The obligate intracellular bacterium Coxiella burnetii causes the zoonotic disease Q-fever. Coxiella pathogenesis depends on a functional type IV secretion system (T4SS). The T4SS effector AnkG inhibits pathogen-induced host cell apoptosis, which is believed to be important for the establishment of a persistent infection. However, the mode of action of AnkG is not fully understood. We have previously demonstrated that binding of AnkG to p32 is crucial for migration of AnkG into the nucleus and that nuclear localization of AnkG is essential for its anti-apoptotic activity. Here, we compared the activity of AnkG from the C. burnetii strains Nine Mile and Dugway. Although there is only a single amino acid exchange at residue 11, we observed a difference in anti-apoptotic activity and nuclear migration. Mutation of amino acid 11 to glutamic acid, threonine or valine results in AnkG mutants that had lost the anti-apoptotic activity and the ability to migrate into the nucleus. We identified Importin-α1 to bind to AnkG, but not to the mutants and concluded that binding of AnkG to p32 and Importin-α1 is essential for migration into the nucleus. Also during Coxiella infection binding of AnkG to p32 and Importin-α1 is crucial for nuclear localization of AnkG.

The ability to inhibit host cell apoptosis is important for the intracellular replication of the obligate intracellular pathogen Coxiella burnetii, as it allows the completion of the lengthy bacterial replication cycle. Effector proteins injected into the host cell by the C. burnetii type IVB secretion system (T4BSS) are required for the inhibition of host cell apoptosis. AnkG is one of these anti-apoptotic effector proteins. The inhibitory effect of AnkG requires its nuclear localization, which depends on p32-dependent intracellular trafficking and importin-α1-mediated nuclear entry of AnkG. Here, we compared the sequences of ankG from 37 C. burnetii isolates and classified them in three groups based on the predicted protein size. The comparison of the three different groups allowed us to identify the first 28 amino acids as essential and sufficient for the anti-apoptotic activity of AnkG. Importantly, only the full-length protein from the first group is a bona fide effector protein injected into host cells during infection and has anti-apoptotic activity. Finally, using the Galleria mellonella infection model, we observed that AnkG from the first group has the ability to attenuate pathology during in vivo infection, as it allows survival of the larvae despite bacterial replication.