Figure 1.
In silico characteristics of rickettsial patatins (Pat1 and Pat2).
A) Schema depicting the major features of R. typhi Pat1 (RT0590, YP_067537) and Pat2 (RT0522, YP_067473) proteins. The conserved Glycine-rich motif (G), Serine hydrolase motif (S) and active site Aspartate (D) of patatins are depicted for both proteins, with a variable region (V) and conserved C-terminal tail (T) illustrated for Pat1. Neither protein was predicted to contain an N-terminal Sec-dependent secretion signal (see text). B) Comparative analysis of Pat1 and Pat2 proteins across 46 Rickettsia genomes. Phylogeny at left is based on whole genome analysis [73], with rickettsial groups as follows: red, ancestral group; aquamarine, typhus group; blue, transitional group, brown, spotted fever group [26], [74]. R. helvetica, previously categorized as spotted fever group rickettsiae, is shown as incertae sedis. Plasmids pRF (R. felis) and pMCE_1 (“Candidatus R. amblyommii”), which encode Pat1 homologs, are also listed (but not included in the phylogeny estimation). The sequences within the conserved G, S and D regions for both Pat1 and Pat2 are shown. For Pat1, the length of the V region is provided, followed by number of predicted repeats [66]. The group designation (A or B) for each Pat1 sequence is provided based on phylogeny estimation (see Fig. 2A). For Pat1, sequences highlighted gray (R. bellii str. RML369-C and R. peacockii str. Rustic) depict the proteins comprised of two ORFs split near the C-terminus (see Fig. 2B). For Pat2, black bars depict pseudogenes, which comprise short annotated ORFs with strong homology to full length Pat2 proteins or smaller fragments identified using TBLASTN. Accession numbers for all sequences are provided in Table S1.
Figure 2.
Evolutionary analysis of rickettsial Pat1 proteins.
A) Phylogeny estimation of rickettsial Pat1 proteins, with closest non-Rickettsia clades shown. The complete tree containing 93 ExoU-like patatins is shown in Fig S2. See Materials and Methods for details of phylogenetic analysis. The Rickettsia Pat1 sequences (n = 33) form two well-supported clades (Pat1A and Pat1B). The Pat1B group (1–10, shaded blue) contains the plasmid-encoded Pat1 proteins (colored red), Pat1 of the R. canadensis strains, and the proteins from REIS, R. helvetica (incertae sedis) and the transitional group rickettsiae. The Pat1A group (11–33, shaded blue) contains R. typhi RT0590 (noted with a black star), the R. prowazekii and R. bellii proteins, and Pat1 from the majority of the spotted fever group rickettsiae. B) Comparative analysis of the C-terminal tails for 33 rickettsial Pat1 proteins. See text for alignment details. The sequences are listed (1–33) as they appear in the phylogenetic tree in panel A. The horizontal bar distinguishes Pat1B (upper) from Pat1A (lower) sequences. The tan bar above the alignment illustrates the delineation between the tail region and the variable region of Pat1 sequences. Coordinates for each sequence are provided at left and right. Conserved positions across the alignment are highlighted yellow, with positions predicted to evolve under positive selection [71] highlighted red and noted at the top of the alignment by asterisks. Positions depicted with dots and highlighted gray illustrate missing sequence for Pat1 proteins encoded by two ORFs (R. bellii str. RML369-C and R. peacockii str. Rustic).
Figure 3.
Translocation of Pat1 during R. typhi infection of Vero76 cells by Western blotting.
R. typhi infected or uninfected Vero76 grown for 48 hr were fractionated by 0.1% Triton X-100 treatment into pellet containing intact rickettsia with host cell debris and supernatant containing rickettsial secreted protein with host soluble proteins and was probed with rabbit anti-Pat1 antibody, rabbit anti-Pat2 antibody (as positive control for R. typhi secreted protein), rabbit anti-rOmpB antibody (as control for R. typhi surface protein), rabbit anti-EF-Ts antibody (as control for R. typhi cytoplasmic protein) or mouse anti-GAPDH monoclonal antibody (as control for host cytoplasmic protein). Lane1: pellet of uninfected Vero76; Lane2: supernatant of uninfected Vero76; Lane3: pellet of R. typhi infected Vero76; Lane4: supernatant of R. typhi infected Vero76. The size of the expected protein bands is shown on the right.
Figure 4.
Secretion of Pat1 and Pat2 into host cells.
Cells were fixed with 4% paraformaldehyde at 24 h postinfection and immunolabeled with anti-Pat1 (panels A to C for R. typhi infected and panel D for uninfected) or anti-Pat2 (panels F to H for R. typhi infected and panel I for uninfected) rabbit antibodies (at 1∶200 dilution) and anti-R. typhi rat serum (at 1∶500 dilution) as primary antibodies. R. typhi infected Vero76 cells were also labeled similarly with rabbit pre-immune serum (at 1∶200 dilution) and anti-R. typhi rat serum (at 1∶500 dilution) shown in panel E. The anti-rat-Alexa Fluor-594 (red) and anti-rabbit-Alexa Fluor-488 (green) antibodies were used as secondary antibodies. The cell nuclei were stained with DAPI (blue). Samples were viewed under a LSM5DUO confocal microscope and images were processed using ZEN imaging and analysis software. The white arrows showed punctate structure indicating translocation of Pat1 or Pat2 from rickettsiae into host cell cytoplasm. Scale Bar = 5 µm.
