The c-di-GMP network of Borrelia burgdorferi, a causative agent of Lyme disease, consists of Rrp1, a diguanylate cyclase/response regulator; Hpk1, a histidine kinase; PdeA and PdeB, c-di-GMP phosphodiesterases; and PlzA, a PilZ domain c-di-GMP receptor. Borrelia hermsii, a causative agent of tick-borne relapsing fever, possesses a putative c-di-GMP regulatory network that is uncharacterized. While B. burgdorferi requires c-di-GMP to survive within ticks, the associated effector mechanisms are poorly defined. Using site-directed mutagenesis, size exclusion chromatography, isothermal titration calorimetry and fluorescence resonance energy transfer, we investigate the interaction of c-di-GMP with the Borrelia PilZ domain-containing Plz proteins: B. burgdorferi PlzA and B. hermsii PlzC. The Plz proteins were determined to be monomeric in their apo and holo forms and to bind c-di-GMP with high affinity with a 1:1 stoichiometry. C-di-GMP binding induced structural rearrangements in PlzA and PlzC. C-di-GMP binding proved to be dependent on positive charge at R of the PilZ domain motif, RxxxR. Comparative sequence analyses led to the identification of Borrelia consensus sequences for the PilZ domain signature motifs. This study provides insight into c-di-GMP:Plz receptor interaction and identifies a possible switch mechanism that may regulate Plz protein effector functions.

SUMMARY Orientia tsutsugamushi causes scrub typhus, a potentially fatal infection that afflicts one million people annually. This obligate intracellular bacterium boasts one of the largest microbial arsenals of ankyrin repeat-containing protein (Ank) effectors, most of which target the endoplasmic reticulum (ER) by undefined mechanisms. Ank9 is the only one proven to function during infection. Here, we demonstrate that Ank9 bears a motif that mimics the GRIP domain of eukaryotic golgins and is necessary and sufficient for its Golgi localization. Ank9 reaches the ER exclusively by retrograde trafficking from the Golgi. Consistent with this observation, it binds COPB2, a host protein that mediates Golgi-to-ER transport. Ank9 destabilizes the Golgi and ER in a Golgi-localization domain (GLD)-dependent manner and induces the ATF4-dependent unfolded protein response. The Golgi is also destabilized in cells infected with O. tsutsugamushi or treated with COPB2 siRNA. COPB2 reduction and/or the cellular events that it invokes, such as Golgi destabilization, benefit Orientia replication. Thus, Ank9/bacterial negative modulation of COPB2 might contribute to the bacterium’s intracellular replication. This report identifies a novel microbial GLD, links Ank9 to the ability of O. tsutsugamushi to perturb Golgi structure, and describes the first mechanism by which any Orientia effector targets the secretory pathway.

Anaplasma marginale causes bovine anaplasmosis, a debilitating and potentially fatal tick-borne infection of cattle. Because A. marginale is an obligate intracellular organism, its adhesins that mediate entry into host cells are essential for survival. Here, we demonstrate that A. marginale outer membrane protein A (AmOmpA; AM854) contributes to the invasion of mammalian and tick host cells. AmOmpA exhibits predicted structural homology to OmpA of A. phagocytophilum (ApOmpA), an adhesin that uses key lysine and glycine residues to interact with ␣2,3-sialylated and ␣1,3-fucosylated glycan receptors, including 6-sulfo-sialyl Lewis x (6-sulfo-sLe x ). Antisera against AmOmpA or its predicted binding domain inhibits A. marginale infection of host cells. Residues G55 and K58 are contributory, and K59 is essential for recombinant AmOmpA to bind to host cells. Enzymatic removal of ␣2,3-sialic acid and ␣1,3-fucose residues from host cell surfaces makes them less supportive of AmOmpA binding. AmOmpA is both an adhesin and an invasin, as coating inert beads with it confers adhesiveness and invasiveness. Recombinant forms of AmOmpA and ApOmpA competitively antagonize A. marginale infection of host cells, but a monoclonal antibody against 6-sulfo-sLe x fails to inhibit AmOmpA adhesion and A. marginale infection. Thus, the two OmpA proteins bind related but structurally distinct receptors. This study provides a detailed understanding of AmOmpA function, identifies its essential residues that can be targeted by blocking antibody to reduce infection, and determines that it binds to one or more ␣2,3-sialylated and ␣1,3-fucosylated glycan receptors that are unique from those targeted by ApOmpA.

Cyclic-di-GMP (c-di-GMP) contributes to the regulation of processes required by the Lyme disease (LD) spirochetes to complete the tick-mammal enzootic cycle. Our understanding of the effector mechanisms of c-di-GMP in the Borrelia is evolving. While most LD spirochete isolates encode a single PilZ domain containing c-di-GMP receptor designated as PlzA, genome analyses have revealed that a subset encode a second PilZ domain protein (PlzB). The c-di-GMP binding potential of PlzB, and its role in LD spirochete biology, have not been investigated. To determine if PlzB binds c-di-GMP, plzB from B. burgdorferi isolate ZS7 was PCR amplified, cloned, and recombinant protein generated. PlzB bound c-di-GMP but not other nucleotides, indicating a specific binding interaction. To determine if PlzA and PlzB are functionally synonymous, a series of allelic-exchange gene deletion and cis-complemented strains were generated in the B. burgdorferi B31 background. B. burgdorferi B31-ΔplzA was competent to infect Ixodes scapularis larvae but not mice when delivered by either needle or tick feeding. B. burgdorferi B31-ΔplzA also displayed an atypical motility phenotype. Complementation in cis of B. burgdorferi B31-ΔplzA with plzA (B31-plzA KI) restored wild-type (wt) phenotype. However, a strain complemented in cis with plzB (B31-plzB KI) did not. The data presented here are consistent with an earlier study that demonstrated that PlzA plays an essential role in spirochete survival in the mammalian environment. We add to our understanding of the c-di-GMP regulatory network by demonstrating that while PlzB binds c-di-GMP, it is not functionally synonymous with PlzA. The absence of plzB from most strains suggests that it is not required for survival. One possibility is that cells that harbor both PlzA and PlzB might have enhanced biological fitness or increased virulence.

No abstract