Control of Mycobacterium tuberculosis growth by activated natural killer cells

Abstract

We characterized the underlying mechanisms by which glutathione (GSH)-enhanced natural killer (NK) cells inhibit the growth of Mycobacterium tuberculosis (M. tb) inside human monocytes. We observed that in healthy individuals, treatment of NK cells with N-acetyl cysteine (NAC), a GSH prodrug in conjunction with cytokines such as interleukin (IL)-2 + IL-12, resulted in enhanced expression of NK cytotoxic ligands (FasL and CD40L) with concomitant stasis in the intracellular growth of M. tb. Neutralization of FasL and CD40L in IL-2 + IL-12 + NAC-treated NK cells resulted in abrogation in the growth inhibition of M. tb inside monocytes. Importantly, we observed that the levels of GSH are decreased significantly in NK cells derived from individuals with HIV infection compared to healthy subjects, and this decrease correlated with a several-fold increase in the growth of M. tb inside monocytes. This study describes a novel innate defence mechanism adopted by NK cells to control M. tb infection.

Fig. 1

Intracellular survival of H37Rv inside natural killer (NK) cell–monocyte co-cultures from healthy individuals: human monocytes were infected with the processed virulent laboratory strain of Mycobacterium tuberculosis, H37Rv at a multiplicity of infection of 10:1. NK cells were purified using mini-magnetic affinity cell sorting (MACS) column. NK cells were treated as follows: (a) no additives; (b) N-acetyl cysteine (NAC) (20 mM); (c) NAC (20 mM) + interleukin (IL)-2 (100 U/ml) and IL-12 (100 U/ml); and (d) buthionine sulphoximine (BSO) (500 µM) for 24 h. NK cells were washed, resuspended in fresh media and then added to the infected monocytes. Infected monocyte–NK cell co-cultures were terminated at 1 h and 5 days post-infection to determine the intracellular survival of H37Rv inside human monocytes. Monocyte lysates were plated on 7H11 medium enriched with albumin dextrose complex (ADC) to estimate the growth or killing of H37Rv. Results shown are averages from five different experiments performed in triplicate. *Significant difference compared to monocytes cultured in the absence of NK cells.

Fig. 2

Expression of natural killer (NK) cytotoxic receptors (NKP44, NKP30 and NKP46) on NK cells from healthy individuals. NK cells were treated as follows: no additives, 20 mM of freshly prepared N-acetyl cysteine (NAC), interleukin (IL)-2 (100 U/ml) + IL-12 (100 U/ml), IL-2 + IL-12 + NAC and buthionine sulphoximine (BSO) (500 µM); 24 h after incubation, NK cells were washed and resuspended in 100 µl of phosphate-buffered saline (PBS) containing phycoerythrin (PE)-labelled NKP44 (a), PE-labelled NKP30 (b) and PE-labelled NKP46 (c), incubated at 4°C for 30 min and analysed by flow cytometry. Values obtained from different treatments are normalized to control. Data in (a–c) represent means ± standard error from four different healthy individuals performed in triplicate. Lower panels (a–c) denote representative fluorescence intensity from one experiment.

Fig. 2

Expression of natural killer (NK) cytotoxic receptors (NKP44, NKP30 and NKP46) on NK cells from healthy individuals. NK cells were treated as follows: no additives, 20 mM of freshly prepared N-acetyl cysteine (NAC), interleukin (IL)-2 (100 U/ml) + IL-12 (100 U/ml), IL-2 + IL-12 + NAC and buthionine sulphoximine (BSO) (500 µM); 24 h after incubation, NK cells were washed and resuspended in 100 µl of phosphate-buffered saline (PBS) containing phycoerythrin (PE)-labelled NKP44 (a), PE-labelled NKP30 (b) and PE-labelled NKP46 (c), incubated at 4°C for 30 min and analysed by flow cytometry. Values obtained from different treatments are normalized to control. Data in (a–c) represent means ± standard error from four different healthy individuals performed in triplicate. Lower panels (a–c) denote representative fluorescence intensity from one experiment.

Fig. 3

Expression of natural killer (NK) activation receptor (NKG2D) and cytotoxic ligands (FasL and CD40L) on NK cells from healthy individuals. NK cells were treated as follows: no additives, freshly prepared N-acetyl cysteine (NAC) (20 mM), IL-2 (100 U/ml) + IL-12 (100 U/ml), IL-2 + IL-12 + NAC and buthionine sulphoximine (BSO) (500 µM); 24 h after incubation, NK cells were washed and resuspended in 100 µl of phosphate-buffered saline (PBS) containing phycoerythrin (PE)-labelled NKG2D (a), PE-labelled FasL (b) and fluorescein isothiocyanate (FITC)-labelled CD40L (c), incubated at 4°C for 30 min and analysed by flow cytometry. Values obtained from different treatments are normalized to control. Data in (a) and (b) represent means ± standard error (s.e.) from four different healthy individuals performed in triplicate. Data in (c) represent means ± s.e. from six different healthy individuals performed in triplicate. Lower panels (a–c) denote representative fluorescence intensity from one experiment.

