Quantitative proteomic screen identifies annexin A2 as a host target for Salmonella pathogenicity island-2 effectors SopD2 and PipB2

Abstract

Intracellular pathogens need to establish an intracellular replicative niche to promote survival and replication within the hostile environment inside the host cell. Salmonella enterica serovar Typhimurium (S. Typhimurium) initiates formation of the unique Salmonella-containing vacuole and an extensive network of Salmonella-induced tubules in order to survive and thrive within host cells. At least six effectors secreted by the type III secretion system encoded within Salmonella pathogenicity island-2 (SPI-2), namely SifA, SopD2, PipB2, SteA, SseJ, and SseF, purportedly manipulate host cell intracellular trafficking and establish the intracellular replicative niche for S. Typhimurium. The phenotypes of these effectors are both subtle and complex, complicating elucidation of the mechanism underpinning host cell manipulation by S. Typhimurium. In this work we used stable isotope labeling of amino acids in cell culture (SILAC) and a S. Typhimurium mutant that secretes increased amounts of effectors to identify cognate effector binding partners during infection. Using this method, we identified the host protein annexin A2 (AnxA2) as a binding partner for both SopD2 and PipB2 and were able to confirm its binding to SopD2 and PipB2 by reciprocal pull down, although there was a low level of non-specific binding of SopD2-2HA and PipB2-2HA to the Ni-Sepharose beads present. We further showed that knockdown of AnxA2 altered the intracellular positioning of the Salmonella containing vacuole (SCV). This suggests that AnxA2 plays a role in the subcellular positioning of the SCV which could potentially be mediated through protein-protein interactions with either SopD2 or PipB2. This demonstrates the value of studying effector interactions using proteomic techniques and natural effector delivery during infection rather than transfection.

Figure 1

ssrB overexpression increases T3SS2 secretion, but not T3SS1 secretion. Images shown are representative of n = 3 experiments. (a) Secreted proteins from indicated strains grown under SPI-1 inducing conditions were separated by SDS-PAGE and stained with Coomassie Brilliant R250. (b,c) Secreted proteins from indicated strains grown under SPI-2 inducing conditions were precipitated from sterilized culture supernatant and separated on an SDS-PAGE gel and immunoblotted with α-SseD (b) or α-SseB (c). (a–c) Images were cropped where indicated by dividing lines. Original Coomassie stained SDS-PAGE gel and western blots are shown in Supplemental Fig. S2. (d) Densitometric analysis of SPI-1 secreted effectors SipA, SipC, and SopE as determined by 3 SDS-PAGE gels stained by Coomassie. Ns not significant as determined by a Kruskal–Wallis test with Dunn’s correction for multiple comparisons. (e) Fold change determined by densitometry of 3 Western blots from independent experiments. Mean fold change ± standard deviation is shown. Fold change = (Densitometry of T3SS2⁺)/(Densitometry of wild type). Dashed line = 1.

Figure 2

T3SS2⁺ infected HeLa cells exhibit increased SIF frequency. HeLa cells were infected at an MOI ≈ 100 with the indicated strains for 8 hours prior to cell fixation. Cells were immunostained for Salmonella (red) and LAMP1 (green), and the nucleus was stained with DAPI (blue). 60 distinct fields of view were used for quantification with at least 1 infected cell per field of view for each experiment. (a) Percentage of cells infected as determined by enumerating the number of both infected and uninfected cells in each field of view. Mean ± standard error of the mean is shown (n = 3). Significance determined by a Kruskall–Wallis test with Dunn’s multiple comparisons post-test. p-values are as indicated on graph. (b) Percent of infected cells with SIFs determined by enumerating the number of infected cells and the number of infected cells with SIFs per field of view. Mean ± standard error of the mean is shown (n = 3). (c) Representative images of HeLa cells infected with a MOI ≈ 100 with indicated strain at 8 h.p.i. White boxes indicate zoomed-in region in inset. Arrowheads indicate LAMP1⁺-tubules. Scale Bar = 10 µm.

