Gamma Delta TCR and the WC1 Co-Receptor Interactions in Response to Leptospira Using Imaging Flow Cytometry and STORM

Abstract

The WC1 cell surface family of molecules function as hybrid gamma delta (γδ) TCR co-receptors, augmenting cellular responses when cross-linked with the TCR, and as pattern recognition receptors, binding pathogens. It is known that following activation, key tyrosines are phosphorylated in the intracytoplasmic domains of WC1 molecules and that the cells fail to respond when WC1 is knocked down or, as shown here, when physically separated from the TCR. Based on these results we hypothesized that the colocalization of WC1 and TCR will occur following cellular activation thereby allowing signaling to ensue. We evaluated the spatio-temporal dynamics of their interaction using imaging flow cytometry and stochastic optical reconstruction microscopy. We found that in quiescent γδ T cells both WC1 and TCR existed in separate and spatially stable protein domains (protein islands) but after activation using Leptospira, our model system, that they concatenated. The association between WC1 and TCR was close enough for fluorescence resonance energy transfer. Prior to concatenating with the WC1 co-receptor, γδ T cells had clustering of TCR-CD3 complexes and exclusion of CD45. γδ T cells may individually express more than one variant of the WC1 family of molecules and we found that individual WC1 variants are clustered in separate protein islands in quiescent cells. However, the islands containing different variants merged following cell activation and before merging with the TCR islands. While WC1 was previously shown to bind Leptospira in solution, here we showed that Leptospira bound WC1 proteins on the surface of γδ T cells and that this could be blocked by anti-WC1 antibodies. In conclusion, γδ TCR, WC1 and Leptospira interact directly on the γδ T cell surface, further supporting the role of WC1 in γδ T cell pathogen recognition and cellular activation.

Keywords: Leptospira; STORM; WC1; gamma delta T cells; gamma delta TCR.

Figure 1

FRET between cell surface molecules. AMNIS imaging flow cytometry of bovine PBMC either quiescent or after culture with Leptospira antigen for 7 days. For (A–C) the top flow cytometry plots after the arrow indicate fluorescence with both lasers on, while the bottom panels show fluorescence with the red AF642- laser off to measure fluorescence-sensitized emission in that channel. Individual cell pictures from the indicated (from dashed arrow) gated population are also shown. Stained by indirect immunofluorescence with (A) anti-CD3 mAb with anti-IgG1-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor), (B) anti-CD45 mAb with anti-IgG2a-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor), or (C) anti-WC1 mAb CC15 with anti-IgG2a-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor). Dot plots are representative of 3 independent experiments for (A–C). (D) TCR acceptor fluorescence of WC1⁺ cells as a result of anti-WC1 (donor) mAb and anti-TCRδ mAb (acceptor) interaction on ex vivo resting cells (red dots) and Leptospira stimulated cells (blue dots). (E) The mean ± SD percentage of cells showing FRET relative to the maximal number possible from 3 independent experiments are shown. No significant FRET was found for CD45 interaction with TCR but it was for CD3-TCR and WC1-TCR interactions (*p ≤ 0.05 **p ≤ 0.01 by Student’s t-test: CD3 p = 0.005, WC1 p = 0.022).

Figure 2

STORM imaging of bovine lymphocytes for WC1 and TCRδ interaction. Bovine PBMC were imaged by STORM after staining by immunofluorescence with anti-WC1 mAb (CC15) and AF647 isotype specific secondary Ab or anti-TCRδ mAb (GB21A) and AF488 isotype specific secondary Ab. Examples are shown here from 4 experiments conducted with two animals. (A) Ex vivo WC1⁺ γδ T cells and (B) WC1⁺ γδ T cells cultured with sonicated Leptospira antigen for 7 days. Cell size is indicated with white bars. The sub-cellular particle in Panel A is likely a RBC. (C) Examples of individual cells with their corresponding Pearson’s coefficient. (D) Comparison of Pearson’s coefficients from a larger sample of cells with the mean and SD shown.

Figure 3

FRET between different variants of WC1 molecules and the γδ TCR. AMNIS imaging flow cytometry of bovine PBMC either ex vivo or following culture with Leptospira antigen for 7 days. For (A–C) the top flow cytometry plots to the right of the arrow indicate fluorescence with both lasers on, while the bottom panels to the right of the arrow show fluorescence with the red AF642-activating laser off. Individual cell pictures from the indicated (from dashed arrow) gated population are also shown. (A) Staining by indirect immunofluorescence with anti-WC1.1 mAb BAG25A with anti-IgM-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor) or (B) anti-WC1.2 mAb CACTB32A with anti-IgG1-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor). (C) Stained by indirect immunofluorescence with anti-WC1-8 (WC1.3) mAb CACT21A with anti-IgG1-PE Ab (donor) and anti-TCRδ mAb with anti-IgG2b-AF647 Ab (acceptor). AMNIS plots in panels (A–C) are representative of 2 experiments. (D) The % of cells showing FRET relative to the maximal number possible for the 2 experiments is shown in the bar graphs; n.v., not visible but was measured.

