The Legionella longbeachae Icm/Dot substrate SidC binds to the LCV through PtdIns(4)P and facilitates the interaction with the ER

The genus Legionella consists of environmental bacteria which are
the causative agents of the severe pneumonia Legionnaires’ disease.
L. longbeachae and L. pneumophila are able to replicate
intracellularly in human alveolar macrophages and aquatic or soil
amoebae. In order to replicate within host cells the bacteria
establish a compartment derived from the endoplasmatic reticulum
(ER) which is called “Legionella-containing vacuole” (LCV). A
bacterial intracellular multiplication/defective in organelle
transport (Icm/Dot) type IV secretion system (T4SS) is essential
for the formation of this LCV. The Icm/Dot T4SS enables
translocation of effector proteins into the host cell. More than
100 effector proteins are presumably translocated during an L.
longbeachae infection whereas around 300 translocated effector
proteins are known for L. pneumophila. During maturation the LCV
communicates with vesicles from the endocytic vesicle trafficking
pathway, avoids fusion with lysosomes and instead fuses with the
ER. Phosphoinositides (PI) such as
phosphatitdylinositol-4-phosphate (PtdIns(4)P) are enriched on the
LCV which mediate the binding of Icm/Dot translocated effector
proteins like SidCLpn (substrate of Icm/Dot transporter) as well as
its paralogous protein SdcALpn. The 73 kDa effector SidM but not
the 106 kDa SidCLpn was found in a previous phosphoinositide
pulldown assay with L. pneumophila lysate to be the major
PtdIns(4)P binding protein. Using L. longbeachae lysate we showed
binding of the 111 kDa SidCLlo to PtdIns(4)P in a phosphoinositide
pulldown. This result was confirmed by protein-lipid overlay assays
using “PIP-strips”. In further analysis the P4C (PtdIns(4)P-binding
of SidC) domain was identified as a 19 kDa domain of SidCLlo
located in the amino acid region 609 to 782. This P4C domain was
located in the same region as the 20 kDa SidCLpn_P4C domain of L.
pneumophila. Both P4C domains can be used as LCV markers. This was
shown with GST-tagged proteins binding to LCVs in a cell
homogenate. The two P4C domains show a sequence identity of only
45% and the full-length protein of 40%. Circular dichroism
measurements revealed that the secondary structure of the two
proteins is similar. Moreover, isothermal titration calorimetric
measurements indicated a 3.4 higher affinity of SidCLlo towards
PtdIns(4)P compared with SidCLpn. In RAW 264.7 macrophages infected
with L. longbeachae we showed that endogenous SidCLlo as well as
heterologously produced SidCLpn is translocated to the LCV in an
Icm/Dot-dependent manner. The deletion of the sidCLlo gene led to a
reduced recruitment of calnexin to the LCV in infected
Dictyostelium discoideum. This effect was complemented by adding
plasmid-encoded SidCLlo, SidCLpn or SdcALpn. The same recruitment
defect for a L. pneumophila strain lacking the sidCLpn and sdcALpn
genes was complemented by the production of SidCLlo and SidCLpn as
published before. Therefore, these effectors play a role for
pathogen-host interactions by promoting the recruitment of ER to
the LCV. L. longbeachae or L. pneumophila wild-type strains
outcompeted their sidC deletion mutant in a competition assay in
Acanthamoeba castellanii. However neither of the deletion mutants
were impaired in their growth in single strain replication
experiments. In summary despite of the small sequence identity and
the higher binding affinity to PtdIns(4)P of SidCLlo compared to
SidCLpn both effector proteins seem to have similar functions
during an infection of Legionella. For the characterization of L.
longbeachae-containing vacuoles through proteomic analysis, LCVs
had to be isolated from infected D. discoideum or RAW 264.7
macrophages. Endogenous SidCLlo or heterologously produced SidCLpn
were used as LCV markers for the isolation. Pathogen vacuoles
harbouring L. longbeachae were isolated by immuno-affinity
purification using antibodies specifically recognizing SidCLlo or
SidCLpn. Future investigations aim at optimizing the LCV
purification protocol for L. longbeachae to determine the proteome
composition of the L. longbeachae-containing vacuole.