Fig. 3

Celecoxib impairs immune memory and affects CD4 T cell phenotype. a 6 to 8-week-old female CB6F1 mice were infected with ~25–50 CFU Mtb Erdman by the aerosol route and fed with celecoxib chow for 8 weeks as previously (week 4–12). At week 10, mice were put on antibiotic treatment with rifampicin/isoniazid for 12 weeks to clear the infection (week 10–22). After a 2-week rest to wash out the antibiotics, mice re-infected with 50–100 CFU Mtb Erdman by the aerosol route. b Mtb bacilli were enumerated in the lungs at week 3 (n = 10) and 6 (n = 16) of the re-infection by plating organ homogenates. c The number of cytokine positive CD4 T cells (producing any combination of IFNγ, TNFα or IL-2, see and Supplementary Fig. 5) was determined by ICS of lung cells stimulated with overlapping ESAT-6 peptides (n = 6). d Female CB6F1 mice were infected with ~25–50 CFU Mtb Erdman by the aerosol route and fed with celecoxib chow as previously (same experiment as in Fig. 1). e The total number of CD4 T cells as well as the frequency of cytokine producing CD4 T cells (IFNγ, TNFα or IL-2) was determined by ICS of lung cells stimulated with overlapping ESAT-6 peptides (week 0 n = 2, week 6 n = 6, week 12 n = 8). f At week 12, the ability of CD4 T cells to home to infected tissue was assessed by enumerating the CD4 T cells that were not stained by i.v. injection of a labeled CD45 antibody (IV⁻). g The Th1 differentiation status of activated CD4 T cells (CD44hi) was assessed in both cytokine producing (red) and non-producing (grey) T cells by staining for the linage-specific transcription factor, T-bet (data show a concatenation of every animal in each group). h (left) the frequency of T-bet⁺ CD4 T cells in the cytokine producing population of each animal, (right) T-bet expression measured by mean fluorescence intensity (MFI) of the population of T-bet⁺ CD4 T cells (n = 6). Bars represent mean ± SEM and p values were calculated with an unpaired two-tailed t-test comparing treated and non-treated animals