Monocyte activation by necrotic cells is promoted by mitochondrial proteins and formyl peptide receptors

Abstract

Objective: Necrotic cells evoke potent innate immune responses through unclear mechanisms. The mitochondrial fraction of the cell retains constituents of its bacterial ancestors, including N-formyl peptides, which are potentially immunogenic. Thus, we hypothesized that the mitochondrial fraction of the cell, particularly N-formyl peptides, contributes significantly to the activation of monocytes by necrotic cells.

Design: Human peripheral blood monocytes were incubated with necrotic cell fractions and mitochondrial proteins to investigate their potential for immune cell activation.

Setting: University Medical Center Research Laboratory.

Subjects: Healthy human adults served as blood donors.

Measurements and main results: Human blood monocyte activation was measured after treatment with cytosolic, nuclear and mitochondrial fractions of necrotic HepG2 cells or necrotic HepG2 cells depleted of N-formyl peptides [Rho(0) cells]. The specific role of the high affinity formyl peptide receptor (FPR) was then tested using specific pharmacologic inhibitors and RNA silencing. The capacity of mitochondrial N-formyl peptides to activate monocytes was confirmed using a synthetic peptide conforming to the N-terminus of mitochondrial nicotinamide adenine dinucleotide subunit 6. The results demonstrated that mitochondrial cell fractions most potently activated monocytes, and interleukin (IL)-8 was selectively released at low-protein concentrations. Mitochondria from Rho(0) cells induced minimal monocyte IL-8 release, and specific pharmacologic inhibitors and RNA-silencing confirmed that FPR contributes significantly to monocyte IL-8 responses to both necrotic cells and mitochondrial proteins. N-formyl peptides alone did not induce monocyte IL-8 release; whereas, the combination of mitochondrial N-formyl peptides and mitochondrial transcription factor A (TFAM) dramatically increased IL-8 release from monocytes. Likewise, high mobility group box 1, the nuclear homolog of TFAM, did not induce monocyte IL-8 release unless combined with mitochondrial N-formyl peptides.

Conclusions: Interactions between mitochondrial N-formyl peptides and FPR in the presence of other mitochondrial antigens (e.g., TFAM) contributes significantly to the activation of monocytes by necrotic cells.

Figure 1. Verification of Subcellular Fraction Purity

(A) Representative Western blot of subcellular fractions (30 μg) showed a lack of cytosolic contamination by mitochondrial proteins (HSP60, COXII) and an absence of β-actin in the mitochondrial fractions. Furthermore, the mitochondrial matrix fraction was free of mitochondrial membrane contamination, as reflected by the absence of COXII. The COXII control was derived from human heart mitochondria (MitoSciences, Inc.). Protein loading was normalized to GAPDH. (B) Representative Western blot examining CEACAM-1 and calnexin immunoreactivity confirmed that the mitochondrial and cytosolic fractions (30 μg) were not contaminated by plasma membrane or endoplasmic reticulum, respectively.

Figure 2. Verification of Recombinant Human TFAM and HMGB1 Proteins

(A) Stained protein gel of TFAM and HMGB1 (3μg). TFAM.myc.6×His and HMGB1.myc.6×His monomers were ∼29 kD. (B) Western blot of proteins (2 μg) using antibodies directed against TFAM and HMGB1. (C-D) Protein gels of TFAM and HMGB1 (2 μg) following incubation with annealed oligonucleotide binding probes [³²P-LSP (C) or ³²P-cruciform (D) DNA].

Figure 3. Necrotic Cells Release Mitochondrial Proteins and Promote Monocyte IL-8 Release

(A) Representative Western blot of cell supernatants derived from HepG2 cells exposed to heat, freeze/thaw or detergent (2% Triton X-100) demonstrating the release of mitochondrial (TFAM, ND1, ND6) and nuclear (HMGB1) proteins. (B) Significant monocyte IL-8 release was observed 6 hrs after treatment with equal supernatant volumes (50 μl) derived from freeze/thaw, or heat-treated HepG2 cells (∼10⁷ cells/ml) compared to matching untreated preparations (*p<0.05, compared to untreated Control). The results were compared to spontaneous release of IL-8 from monocytes in culture medium.

Figure 4. Monocyte Cytokine Responses to Treatment with Nuclear, Cytoplasmic and Mitochondrial Components of Necrotic Cells

(A) Six hrs after exposure to increasing titrations of HepG2 subcellular fractions, dose-dependent monocyte IL-8 release was limited to the mitochondrial compartment, particularly the membrane fraction. (B) Monocyte IL-8 release was comparable when treated with mitochondria derived from HepG2 cells or human liver homogenates. LPS (10 ng/ml) was used as a positive control.

