SEARCH

SEARCH BY CITATION

FilenameFormatSizeDescription
cmi12100-sup-0001-FigS1.tif6188K

Fig. S1. Prediction and comparison of the α-helical coiled-coil of YopD and isogenic site-directed mutants. Different propensities to form dimeric (red line), trimeric (green) and tetrameric (blue) coiled-coil structures for each protein sequence is reported as a graph of probability scores that has been calculated using the method of Lupas et al. (1991) hosted by the Pôle Bio-Informatique Lyonnais web server (http://pbil.univ-lyon1.fr/). The bold black line indicates the C-terminal coiled-coil under focus in this study. With far less confidence, a second coil-coil is predicted for the YopD N-terminus, which has been under investigation elsewhere (Costa et al., 2012).

cmi12100-sup-0002-FigS2.tif290K

Fig. S2. Intrabacterial stability of pre-formed pools of YopD mutants with substituted proline at the C-terminus. Bacteria were first cultured for 1 h in non-inducing (plus 2.5 mM CaCl2) BHI broth at 37°C. The protein synthesis inhibitor chloramphenicol (50 μg ml–1) was added at time point 0 min (min). Samples were then collected at this and subsequent time points. Protein levels associated with pelleted bacteria were detected by Western blot using polyclonal anti-YopD antiserum. Panels: Parent (YopDwt), YPIII/pIB102; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304. All variants displayed a comparable intrabacterial stability. The experiment was repeated two times.

cmi12100-sup-0003-FigS3.tif505K

Fig. S3. Low calcium response growth phenotypes of Y. pseudotuberculosis producing YopD variants deficient in a C-terminal α-helix. Bacteria were grown at 37°C in TMH medium supplemented with 2.5 mM CaCl2 (A) or non-supplemented (without Ca2+; B). Two different growth phenotypes were detected: TS – bacteria are sensitive to elevated temperature regardless of the presence or absence of calcium (YopDΔ4–303) and, CD – calcium-dependent growth (all remaining strains). Strains: Parent (YopDwt), YPIII/pIB102; YopDΔ4–303, (full-length null mutant) YPIII/pIB621; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304. The experiment was repeated two times.

cmi12100-sup-0004-FigS4.tif576K

Fig. S4. Cell contact-dependent feedback inhibition of Yop synthesis by Y. pseudotuberculosis. For each bacterial strain, HeLa cells (+) were seeded in two wells of a six-well plate, while the remaining well lacked any cells. Pre-induced bacteria were added to all three wells. To prevent de novo protein synthesis, chloramphenicol (Cml, +) was added to one of the wells containing both bacteria and HeLa cells. At 90 min post infection, the total well content was recovered in loading buffer and analysed by Western blotting with a polyclonal YopE antibody. As a control for general protein synthesis (i.e. T3SS-independent), we detected the molecular chaperone DnaJ with rabbit polyclonal DnaJ antibodies. We also utilized the ΔyopK null mutant as a control for disruption of the feedback inhibitory loop, which subsequently leads to derepression of Yops synthesis regardless of the interaction with host cells (Aili et al., 2008; Isaksson et al., 2009). The feedback inhibitory mechanism remains intact in our YopD mutants i.e. lower YopE levels in the presence of target cells. The one exception being the full-length ΔyopD null mutant, where Yop synthesis is de-repressed in both in vitro and in vivo conditions (Francis and Wolf-Watz, 1998; Williams and Straley, 1998; Anderson et al., 2002). Strains: Parent (YopDwt), YPIII/pIB102; ΔyopD full-length null mutant (YopDΔ4–303), YPIII/pIB621; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304; ΔyopK null mutant, YPIII/pIB155; ΔyopE null mutant, YPIII/pIB526. The experiment was repeated three times.

