Human lung epithelial cells contain Mycobacterium tuberculosis in a late endosomal vacuole and are efficiently recognized by CD8⁺ T cells

Abstract

Mycobacterium tuberculosis (Mtb) is transmitted via inhalation of aerosolized particles. While alveolar macrophages are thought to play a central role in the acquisition and control of this infection, Mtb also has ample opportunity to interact with the airway epithelium. In this regard, we have recently shown that the upper airways are enriched with a population of non-classical, MR1-restricted, Mtb-reactive CD8⁺ T cells (MAIT cells). Additionally, we have demonstrated that Mtb-infected epithelial cells lining the upper airways are capable of stimulating IFNγ production by MAIT cells. In this study, we demonstrate that airway epithelial cells efficiently stimulate IFNγ release by MAIT cells as well as HLA-B45 and HLA-E restricted T cell clones. Characterization of the intracellular localization of Mtb in epithelial cells indicates that the vacuole occupied by Mtb in epithelial cells is distinct from DC in that it acquires Rab7 molecules and does not retain markers of early endosomes such as Rab5. The Mtb vacuole is also heterogeneous as there is a varying degree of association with Lamp1 and HLA-I. Although the Mtb vacuole shares markers associated with the late endosome, it does not acidify, and the bacteria are able to replicate within the cell. This work demonstrates that Mtb infected lung epithelial cells are surprisingly efficient at stimulating IFNγ release by CD8⁺ T cells.

Figure 1. The human lung epithelial cell line BEAS-2B is less efficiently infected with Mtb than human DC.

A–B) Primary human DC or the human bronchial epithelial cell line, BEAS-2B, were infected with dsRED Mtb at an MOI of 30, 10, and 3. Fixed cells were assessed for dsRED-H37Rv infection by flow cytometry after 18 hours. FlowJo (TreeStar) was used to calculate the change in geometric mean in BEAS-2B cells or DC for MOI:10. Results are representative of all experiments (N = 3). C–E) DC, BEAS-2B cells, or primary LAEC were infected with Mtb (MOI:7). Infected cells were imaged in an unbiased manner, and the number of bacteria per cell was enumerated. Results are representative of all experiments (N = 3 for DC and BEAS-2B cells, N = 2 for LAEC).

Figure 2. Mtb-infected human AEC are efficiently recognized by Mtb-specific CD8⁺ T cells.

A–B) Mtb-infected DC, BEAS-2B cells, or LAEC (MOI:10) were used as APCs in an IFN-γ ELISPOT assay. A) IFN-γ release by the HLA-B45-restricted CD8⁺ T cell clone D466H4 was assessed by ELISPOT assay. 10,000 T cells were used with a titration of APCs from 10,000 - 1,250 cells per well. Results are representative of all experiments (N = 3). Error bars represent the mean and standard error from duplicate wells. B) IFN-γ release by the MR1-restricted (D426B1), HLA-E-restricted (D160 1-23), and HLA-B45-restricted (D466H4) CD8⁺ T cell clones was assessed for Mtb-infected DC, BEAS-2B, or primary LAEC (except D466H4). The fold-difference in IFN-γ release was normalized to the T cell response to Mtb-infected DC (5,000 DC/well) for each APC type and T cell clone. Results are representative of all experiments (N = 3 for DC and BEAS-2B cells and N = 1 for LAEC). C–F) Following fixation in 4% PFA, uninfected or infected BEAS-2B cells, DC, or LAEC were surface stained with either the pan-HLA-I antibody W6/32 (C-F) or the anti-MR1 antibody 26.5 (G) and analyzed by flow cytometry. C-E: shaded histogram, UI; black line, Mtb. F-G: shaded histogram, isotype staining for each cell type. Results are representative of all experiments (N = 5 for W6/32 and N = 4 for 26.5). H) DC or BEAS-2B cells were pulsed with CFP10_2-9 peptide at the indicated concentrations for 1 hour and assessed for their ability to stimulate IFN-γ production by the HLA-B45 restricted, CFP10_2-9 specific, D466H4 T-cell clone. APCs were added at 10,000 cells/well, while T cells were kept constant at 500 T cells per well. Results are representative of all experiments (N = 3). Error bars represent the mean and standard error from duplicate wells.

Figure 3. Intracellular Mtb are required for T cell activation.

A) BEAS-2B cells were seeded in a 12-well tissue culture plate and a 0.4 uM pore size transwell insert. Cells in the transwell were infected with Mtb (MOI:10) for 18 hrs. Cells from both chambers were then used as APCs (10,000 cells/well) in an IFN-γ Elispot assay. Results are representative of three independent experiments. B-C) Mtb were labeled with streptavidin coated magnetic microbeads prior to infection (MOI:30). A fraction enriched for infected BEAS-2B cells was then obtained by magnetic sorting (adherent). Mtb CFUs were enumerated from each fraction (B) and cells were used as APCs (5,000 cells/well) in an IFN-γ Elispot assay (C). (N = Non-adherent, A = Adherent). Results are representative of three independent experiments. D) The T cell response to the enriched fraction of Mtb-infected BEAS-2B cells (MOI:30, 2,000 cells/well) was compared to an equal number of Mtb-infected DC (MOI:30) in an IFN-γ Elispot assay. Results are representative of two independent experiments. For all assays, error bars represent the mean and standard error from duplicate wells.

Figure 4. The Mtb vacuole in AEC retains features of a late endosome.