Figure 5.
Cytotoxicity in yeast strain INVSc1 by CFU assay.
The yeast strain INVSc1 transformed with the plasmids carrying genome sequence of pat1 (Pat1-wt), carrying codon optimized sequence of pat1 (Pat1-co) or carrying mutation at catalytic sites of pat1-co (Pat1-SD). A) The percentage of CFU in transformed yeast cells was determined as described (see Materials and Methods). B) Western blot analysis of the expression of Pat1 and its mutant derivatives in INVSc1 under inducing condition (SC-U+Gal medium). The total proteins from yeast cells carrying the appropriate plasmid were probed with anti-Pat1 antibody using WesternBreeze chemiluminescent immunodetection kit (Invitrogen). LaneM: MagicMark XP Western Protein Standard (Invitrogen), Lane1: Pat1-SD/INVSc1, Lane2: Pat1-co/INVSc1, Lane3: Pat1-wt/INVSc1. The size of the expected recombinant protein (including C-terminal V5 epitope and 6× His tag) is mentioned on the left (59.46 kDa). The bands below the expected protein size may have resulted from degradation of recombinant protein or nonspecific binding to proteins from yeast cell lysate.
Figure 6.
Phospholipase A activity assay of recombinant proteins.
The phospholipase A (PLA) activity was determined via calculation of fluorescent emission/µg recombinant protein/min (see Materials and Methods). Error bars represent standard errors of the means. A) The PLA activity of recombinant proteins Pat2 or Pat1 increased significantly (P<0.05[two-tail t-test]) in presence of Vero76 cell lysate compared to that in absence of Vero76 cell lysate. The PLA activity of Pat1 was found to be significantly different (P<0.05[two-tail t-test]) from that of mutant Pat1-SD or in presence of 0.5 µM MAPF (PLA2 inhibitor). B) PLA activity assay of recombinant proteins: Pat2 and Pat1 in presence of bovine liver superoxide dismutase (SOD) as described in materials and methods.
Figure 7.
Effect of anti-Pat1 and anti-Pat2 pretreatment on R. typhi infection by plaque assay.
Rickettsiae were treated with affinity purified anti-Pat1, anti-Pat2 or pre-immune IgG (PI) for 30 min on ice, followed by incubation for infection into Vero76 for 1 hour at 34°C and 5% CO2. R. typhi infected cells were left unwashed (A) or washed (B) before addition of agar for plaque assay as described (see Materials and Methods). The effect of antibody treatment on R. typhi was determined in percent (%) plaque formation by antibody treated rickettsiae with respect to that by no treatment. Error bars represent standard errors of the means. In both Panel A and B, percent (%) plaque formation by antibody pretreated R. typhi were found to be significantly different (P<0.05[two-tail t-test]) from that by PI.
Figure 8.
Effect of anti-Pat1 and anti-Pat2 antibody pretreatment on R. typhi infectivity of Vero76 cells by IFA.
Rickettsiae were treated with 2 µg of affinity purified anti-Pat1, anti-Pat2 or pre-immune IgG (PI) for 30 min on ice, followed by incubation for infection of Vero76 for 18 hour at 34°C and 5% CO2. The infected cells were labeled for IFA and % infectivity was determined as described (see Materials and Methods). Infectivity (%) was determined in % host cells infected by R. typhi pretreated with anti-Pat1 or anti-Pat2 antibody with respect to that with PI. Error bars represent standard errors of the means. Infectivity (%) of anti-Pat1 or anti-Pat2 antibody pretreated R. typhi was found to be significantly different (P<0.05[two-tail t-test]) from that of PI.
Figure 9.
Effect of anti-Pat1 and anti-Pat2 pretreatment on R. typhi phagosome escape by IFA.
Rickettsiae were incubated with 2 µg of affinity purified anti-Pat1, anti-Pat2 or pre-immune IgG (PI) for 30 min on ice. Pretreated rickettsiae were added onto Vero76 monolayer followed by incubation at 34°C and 5% CO2. At 30 min postinfection, the cells were fixed with 4% paraformaldehyde and immunolabeled with anti-LAMP-1 (at 1∶100 dilution) mouse monoclonal antibodies (Abcam Inc) and anti-R. typhi rat serum (at 1∶500 dilution) as primary antibodies. The anti-rat-Alexa Fluor-488 (green) and anti-mouse-Alexa Fluor-594 (red) were used as secondary antibodies. The cell nuclei were stained with DAPI (blue). Samples were viewed under a LSM5DUO confocal microscope and images were processed using ZEN imaging and analysis software. (A) Representative micrograph of anti-Pat1 (panels a to c), anti-Pat2 (panels d to f) or pre-immune IgG (panels g to i) treated rickettsiae that were infected of Vero76 cells. The white arrows show rickettsiae in LAMP-1 positive phagosome, with the square head arrow showing rickettsiae escaping or escaped from the phagosome. Scale Bar = 5 µm. (B) Quantitation of percent rickettsiae in LAMP-1 positive phagosome was determined as percent LAMP-1 positive R. typhi by scoring 100 bacteria for each treatment. Each treatment was repeated four times. Error bars represent standard error of the means. Percent LAMP-1 positive R.typhi for anti-Pat1 or anti-Pat2 pretreatment was found to be significantly different (P<0.05[two-tail t-test]) from that of PI.
Table 1.
Primers used in this study.