PARK2 mediates interleukin 6 and monocyte chemoattractant protein 1 production by human macrophages

Abstract

Leprosy is a persistent infectious disease caused by Mycobacterium leprae that still affects over 200,000 new patients annually. The host genetic background is an important risk factor for leprosy susceptibility and the PARK2 gene is a replicated leprosy susceptibility candidate gene. The protein product of PARK2, Parkin, is an E3 ubiquitin ligase that is involved in the development of various forms of Parkinsonism. The human macrophage is both a natural host cell of M. leprae as well as a primary mediator of natural immune defenses, in part by secreting important pro-inflammatory cytokines and chemokines. Here, we report that down-regulation of Parkin in THP-1 macrophages, human monocyte-derived macrophages and human Schwann cells resulted in a consistent and specific decrease in interleukin-6 (IL-6) and monocyte chemoattractant protein 1 (MCP-1/CCL2) production in response to mycobacteria or LPS. Interestingly, production of IL-6 at 6 hours by THP-1 cells stimulated with live M. leprae and M. bovis BCG was dependent on pretreatment with 1,25-dihydroxyvitamin D(3) (VD). Parkin knockdown in VD-treated cells blocked IL-6 induction by mycobacteria. However, IκB-α phosphorylation and levels of IκB-ξ, a nuclear protein required for IL-6 expression, were not affected by Parkin silencing. Phosphorylation of MAPK ERK1/2 and p38 was unaffected by Parkin silencing while JNK activation was promoted but did not explain the altered cytokine production. In a final set of experiments we found that genetic risk factors of leprosy located in the PARK2 promoter region were significantly correlated with M. leprae sonicate triggered CCL2 and IL6 transcript levels in whole blood assays. These results associated genetically controlled changes in the production of MCP-1/CCL2 and IL-6 with known leprosy susceptibility factors.

Figure 1. Parkin-silenced THP-1 macrophages cytokine screen.

(A) Parkin was detected by indirect immunofluorescence of THP-1 cells following transfection with either scrambled siRNA (upper panel) or siRNA targeting Parkin (lower panel). Insets represent DAPI-stained nuclei. (B) PMA-differentiated THP-1 macrophages were transfected with control or Parkin-silencing siRNA. After 48 hours, cells were treated with H37Ra at an MOI of 10, M. leprae (ML) at an MOI 50, or left untreated (Neg.). After 6 hours, supernatants were collected and analyzed with a Milliplex 42-cytokine assay. Cytokines with detectable values (12 out of 42) are plotted on the graph. Cytokine production is expressed as ratio of cytokine secreted by cells transfected with siRNA for PARK2 (Parkin) to cytokine secreted by cells transfected with control siRNA (scrambled). (C) PMA- differentiated THP-1 macrophages were transfected with control or Parkin-silencing siRNA. After 48 hours, cells were treated with LPS (10 ng/ml), M. bovis BCG at an MOI of 10, M. leprae (ML) at an MOI 50, or left untreated (Neg). After 6 hours supernatants were collected and analyzed with a Q-Plex custom cytokine multiplex assay. Values represent the ratio of concentrations produced by Parkin-silenced cells over controls ± SD of at least three independent experiments. (D) As described for C except that supernatants were collected after 24 hrs incubation with stimulants. * p<0.05, non-parametric t test of unpaired samples.

Figure 2. Parkin knockdown inhibits IL-6 and MCP-1 induction in macrophages and Schwann cells.

PMA-differentiated THP-1 macrophages, VD-treated THP-1 macrophages, VD-treated human monocyte-derived macrophages and human Schwann cells were transfected with control or Parkin-silencing siRNA. Due to the limited number of cells available for each experiment, only one PARK2 siRNA was used for MDM. After 48 hours, cells were treated with LPS (10 ng/ml), M. bovis BCG at an MOI of 10, M. leprae (ML) at an MOI 50, or left untreated (Neg.). Supernatants were collected after 6 hours and analyzed for (A) IL-6 and (B) MCP-1. The charts show the average concentration of cytokine in the supernatant in pg/ml ± SD of at least three independent experiments for each cell type. Scrambled: control siRNA, Parkin A: siRNA(A) for Parkin, Parkin B: siRNA(B) for Parkin. * p<0.05, ** p<0.01, non-parametric t test of unpaired samples.

Figure 3. Parkin knockdown effect on NF-κB and MAPK signaling.

(A) siRNA-transfected THP-1 macrophages were stimulated with 100 ng/ml LPS for the indicated times, then lysed and analysed by 12% SDS-PAGE followed by Western blotting. Membranes were stained for phospho-IκB-α then reprobed for IκB-ζ and β-actin (B) Nuclear extracts were also analyzed by transcription factor ELISA. THP-1 cells were treated with 100 ng/ml LPS for 4 hours, then extracted and analysed for NF-κB (p65) and AP-1 (phospho-c-Jun) binding to consensus DNA oligomers. Values are expressed as % binding of positive standard extracts and are representative of three experiments. Extracts from diluent (RPMI)-treated cells did not show DNA binding above background. (C) Membranes were stained for phospho-ERK1/2, phospho-p38 and phospho-JNK then reprobed for total JNK and β-actin. The JNK2 (p54) isoform is predominantly detected in these lysates although some phospho-JNK1 (p46) bands can be observed. Blots are representative of at least three experiments. (D) Transfected cells were pretreated with 30 µM JNK inhibitor SP600125 (SP) for 1 hour and then treated with LPS or diluent for 6 hours. IL-6 from the supernatants was measured by ELISA.

Figure 4. Whisker plots of CCL2 and IL6 transcript levels in whole blood cultures in the presence an absence of M. leprae sonicate.

Whole blood from 62 Vietnamese subjects was stimulated with 10 µg/ml M. leprae sonicate and transcript levels of CCL2 and IL-6 were determined by real time PCR. (A) Transcript levels were normalized with the HPRT house keeping gene and expressed as ΔC_t in the absence (NON-STIM) and presence (STIM) of M. leprae sonicate. The median of the distribution is indicated by a solid line within the box. The resulting subdivision of the box indicates the distribution of the flanking 25% percentile in each direction while the error bars give the distribution of the upper and lower 25% of the ΔC_t values. (B) The increase of CCL2 and IL6 transcripts resulting from stimulation with M. leprae sonicate expressed as ΔΔC_t. Plots as described in A.