Fig. 3

Expression of natural killer (NK) activation receptor (NKG2D) and cytotoxic ligands (FasL and CD40L) on NK cells from healthy individuals. NK cells were treated as follows: no additives, freshly prepared N-acetyl cysteine (NAC) (20 mM), IL-2 (100 U/ml) + IL-12 (100 U/ml), IL-2 + IL-12 + NAC and buthionine sulphoximine (BSO) (500 µM); 24 h after incubation, NK cells were washed and resuspended in 100 µl of phosphate-buffered saline (PBS) containing phycoerythrin (PE)-labelled NKG2D (a), PE-labelled FasL (b) and fluorescein isothiocyanate (FITC)-labelled CD40L (c), incubated at 4°C for 30 min and analysed by flow cytometry. Values obtained from different treatments are normalized to control. Data in (a) and (b) represent means ± standard error (s.e.) from four different healthy individuals performed in triplicate. Data in (c) represent means ± s.e. from six different healthy individuals performed in triplicate. Lower panels (a–c) denote representative fluorescence intensity from one experiment.

Fig. 4

Neutralization of CD40L and FasL in natural killer (NK) cells treated with interleukin (IL)-2 + IL-12 + N-acetyl cysteine (NAC). Human monocytes isolated from healthy individuals were infected with processed H37Rv at a multiplicity of infection of 10:1. NK cells were purified using a mini-magnetic affinity cell sorting (MACS) column. NK cells were treated as follows: (a) no additives; (b) NAC (20 mM) + IL-2 (100 U/ml) and IL-12 (100 U/ml); (c) IL-2 + IL-12 + NAC + anti-FasL (100 µg/ml); (d) IL-2 + IL-12 + NAC + anti-CD40L (100 µg/ml); (e) IL-2 + IL-12 + NAC + anti-FasL + anti-CD40L (100 µg/ml); and (f) buthionine sulphoximine (BSO) (500 mM). Neutralizing antibodies-treated NK cells were added to H37Rv-infected monocytes. Infected cultures were terminated at 1 h and 5 days post-infection to determine the effects of neutralization of FasL and CD40L in abrogating the growth inhibition of H37Rv inside human monocytes. Macrophage lysates were plated on 7H11 medium enriched with albumin dextrose complex (ADC) to estimate the growth or killing of H37Rv. Results shown in are averages from five different experiments performed in triplicate. *Significant difference compared to monocytes cultured in the absence of NK cells.

Fig. 5

Assay of glutathione (GSH) levels in natural killer (NK) cells isolated from healthy subjects and individuals with human immunodeficiency virus (HIV) infection. Intracellular levels of GSH in NK cell lysates from healthy subjects and individuals with HIV infection were determined by spectrophotometry, using an assay kit from Calbiochem. NK cells (2 × 10⁵/well) lysed with ice-cold 5% metaphosphoric acid (MPA) was added to the pellet. Supernatants were collected after centrifugation and analysed for total GSH as per the manufacturer's instructions. Total GSH in the samples were normalized with protein. Proteins in the samples were estimated by Bradford's method using BioRad reagent. Results shown are averages from experiments performed in five healthy subjects and 10 individuals with HIV infection.

Fig. 6

Intracellular survival of H37Rv inside natural killer (NK) cell–monocyte co-cultures from individuals with human immunodeficiency virus (HIV) infection. Human monocytes isolated from individuals with HIV infection were infected in vitro with processed H37Rv at a multiplicity of infection of 10:1. NK cells were purified using a mini-magnetic affinity cell sorting (MACS) column. NK cells were treated as follows: (a) no additives; (b) N-acetyl cysteine (NAC) (20 mM) + interleukin (IL)-2 (100 U/ml) and IL-12 (100 U/ml); (c) IL-2 + IL-12 + NAC + anti-FasL (100 µg/ml); (d) IL-2 + IL-12 + NAC + anti-CD40L (100 µg/ml); (e) IL-2 + IL-12 + NAC + anti-FasL + anti-CD40L (100 µg/ml) and (f) BSO (500 mM). Neutralizing antibodies-treated NK cells were added to H37Rv-infected monocytes. Infected cultures were terminated at 1 h and 5 days post-infection to determine the effects of neutralization of FasL and CD40L in abrogating the growth inhibition of H37Rv inside human monocytes. Macrophage lysates were plated on 7H11 medium enriched with albumin dextrose complex (ADC) to estimate the growth or killing of H37Rv. Results shown are averages from seven different experiments performed in triplicate.