Fig 1.
PI3K inhibition augments Mtb-driven macrophage MMP-1 secretion.
A. Global inhibition of the PI3K kinase pathway with LY294002 significantly increases macrophage MMP-1 secretion at 72h. B. PI3K inhibition increases macrophage MMP-1 gene expression at 24 h. C. Targeted inhibition of PI3Kδ with IC87114 similarly increases MMP-1 secretion at 72h. D. Increased secretion with IC87114 is secondary to increased mRNA accumulation at 24h. E. Mtb infection increases AKT phosphorylation from 30 minutes, peaking at 60 minutes and declining after 4 h. AKT phosphorylation is completely inhibited by LY294002 at 30 minutes and 4 h. F. Targeted inhibition of PI3K / PDK1 with NVP-BAG956 also increases MMP-1 secretion at 72h. Data show mean and standard deviation of experiments performed in triplicate and are representative of experiments performed on a minimum of three occasions. P values are Student’s t-test.
Fig 2.
PI3K inhibition has global effects on MMP, cytokine and chemokine secretion by macrophages.
Luminex profiling of MMPs, cytokines, chemokines and growth factors demonstrates that Mtb infection significantly upregulates a wide range of mediators from macrophages at 72h. Secretion of the majority of mediators was further augmented when PI3K signalling was globally inhibited. Median values normalised to Mtb-infected samples of an experiment performed in triplicate are shown, and are representative of an experiment performed on two separate occasions.
Fig 3.
PI3Kδ expression is suppressed in human TB granulomas and infected macrophages.
A, B. Immunohistochemistry analysis of normal human lung tissue demonstrates that alveolar macrophages express both CD68 and PI3Kδ. C, D. Within human lung TB granulomas, CD68 is widely expressed, but no PI3Kδ immunoreactivity is observed. E, F. Both epithelioid macrophages and multinucleate giant cells around TB granulomas express CD68, but neither express PI3Kδ. G. In primary macrophages, Mtb infection significantly upregulates MMP-1 expression at 24h, whilst (H) concurrently suppresses PIK3CD mRNA accumulation. Immunohistochemistry images are representative of six donors with histopathologically confirmed TB who had lung biopsies as part of their routine clinical care. Scale bars: A, B, E, F: 100μm, C, D: 500 μm. Cellular experiments were performed in triplicate and are representative of at least three donors. P values are t-test comparisons.
Fig 4.
Negative regulation of MMP-1 is via the AKT and mTORC1 pathways, but the PI3K pathway does not cross talk with p38, COX-II or NFκB signalling.
A. Inhibition of AKT significantly increases MMP-1 secretion from macrophages at 72h. B. Inhibition of mTORC1 signalling with rapamycin similarly augments macrophage MMP-1 secretion at 72h. C. MMP-1 mRNA accumulation is increased in rapamycin-treated cells at 24h. D. Mtb increases p38 phosphorylation in macrophages analysed by western blotting, but PI3K pathway inhibition does not affect this at 30 minutes or 4h. E. PI3K inhibition does not augment constitutive ERK phosphorylation. F. PI3K inhibition has no effect on accumulation of COX-II protein in infected macrophages at 24h, analysed by flow cytometry. G. Mtb increases NFκB activation, but no difference is observed after PI3K inhibition with LY294002. Experiments have been performed on three occasions (A, B, D, E) and two occasions (C, F, G). Mean and standard deviations are shown and p values are by Student’s t-test.
Fig 5.
The MNK pathway is also a negative regulator of MMP-1 expression.
A. Chemical inhibition of the MNK pathway by MNK-I1 significantly increases MMP-1 secretion from macrophages at 72h. B. Increased MMP-1 secretion is secondary to increased mRNA accumulation at 24h. C. Bone marrow derived macrophages from wild type, MNK1-KO, MNK2-KO or double KO mice were uninfected or infected with Mtb at MOI 1. Supernatants were harvested at 24h, sterile filtered and MMP-3 concentration analyzed by Luminex array. MMP-3 secretion was significantly elevated in the MNK double KO cells. Representative experiment of two each performed in triplicate. D. MNK inhibition suppresses phosphorylation of eIF4E at 180 mins. E. Inhibition of the p90RSK pathway does not significantly change MMP-1 secretion at 72h. A, B: data represent experiments performed in triplicate on a minimum of two occasions, D: 3 separate donors, E: two donors in triplicate. Mean and standard deviation are shown and p values are Student’s t-test.
Fig 6.
Negative regulation of MMPs by MNK is more specific than PI3K/AKT/mTORC1 signalling.
Global analysis of MMP, cytokine, chemokine and growth factors secretion from macrophages at 72h by Luminex array demonstrates that MNK inhibition augments MMP-1, MMP-3 and MMP-10 secretion significantly, but conversely suppresses TH1, TH2 and chemokine secretion. Median values normalised to Mtb values of an experiment performed in triplicate are shown, and are representative of an experiment performed on two occasions.
Fig 7.
The PI3K and MNK pathways converge at the level of eIF4E phosphorylation.
A. Inhibition of eIF4F complex formation significantly suppresses MMP-1 secretion from Mtb-infected macrophages at 72h. B. Insulin, a PI3K pathway activator, increases macrophage eIF4E phosphorylation, while LY294002 abrogates phosphorylation. Mtb infection does not change basal eIF4E phosphorylation levels, but either MNK or PI3K/PD-1 inhibition suppresses eIF4E phosphorylation at 90 mins. A: representative data from an experiment performed in 2 donors in triplicate, B: performed on three occasions in different donors. Mean and standard deviation shown and p values are by Student’s t-test.
Fig 8.
Mtb infection reduces stability of negative regulatory pathway genes and increases microRNA accumulation.
A, B, C. Mtb infection of macrophages suppresses total PIK3CD, MLST8 and MKNK1 mRNA levels at 24h after infection. D. In contrast, synthesis of newly transcribed mRNA for each gene is increased at 24h. E. Mtb infection of macrophages upregulates multiple microRNAs that are predicted to target the negative regulatory pathways analyzed at 48h. F. miR-7 targeting the 3’UTR of PIK3CD suppresses luciferase activity. Hela cells were co-transfected with a miR-7 expressing vector (miR-7) and a Renilla Luciferase construct harbouring either a PIK3CD-3’UTR fragment containing the miR-7-5p binding site (bp 268–275) (Wild Type, WT), the same PIK3CD-3’UTR construct with mutated miR-7-5p binding site (bp 268–275) (Mutant), or empty construct (Control). RLA: Relative Luciferase Activity. Mean and standard deviation are shown, p values are by Student’s t-test and data are representative of experiments performed on at least two occasions in triplicate. F. Combined results from 4 independent experiments.
Fig 9.
Mtb inhibits negative regulatory pathways in macrophages.
Mtb upregulates MMP-1 via the MAPK pathways, whilst concurrently inducing microRNAs that target the negative regulatory pathways that limit MMP-1 expression, leading to excessive protease production and tissue destruction.