Resident Th1-like effector memory cells in pulmonary recall responses to Mycobacterium tuberculosis

Abstract

We recently described a model of Th1 recall responses based on segmental antigen challenge with purified protein derivative of Mycobacterium tuberculosis (PPD). Bronchoscopic instillation of 0.5 tuberculin units of PPD resulted in localized lymphocytic inflammation in PPD-positive subjects only. Recruited lymphocytes were predominantly CD4+ and were enriched for cells capable of PPD-specific interferon (IFN)-gamma production. In the current study, we investigated the mechanisms by which this localized recall response is mobilized. Bronchoscopic PPD challenge of skin test-positive subjects resulted in the production of CXCR3 ligands IFN-gamma-inducible protein (IP)-10 and monokine induced by IFN-gamma (Mig), but not of CCR5 ligands macrophage inflammatory protein-1alpha and regulated-upon activation, normal T-cell expressed and secreted, whereas skin test-negative subjects produced none of these chemokines. Baseline bronchoalveolar lavage (BAL) cells of skin test-positive subjects produced IP-10 and Mig in response to in vitro stimulation as well. Because IP-10 and Mig are IFN-gamma-inducible chemokines, these findings suggested that chemokine responses to PPD were facilitated by resident memory cells of the lung. Further studies confirmed that baseline BAL lymphocytes of PPD-positive subjects produce IFN-gamma in response to PPD, and that, compared with peripheral blood, BAL cells are preferentially enriched for PPD-specific lymphocytes. This IFN-gamma production is predominantly a function of CD4+ T cells that display the CD45RO+/CCR7- surface phenotype characteristic of effector memory cells.

Figure 1.

Segmental antigen challenge with PPD stimulated local production of IP-10 and Mig in skin test–positive subjects only. Nonconcentrated BAL fluid of PPD-positive subjects contains large amounts of CXCR3 chemokine ligands IP-10 and Mig, but not CCR5 ligands MIP1α and RANTES (A). In contrast, BAL fluid of PPD-negative subjects did not contain measurable amounts of any of the chemokines measured (B). Chemokines were measured in BAL fluid obtained before PPD challenge (white bars), and 48 h after instillation of saline into control segments (striped bars) and of 0.5 TU of PPD into challenged segments (black bars).

Figure 1.

Segmental antigen challenge with PPD stimulated local production of IP-10 and Mig in skin test–positive subjects only. Nonconcentrated BAL fluid of PPD-positive subjects contains large amounts of CXCR3 chemokine ligands IP-10 and Mig, but not CCR5 ligands MIP1α and RANTES (A). In contrast, BAL fluid of PPD-negative subjects did not contain measurable amounts of any of the chemokines measured (B). Chemokines were measured in BAL fluid obtained before PPD challenge (white bars), and 48 h after instillation of saline into control segments (striped bars) and of 0.5 TU of PPD into challenged segments (black bars).

Figure 2.

IP-10 and Mig are produced by baseline BAL cells of skin test–positive subjects in response to in vitro stimulation with PPD. White bars indicate chemokine levels in supernatants of unstimulated cell cultures, whereas black bars show chemokine levels in cultures stimulated with PPD. After 48 h of in vitro stimulation with PPD, resident alveolar cells obtainable by BAL of skin test–positive subjects have the ability to produce IFN-γ–dependent chemokines Mig (A) and IP-10 (B), whereas those of PPD-negative subjects do not. Figures show mean and SD values (n = 3 for each subject group).

Figure 2.

IP-10 and Mig are produced by baseline BAL cells of skin test–positive subjects in response to in vitro stimulation with PPD. White bars indicate chemokine levels in supernatants of unstimulated cell cultures, whereas black bars show chemokine levels in cultures stimulated with PPD. After 48 h of in vitro stimulation with PPD, resident alveolar cells obtainable by BAL of skin test–positive subjects have the ability to produce IFN-γ–dependent chemokines Mig (A) and IP-10 (B), whereas those of PPD-negative subjects do not. Figures show mean and SD values (n = 3 for each subject group).

Figure 3.

Baseline BAL cells of PPD-positive subjects display antigen-specific production of IFN-γ in response to PPD. ELISPOT for IFN-γ was performed using baseline BAL cells from both PPD-positive and PPD-negative subjects as described in the text. White bars indicate responses of unstimulated cells and black bars represent responses of cells incubated with PPD. PPD-positive subjects displayed significantly more IFN-γ–positive cells in response to stimulation with PPD than did skin test–negative subjects. For the later group, IFN-γ production in response to PPD was not significantly greater than that observed in cells cultured in medium alone. Results are expressed as number of IFN-γ–producing cells per 10 K total BAL cells (n = 4 for each group).

Figure 4.

CD4+ T cells are the predominant source of antigen-specific IFN-γ production in response to PPD. Production of IFN-γ by baseline BAL cells was assessed using intracellular staining for IFN-γ in combination with surface markers for specific lymphocyte subsets as described. (A) Mean data of total BAL lymphocytes expressing surface markers of the various subsets for four PPD-positive subjects studied. In this stacked bar format, black portions of each bar indicate lymphocytes that were negative for IFN-γ production, whereas white portions of each bar indicate lymphocytes that displayed positive intracellular staining for IFN-γ after PPD stimulation. B Re-expresses these data in terms of the percentage of all IFN-γ–producing lymphocytes that were accounted for by each lymphocyte subset. CD4+ T cells accounted for over 80% of the BAL lymphocytes producing IFN-γ in response to stimulation with PPD, as illustrated.

Figure 4.

CD4+ T cells are the predominant source of antigen-specific IFN-γ production in response to PPD. Production of IFN-γ by baseline BAL cells was assessed using intracellular staining for IFN-γ in combination with surface markers for specific lymphocyte subsets as described. (A) Mean data of total BAL lymphocytes expressing surface markers of the various subsets for four PPD-positive subjects studied. In this stacked bar format, black portions of each bar indicate lymphocytes that were negative for IFN-γ production, whereas white portions of each bar indicate lymphocytes that displayed positive intracellular staining for IFN-γ after PPD stimulation. B Re-expresses these data in terms of the percentage of all IFN-γ–producing lymphocytes that were accounted for by each lymphocyte subset. CD4+ T cells accounted for over 80% of the BAL lymphocytes producing IFN-γ in response to stimulation with PPD, as illustrated.

Figure 5.

IFN-γ–producing CD4+ BAL T cells of PPD-positive subjects display memory-effector cell phenotype. BAL cells were incubated overnight with PPD. Intracellular staining for IFN-γ was performed, as was surface staining for various markers as described in MATERIALS AND METHODS. Representative results gated on CD4+ lymphocytes are shown. (A) Assessment of intracellular IFN-γ (x axis) and expression of the memory marker CD45RO (y axis) indicates that 94.7% of CD4+ T cells that produce IFN-γ in response to PPD express CD45RO. (B) Assessment of intracellular IFN-γ (x axis) and expression of CCR7 (y axis) of CD45RO+ CD4 T cells indicates that 99.3% of memory CD4+ T cells capable of PPD-specific IFN-γ production are negative for CCR7.