Figure 3

Complementation of SIF biogenesis. HeLa cells were infected at an MOI ≈ 100 with the indicated strains for 8 hours prior to cell fixation. Cells were immunostained for Salmonela (red) and LAMP1 (green), and the nucleus was stained with DAPI (blue). 60 distinct fields of view were used for quantification with at least 1 infected cell per field of view for each experiment. (a–g) Quantification of LAMP1⁺-tubule frequency (SIFs) in HeLa cells infected with single effector deletion mutants in the T3SS2⁺ background and their corresponding complemented strains after 8 h of infection. The average frequency of infected cells with LAMP1⁺-tubules ± standard error of the mean is shown (n = 3). Strains were analyzed in a single repeated experiment but have been divided into separate panels for the sake of clarity. At least 60 infected cells per strain were blindly analyzed in each experiment. p-values are as indicated as determined by two-tailed Mann–Whitney test with a 95% confidence level. (h) Representative images for select strains. White boxes indicate zoomed-in region in inset. Arrowheads indicate LAMP1⁺-tubules. Scale Bar = 10 µm.

Figure 4

SopD2 and PipB2 immunoprecipitation reveals common proteins that are directly or indirectly linked. STRING analysis on the top hits from the SopD2-IP (a) and PipB2-IP (b) as identified by mass spectrometry. Median SILAC ratios are shown for each protein as determined from 3 independent experiments. STRING networks show both functional and physical protein associations. Line thickness indicates the strength of data supporting the interaction with a minimum confidence of interaction score of 0.4.

Figure 5

SopD2 targets AnxA2 in vitro. (a) Representative blot of reciprocal pull-down of AnxA2 and SopD2 or PipB2. Concentrated culture supernatants of S. Typhimurium strains secreting SopD2-2HA, PipB2-2HA, or HA empty vector were mixed with His₆-AnxA2. Pull-downs were analyzed by Western blot using α-HA or α-AnxA2. IB: Immunoblot. Leftmost lane: pull-down of SopD2-2HA in the absence of AnxA2. Lane second from the left: pull-down of SopD2-2HA in the presence of AnxA2. Middle lane: pull-down of PipB2-2HA in the absence of AnxA2. Lane second from the right: pull-down of PipB2-2HA in the presence of AnxA2. Rightmost lane: pull-down of empty vector expressing a HA-tag in the presence of AnxA2. Original Western blots are shown in Supplemental Fig. S4. (b) Fold change determined by densitometry of 3 Western blots from independent experiments. Mean fold change ± standard deviation is shown. Fold change = (Densitometry of effector + AnxA2)/(Densitometry of effector). Dashed line = 1.

Figure 6

Annexin A2 contributes to the precise intracellular positioning of the SCV during infection. (a) STRING network showing both direct and indirect protein associations between AnxA2 and previously identified host proteins known to be involved in SIF biogenesis. Line thickness indicates the strength of data supporting the interaction with a minimum confidence of interaction score of 0.4. (b) Annexin A2 expression in Hela cells treated with either control shRNA (left lane) or with Annexin A2 shRNA (right lane) as determined by Western blot from whole cell lysate. Beta Tubulin was used as a loading control. Original Western blots are shown in Supplemental Fig. S5. (c) Distribution of Salmonella in infected HeLa cells treated with either control shRNA or Annexin A2 shRNA. shRNA-treated HeLa cells infected with wild type Salmonella were fixed at 8 h post infection, immunostained for Salmonella (red), LAMP1 (green), and actin (grey), and the nucleus was stained with DAPI (blue). The nearest edge distance from Salmonella to nucleus was measured in infected cells. The mean plus or minus SEM for three experiments is shown (n = 3). At least 200 distances per experiment were measured. The p-value is as indicated as determined by a Mann–Whitney test. (d) Select representative images of wild type Salmonella-infected HeLa cells treated with either control shRNA or Annexin A2 shRNA.