Figure 4

FRET between different variants of WC1 molecules. Bovine PBMC were either ex vivo or cultured with Leptospira antigen for 7 days and imaged with Amnis imaging flow cytometry. The top flow cytometry plots after the arrows indicate fluorescence with both lasers on as a positive control, while the bottom panels after the arrows show fluorescence with the red AF642- laser off to measure fluorescence-sensitized emission in that channel. Individual cell pictures from the indicated (dashed arrow) gated population are also shown. Cells were stained by immunofluorescence with (A) anti-WC1.1 mAb (BAG25A does not react with WC 1-8) with anti-IgM-PE Ab and anti-WC1-8 (i.e. WC1.3⁺ cells, mAb CACT21A) with anti-IgG1-AF647 secondary Ab. (B) The % of cells showing FRET relative to the maximal number possible is shown in the bar graphs for the 2 experiments performed.

Figure 5

Temporal development of FRET between various cell surface molecules. (A) Flow cytometry of PBMC loaded with efluor670 cell division dye and then cultured with Leptospira sonicate for up to 7 days and stained by indirect immunofluorescence with anti-TCRδ mAb. (B, C) are results ofAMNIS imaging flow cytometry of bovine PBMC after culturing with Leptospira for variable lengths of time; indirect immunofluorescence with mAb as indicated included GB21A (anti-TCRδ), CC15 (anti-pan-WC1), CACTB32A (anti-WC1.2), BAG25A (anti-WC1.1), CACT21A (anti-WC1-8, marking WC1.3⁺ cells), and MM1A (anti-CD3) and appropriate isotype-specific secondary Abs. The mean percentage of γδ T cells showing FRET relative to the maximal number possible between the molecules is indicated in the bar graphs: (B) anti-CD45, anti-CD3 or anti-WC1 mAbs with anti-TCRδ mAb, (C) anti-WC1.1, anti-WC1.2 or anti-WC1.3 mAbs with anti-TCRδ mAb. (B) shows the mean ± SD of 3 independent experiments while (C) is the average of 2 experiments and thus no SD shown. Significant differences by Student’s t-test are shown (*p ≤ 0.05; **p ≤ 0.01) with specific values being: CD3 at three days p = 0.016, CD3 at five days p = 0.007, CD3 at seven days p = 0.005, and WC1 at seven days p = 0.022.

Figure 6

Leptospira binding to WC1⁺ cells. Leptospira interrogans serovar hardjo bacteria were biotinylated and stained with either streptavidin-PE or streptavidin-AF488. (A) Flow cytometry of unlabeled or biotinylated-streptavidin-PE Leptospira alone. (B) Bovine PBMC incubated with biotinylated Leptospira for 4 hr and stained by indirect immunofluorescence with anti-WC1 mAb CC15 and anti-IgG2a-AF647 secondary Ab. Top panels (unstained controls) had no streptavidin-PE added, while it was added to the bottom panels. (C) AMNIS images of double positive cells (WC1⁺/Leptospira ⁺) from (B). (D) Flow cytometry sorting strategy of WC1⁺-AF647⁺ lymphocytes that bound biotinylated Leptospira ⁺-Steptavidin-AF488 Representative of 2 independent experiments performed. (E) Sorted cells shown in (D) were then imaged by STORM.

Figure 7

Blocking of Leptospira binding to lymphocytes by anti-WC1 Ab. (A) Leptospira interrogans serovar hardjo was biotinylated and stained with Streptavidin-PE. Experimental design of blocking adherence of bacteria to cells by mAbs added either before (pre-stained) or after (post-stained) incubation with the bacteria. (B) Cells post-stained with the indicated mAbs reactive with lymphocyte surface antigens and isotype-specific secondary Abs were compared to those pre-stained before the 3-hr incubation with Leptospira-biotin-streptavidin-PE. Percentages are from gated populations of mAb⁺ cells and evaluated for Leptospira binding from that population. The results show the binding of bacteria for the indicated lymphocyte population. Lines connect the results within an experiment (n=4 independent experiments). Significant differences were done by the Mann-Whitney ranked sum (*p ≤ 0.05 **p ≤ 0.01; mAb CC15 p = 0.007, mAb BAG25A p = 0.069 and mAb ILA29 p = 0.028) NS, not significant. (C) Percentage of blocking by the mAbs in (B) was the difference between percentage of cells binding Leptospira post-stained and pre-stained. A positive number indicates blocking, while a negative number indicates enhanced binding. The mean percentage of blocking is indicated above each treatment.