remains poorly understood. However, it has been shown that disrupting genes within the FPI prevents the bacterium from phagosomal escape (Spidlova and Stulik, 2017). Thus, our findings suggest that the wbtF mutant could not escape from the phagosomes, indicating that wbtF is required for intracellular replication of F. novicida.

Mitochondria are double-membraned organelles vital for various cellular processes and cell survival. Their metabolic functions are crucial for maintaining tissue balance across different cell types and organisms. Unsurprisingly, these metabolic pathways often shift in response to stressors, like nutrient deprivation. Beyond their role in energy production, mitochondria are also highly involved in regulating cell death during infection and have been increasingly recognized as important hubs for immune signaling (Spinelli and Haigis, 2018; Mills et al., 2017). We observed that the F. novicida WT strain associates with mitochondria in HeLa–FcγRII cells and THP-1 cells, and underwent intracellular replication. Furthermore, it did not colocalize with mitochondria (MitoTracker) at 2–6 h after infection. In contrast, the wbtF mutant colocalized with the mitochondria during the initial stages of infection (2–6 h) in HeLa-FcγRII cells (61%–76%) and THP-1 cells (69%–78%).

Mitochondria are highly susceptible to damage, and their dysfunction disrupts cellular homeostasis. Such disturbances are closely linked to the development of various diseases, including neurodegenerative and cardiovascular conditions, cancer, metabolic disorders, and infections. Thus, the impaired mitochondria must be promptly isolated and selectively eliminated. Mitophagy (mitochondrial autophagy) is a critical cellular process that eliminates damaged mitochondria, thereby preserving mitochondrial and overall cellular homeostasis. When cells are exposed to stressors such as ROS, nutrient scarcity, or aging, mitochondria become depolarized and dysfunctional, triggering their selective removal through mitophagy (Doblado et al., 2021; Onishi et al., 2021; Xu et al., 2020; Lu et al., 2023).

Pathogens stimulate mitophagy to suppress the host defense mechanism. They likely employ a common mechanism that inhibits innate immune responses and promotes persistent infection by directly or indirectly facilitating mitophagy (Ma et al., 2024). The initiation of mitophagy to eradicate intracellular mitochondrial ROS (mtROS) is an essential survival tactic for intracellular infections. mtROS are essential for innate host defense, functioning as effectors by harming pathogens (Yang et al., 2025; Silwal et al., 2020). Alqahtani et al. (2023) showed that infection with F. tularensis live vaccine strain induced mitophagy. Generally, mitophagy is considered to limit the accumulation of mtROS, a significant source of cellular ROS. Microbial pathogens often evade or suppress the host’s protective mtROS pathway and neutralize bactericidal free radicals (Su et al., 2023; Silwal et al., 2020). Other intracellular bacteria, such as Staphylococcus aureus, promote mitophagy to clear mtROS. S. aureus not only survives within macrophages but also stimulates mitophagy in host cells to clear mtROS. It uses HDAC11 to augment interleukin 10 (IL-10) transcription, promoting mitophagy to clear mtROS, reduce mitochondrial damage, and maintain an ecological environment favorable for bacterial survival. Listeria monocytogenes diminishes mtROS and inhibits cell death by promoting mitophagy through the secretion of its virulence factor, listeriolysin O (Yang et al., 2025; Zhang et al., 2019). These findings suggest that the intracellular survival of Francisella WT depends on mitochondrial modulation and escape. Indeed, the effector protein IglJ has been reported to interact directly with mitochondrial targets (Prokšová et al., 2020). In contrast, the wbtF mutant cannot modulate mitochondria and is therefore detected in association with mitochondria. The wbtF gene might be involved in this kind of mitochondrial modulation.

The Golgi complex is a crucial membrane-bound organelle in the perinuclear region that consists of flattened disk-shaped cisternae connected in a ribbon-like structure. It comprises three distinct subcompartments: the cis-Golgi, medial-Golgi, and trans-Golgi network (Ahat et al., 2019). The Golgi complex is the modification and sorting hub for proteins and lipids. Newly synthesized proteins and lipids are delivered to the Golgi complex, where they undergo posttranslational modifications and are sorted into vesicles for transport to other organelles or extracellular secretion (Makhoul et al., 2019).

To explore the activities of the F. novicida WT and wbtF mutant strains within the Golgi complex, infected HeLa-FcγRII and THP-1 cells were stained with BODIPY TR ceramide. The Francisella WT strain only showed colocalization with the Golgi complex in 2%–5% of HeLa-FcγRII cells and 4%–12% of THP-1 cells. However, the wbtF mutant was colocalized with the Golgi complex in 57%–75% of HeLa-FcγRII cells and 61%–75% of THP-1 cells.