cmi12100-sup-0005-FigS5.tif6236K

Fig. S5. Impaired YopE-induced changes in HeLa cell morphology. Strains were allowed to infect a monolayer of growing HeLa cells. At 30, 60 and 120 min post infection, the effect of the bacteria on the HeLa cells was recorded by phase-contrast microscopy. Translocation of the YopE cytotoxin, a GTPase-activating protein, causes a distinct morphological change from oblong to rounded of affected HeLa cells (see A, E and G). HeLa cells not intoxicated with YopE show normal uninfected cell morphology (see B and C; indicated with an arrow). Some YopD variants are dramatically impaired in their ability to alter cell morphology (see D and F; indicated with an arrowhead). Panels: Parent (YopDwt), YPIII/pIB102; YopDΔ4–303, (full-length null mutant) YPIII/pIB621; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304.

cmi12100-sup-0006-FigS6.tif676K

Fig. S6. Quantification of native YopE translocation into eukaryotic cell monolayers.

A. HeLa cells were infected in duplicates with bacteria for 2 h. Proteinase K was added to remove all extracellular proteins followed by specific target cell lysis of one set of HeLa cells using digitonin. Samples were clarified to remove intact bacterial cells and cellular debris and then analysed by Western blotting with a polyclonal YopE antibody and a monoclonal β-Actin antibody. β-Actin serves as a control of cell lysis and equal loading. The difference between digitonin-treated HeLa cells (+) and untreated (−) is considered to be translocated YopE (Nordfelth and Wolf-Watz, 2001). Represented is one experiment that was performed a total of five times. For each experiment, the extent of translocated YopE was quantified by densitometry (Bio-Rad Quantity One software, version 4.6.2) and the calculated mean ± SEM was tabulated. The degree of YopE translocation by mutant bacteria was then compared with parent bacteria. The asterisk (*) indicates instances where YopE translocation was significantly low (non-parametric Mann–Whitney U test, P < 0.05, two-tailed), while ‘ns’ indicates no statistically different translocation levels.

B. To ensure that this detected YopE is not contaminated by protein derived from chemically lysed bacteria cells, we performed a bacterial lysis control in which parental Y. pseudotuberculosis and the yopD full-length null mutant were grown in the presence (+) or absence (−) of proteinase K, digitonin and SDS. Samples were divided into pelleted (P) and trichloroacetic acid-precipitated supernatant (S) pools that were fractionated by SDS-PAGE and analysed by Western blot using antibodies recognizing the cytoplasmic molecular chaperone DnaJ. Y. pseudotuberculosis Lanes: Parent (YopDwt), YPIII/pIB102; YopDΔ4–303, (full-length null mutant) YPIII/pIB621; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304. The predicted molecular weight of YopE, DnaJ and β-Actin is specified in parentheses.

cmi12100-sup-0007-FigS7.tif1211K

Fig. S7. Evaluation of the infected erythrocyte membrane purification strategy.

A. Western blot analysis of samples extracted at different stages in the sheep red blood cell membrane purification protocol from a haemolytic reaction with the parental strain (YPIII/pIB102 + SRBC), non-infected erythrocytes (SRBC alone) and parental bacteria alone (YPIII/pIB102 alone). Lanes: 1, total lysate after RBC lysis (bacteria and lysed RBCs); 2, lysate after low-spin centrifugation (bacteria and cellular debris removed); 3, clarified lysate after high speed centrifugation (crude membrane fraction removed); 4, resuspended crude membranes in 50% sucrose-TBS; 5, 50% sucrose-TBS fraction after 16 h swing-out centrifugation; 6, purified SRBC membrane fraction from the 44–25% sucrose-TBS interface after 16 h swing-out centrifugation. The loading was corrected for the volume changes in the purification protocol (i.e. 4.5 times less of lanes 4, 5 and 6 were loaded).