A–F) BEAS-2B cells were infected with dsRED-expressing Mtb or RFP-expressing M. smegmatis (Msm) at MOI:5 for 4 or 18 hours. Infected cells were fixed and stained for early endosomal markers Rab5 and EEA1. G-I) BEAS-2B cells were transfected with a plasmid expressing a GFP-Rab7 fusion construct for 6 hours prior to infection and fixed following a 4 hour infection. For each condition, at least 200 intracellular bacteria were counted from 3 independent experiments. Mtb-containing vacuoles were categorized as positive or negative, the mean percent positive and standard error were determined, and a Student's t-test was used to determine statistical significance between groups. A,D,G (Mtb); B,E,H (Msm). *p<0.05. J–L) Primary LAEC were infected with dsRED Mtb at MOI:5 for 18 hours and analyzed as described above. N = 2. Scale bar = 10 uM. For all assays, error bars represent the mean and standard error of the mean (SEM) for all events. Images shown are representative of Mtb or Msm that were positively (B,D,G,H,J,L) or negatively (A,C,K) associated with each marker.

Figure 5. Rab7 accumulates in Msm-containing vacuoles.

From 0–2 hours following synchronized infection, individual Rab7-GFP expressing BEAS-2B cells containing RFP-Msm were imaged sequentially. Each cell was imaged every 30 sec for 5 minutes. Shown is a representative 5 minute time-lapse sequence of Rab7 accumulation on an RFP-Msm compartment. In each experiment, we obtained time-lapse images from 8–10 cells. This experiment was repeated four times. The graph represents the GFP relative fluorescent units (RFU) calculated at the perpendicular transection of the bacterium in a single 0.2 uM z-plane at each time point, as shown in the first panel. Scale bar = 3 uM.

Figure 6. Transferrin receptor (TfR) does not accumulate at high levels in the Mtb vacuole.

A–C) BEAS-2B cells were infected with dsRED-expressing Mtb or RFP-expressing Msm at MOI:5 for 4 or 18 hours. Infected cells were fixed and stained for TfR. At least 200 intracellular bacteria were counted from 3 independent experiments. Mtb-containing vacuoles were categorized as positive or negative, the mean percent positive and standard error were determined, and a Student's t-test was used to determine statistical significance between groups. A (Mtb); B (Msm). Images are representative of TfR-positive compartments for Mtb and Msm. Error bars represent the mean and SEM for all events. Scale bar = 10 uM. D) Primary LAEC were infected with dsRED-expressing Mtb at MOI:5 for 18 hours. Infected cells were fixed and stained for TfR and analyzed as described above. Image is representative of a TfR-negative compartment. Scale bar = 10 uM. E-F) BEAS-2B cells were infected with dsRED-Mtb or RFP-Msm (MOI:10). Infection was synchronized by centrifugation. After a 2 hour infection, cells were washed three times with PBS, and wells replenished with media. Cells were harvested and counted, and lysates were plated in triplicate after the 2 hour infection, and at 24, 48, and 72 hours following infection. Error bars represent the mean and SEM from three experiments (Mtb) or the mean and standard error from triplicate wells (Msm).

Figure 7. The Mtb vacuole in BEAS-2B cells acquires Lamp1, but does not acidify.

A–C) BEAS-2B cells were infected with dsRED-expressing Mtb or RFP-expressing Msm at MOI:5 for 4 or 18 hours. Infected cells were fixed and stained for the lysosomal associated protein, Lamp1. Over 200 intracellular bacteria were counted from 4 independent experiments. Mtb-containing vacuoles were categorized as positive or negative, the mean percent positive and standard error were determined, and a Student's t-test was used to determine statistical significance between groups. Error bars represent the mean and SEM for all events (**p<0.01) D) Primary LAEC were infected with dsRED-expressing Mtb at MOI:5 for 18 hours and analyzed as described above. Scale bar = 10 uM. Images are representative of Lamp1-positive (A,D) or Lamp1-negative (B) Mtb containing compartments. E-F) DC or BEAS-2B cells were infected with GFP-expressing Mtb or GFP-expressing Msm linked to the pH sensitive dye pHrodo Red. Cells were fixed 18 hours after infection and images were acquired on a DeltaVision Core DV wide-field microscope. E) pHrodo signal from perpendicular transection on a single 0.2 µm z-stack was used to generate the relative fluorescence unit (RFU) data point for each individual bacterium. The image on the left is an example of a GFP+ bacterium (green) associated with low pHrodo Red (red) signal, while the image on the right is an example of a bacterium with low GFP signal, but high pHrodo Red signal. Lines on the image indicate the perpendicular transection of the bacterium used to generate the RFU for GFP and pHrodo Red indicated in the panels below the images. F) The pHrodo Red RFU for at least 200 individual bacteria from 4 independent experiments for each condition was plotted on a log scale. Error bars represent the mean RFU and SEM for all events. (*p<0.001)

Figure 8. The Mtb vacuole in epithelial cells acquires Class I molecules.

A–C) BEAS-2B cells were infected with dsRED-expressing Mtb or RFP-expressing M. smegmatis at MOI:5 for 4 or 18 hours. Infected cells were fixed and stained for Class I molecules (W6/32). At least 200 intracellular bacteria were counted from 3 independent experiments. Mtb-containing vacuoles were categorized as positive or negative, the mean percent positive and standard error were determined, and a Student's t-test was used to determine statistical significance between groups. A (Mtb); B (Msm). Scale bar = 10 uM. D) Primary LAEC were infected with dsRED-expressing Mtb at MOI:5 for 18 hours and analyzed as described above. Infected cells were fixed and stained with W6/32. Scale bar = 10 uM. Images are representative of Mtb or Msm associated with HLA-I. Error bars represent the mean and SEM for all events.