The Golgi complex is a crucial organelle in the host defense systems against intracellular infections. Additionally, recent research indicates that the Golgi complex plays a critical role in immunological signaling (Tao et al., 2020). Disruption of the structure of the Golgi complex catalyzes the pathogenesis of infectious diseases. Various intracellular bacterial, viral, and protozoan infections trigger Golgi complex destabilization. This evolutionarily conserved strategy enables pathogens to access nutrients, hijack organelle membranes for replication, and subvert host immune defenses (Read et al., 2024). Type I interferons (IFNs) are induced by the Golgi complex in response to various stimuli, including Francisella infection (Gregory et al., 2020). Type I IFNs are a group of cytokines crucial for shaping innate and adaptive immune responses during viral and bacterial infections. However, they also suppress antibacterial immunity by promoting IL-10 production, inhibiting proinflammatory cytokines and chemokines, and promoting immune cell apoptosis, thereby exacerbating Francisella infections (Xia et al., 2025; Li et al., 2024). The translocation of the signaling intermediate stimulator of IFN genes (STING, TMEM173) from the endoplasmic reticulum to the cis-Golgi is initiated by cyclic GMP-AMP synthase or interferon-γ inducible protein 16 following the identification of cytosolic exogenous DNA (Storek et al., 2015; Meunier et al., 2015; Man et al., 2015). STING activates TANK-binding kinase 1 (TBK1) at the Golgi complex by inducing its autophosphorylation at serine 172. Following activation, TBK1 phosphorylates interferon regulatory factor 3, facilitating dimerization and translocation to the nucleus, which drives the expression of immune-related genes, including type I IFNs (Barber, 2011). Interestingly, structural disruption of the Golgi complex using golgicide A blocks TBK1 phosphorylation in response to F. tularensis infection and LPS (Gregory et al., 2020). These results suggest that F. novicida WT escapes from the Golgi complex, but wbtF, which was detected in the Golgi complex, cannot escape from the immune response.

Since wbtF is a gene involved in O-antigen synthesis, we investigated the cytokine production of the wbtF deletion mutant to confirm whether the attenuated cytotoxicity and intracellular proliferation of the mutant were due to the LPS-mediated activation of immune responses. Infection of THP-1 cells with F. novicida WT, wbtF mutant, and the complemented strain produced similar TNF-α and IL-6 responses, indicating that wbtF does not significantly affect cytokine production. The wbtF gene, a consecutive gene located within the O-antigen-biosynthetic gene clusters spanning a 17-kb chromosomal region of F. tularensis, is directly implicated in LPS biosynthesis. The O-antigen constitutes one of the three components of LPS, along with the lipid A and the oligosaccharide core (Clemens et al., 2012; Krzyżewska-Dudek et al., 2024). Francisella LPS does not function as an antagonist of TLR4-mediated innate immune signaling in either humans or mice. Francisella LPS is tetraacylated, and it elicits only weak activation of TLR4 and demonstrates a markedly reduced ability to induce the secretion of inflammatory cytokines like TNF-α and IL-6 by murine and human innate immune cells (Hajjar et al., 2006; Gunn and Ernst, 2007; Noah et al., 2010). It seems that Francisella LPS represents, at most, a minor contributor to the induction of proinflammatory responses within the innate immune system. Although deletion of LPS synthesis genes, such as kdsA, have been reported to enhance cytokine production (Nakamura et al., 2022), the absence of wbtF did not impact cytokine production in this study. Taken together, these results imply that the attenuated cytotoxicity and intracellular growth ability observed in wbtF deletion mutant are likely due to mechanisms distinct from LPS-mediated immune activation. Although comprehensive understanding of the detailed mechanisms necessitates further investigation, our results may suggest that wbtF is important for modulating LPS structure, potentially facilitating the evasion of recognition and uptake by cellular organelles, including phagosomes, mitochondria, and the Golgi apparatus.

In summary, our research focused on identifying a virulence factor of F. novicida by screening a transposon mutant library. We used a new infection model of Francisella using HeLa cells expressing mouse FcγRII (HeLa-FcγRII) (Nakamura et al., 2024). We identified wbtF as a pathogenic factor of F. novicida. wbtF is required for escaping from organelles such as phagosomes, mitochondria, and the Golgi apparatus, followed by intracellular replication in HeLa-FcγRII and THP-1 cells. Thus, it may be a potential target for controlling Francisella infection. However, further research is necessary to elucidate the detailed mechanism of wbtF function in Francisella infection.