B. Overexposed Western blot analysis of samples extracted at different stages in the RBC membrane purification protocol from a haemolytic reaction with the parental strain (YPIII/pIB102 + RBC) loaded as in (A) except that the loading was not corrected for volume changes in the purification scheme as indicated by the theoretical fold enrichment below the blots. Arrowheads indicate the detection of two bands with the anti-Y. pseudotuberculosis DnaJ antisera; the upper band corresponds to bacterial DnaJ and the lower band to a soluble RBC protein around 40 kDa as indicated by the presence of this species also in the soluble fractions of the non-infected RBC preparation (SRBC alone). No significant enrichment of any of the examined proteins was seen in the purified membranes except for the native Na+/K+-ATPase α-1 chain present in the RBC membranes. The predicted molecular masses of the proteins are specified in parentheses.

cmi12100-sup-0008-FigS8.tif3987K

Fig. S8. T3S translocon components inserted into target cell membranes. Sheep red blood cells were infected with various isogenic strains of Y. pseudotuberculosis prior to the purification of membraneous material via sucrose gradient ultracentrifugation. Membrane fractions either with (+) or without (−) prior thorough washing with 100 mM NaCO3 (pH 10) were resolved on SDS-PAGE and immunoblotted with monoclonal antiserum recognizing YopB and rabbit polyclonal antiserum recognizing YopD and YopE. Lanes: Parent (YopDwt), YPIII/pIB102; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304; full-length ΔyopD null mutant (YopDΔ4–303), YPIII/pIB621; full-length ΔyopB null mutant (YopBΔ13–399), YPIII/pIB615; double ΔyopB, yopD null mutant, YPIII/pIB619; full-length ΔyopK null mutant, YPIII/pIB155. Below the blots and representing loading controls are Coomassie-stained bands corresponding to band 3 (the erythrocyte membrane anion channel) and band 5 (erythrocyte associated actin) (Fairbanks et al., 1971) from stained samples separated on 12% SDS-PAGE. The predicted molecular masses of relevant proteins are specified in parentheses.

cmi12100-sup-0009-FigS9.tif3389K

Fig. S9. Analysis of secreted Yop complexes under native conditions. Autoradiogram of 35S-labelled secreted Yops resolved on a 4–13% BN-PAGE gradient gel (Wittig et al., 2006) from the supernatants of Y. pseudotuberculosis strains grown under secretion permissive conditions. Lanes: Parent (YopDwt), YPIII/pIB102; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304; full-length ΔyopD null mutant (YopDΔ4–303), YPIII/pIB621; full-length ΔyopB null mutant (YopBΔ13–399), YPIII/pIB615; double ΔyopB, yopD null mutant, YPIII/pIB619; multiple ΔyopHMEK null mutant, YPIII/pIB29MEK. Apparent molecular weights are presented to the right of the autoradiogram. Arrowheads indicate YopB–YopD-dependent high-molecular-weight complexes.

cmi12100-sup-0010-FigS10.tif1762K

Fig. S10. Oligomerization of secreted YopD after chemical cross-linking. Yersinia bacteria were grown in T3S permissive conditions and the protein secreted into the external milieu was recovered and incubated in the absence (−) or presence (+) of the cross-linking agent EGS. Samples were then separated by SDS-PAGE and immunoblotted with affinity-purified polyclonal rabbit anti-YopD antiserum. Lanes: Parent (YopDwt), YPIII/pIB102; YopDΔ256–275, YPIII/pIB633; YopDI262P, YPIII/pIB63301; YopDA263P, YPIII/pIB63302; YopDK267P, YPIII/pIB63303; YopDA270P, YPIII/pIB63304. The positions of the protein standards with their approximate molecular mass (kDa) are marked at left. Asterisks (*) at right mark protein bands assumed to represent higher-order YopD oligomers. As observed in our earlier analyses (Costa et al., 2010), monomeric YopD migrates differently depending on the individual YopD mutant (◂). Internal cross-linking of monomeric YopD is presumably a cause of retarded migration in the presence of cross-linker (¤). The experiment was repeated three times.

cmi12100-sup-0011-TableS1.docx19K

Table S1. Oligonucleotides used in this study.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.