Figure 5. Lack of Contribution of Mitochondrial DNA and Lipids to Monocyte Activation

Monocyte IL-8 release 6 hrs after treatment with mitochondrial lysates was unchanged when compared with matching preparations treated with CHAPS and benzonase to remove lipids and DNA, respectively (*p<0.05, compared to untreated Control; †p<0.05, relative to Control and corresponding lower dose treatment). CHAPS-treated mitochondrial preparations had a 33.1 ±1.8% reduction in total lipid content compared to untreated mitochondria (p<0.01). DNA content was reduced by 83.9 ±2.8% (p<0.01) after benzonase treatment.

Figure 6. Dose-response Relationship between Mitochondrial Membrane Proteins and Monocyte Cytokine Release

Human monocytes were treated with increasing titrations of mitochondrial membrane protein fractions derived from HepG2 cells. IL-8 and IL-6 release were measured 6 hrs after treatment; whereas, TNFα release was measured 3 hrs post-treatment. Significant dose-dependent IL-8 release was observed at much lower protein concentrations than those required for IL-6 release; TNFα release was not significantly increased at any concentration.

Figure 7. Treatment with Mitochondria Depleted of N-formyl Peptides Elicits Attenuated Monocyte IL-8 Release

Rho(0) HepG2 cells exhibited reduced mitochondrial DNA-encoded cytochrome c oxidase subunit 2 (COXII) expression (A), and had diminished capacity for electron transport (B). Compared to proteins isolated from normal HepG2 mitochondria, mitochondrial proteins derived from Rho(0) HepG2 cells (500 ng/ml) elicited reduced monocyte IL-8 release (C), and the same relationship was observed when whole cell lysates from normal and Rho(0) cells (5 μg/ml) were compared (not shown) (*p <0.01, compared to Normal HepG2 mitochondrial protein).

Figure 8. FPR Inhibition Significantly Reduces Monocyte IL-8 Release Induced by Cell Lysates or Mitochondrial Proteins

(A) The pharmacological inhibitor of FPR, CsH (3 μM), significantly decreased monocyte IL-8 release in response to treatment with necrotic HepG2 cell lysates (5 μg/ml) (*p<0.05, relative to Control; †p<0.05, compared to cell lysate alone). LPS (10 ng/ml) was used as the positive control. (B) Pretreatment with CsH (3 μM) or Boc-FLFLF (10 μM) attenuated monocyte IL-8 release in response to mitochondrial membrane proteins (50 ng/ml); whereas, the small G-protein inhibitor, pertussis toxin (1 μg/ml) had no significant effect (*p<0.05, relative to Control; †p<0.05, compared to mitochondrial membrane alone; ‡p<0.05, relative to Control and mitochondrial membrane alone).

Figure 9. FPR Plays a Significant Role in the Induction of Monocyte IL-8 Release by Mitochondrial Proteins

(A) RT-PCR (left) and real-time PCR (right) analyses demonstrated significant suppression of monocyte FPR mRNA after treatment with FPR siRNA; whereas, transfection with the non-silencing control siRNA (nc-siRNA) had no effect (*p<0.01, compared to matching Control). (B) Attenuation of monocyte IL-8 release following suppression of monocyte FPR mRNA in response to treatment with mitochondrial membrane proteins (500 ng/ml) but not to LPS (10 ng/ml) (*p<0.05, relative to matching Control; †p<0.05, compared to matching non-silencing treatment).

Figure 10. N-formyl Peptides Alone Are Insufficient to Induce Monocyte IL-8 Release

(A) Representative Western blot analysis showed TFAM to be concentrated in the mitochondrial membrane fraction of the cell (protein load: 30 μg). (B) Monocyte IL-8 responses to mitochondrial (fMMYALF) or bacterial (fMLP) N-formyl peptides (100 ng/ml) in the presence or absence of mitochondrial (TFAM) or nuclear (HMGB1) chromatin-binding proteins (5 μg/ml) and CsH (3 μM) treatment (*p<0.05, relative to Control; †p<0.05, compared to matching treatment with TFAM or HMGB1 alone; ‡p<0.05, relative to matching treatment in the absence of CsH). LPS (10 ng/ml) was used as a positive control.