Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Feb 23;9:306.
doi: 10.3389/fimmu.2018.00306. eCollection 2018.

Receptor-Interacting Protein Kinases 1 and 3, and Mixed Lineage Kinase Domain-Like Protein Are Activated by Sublytic Complement and Participate in Complement-Dependent Cytotoxicity

Affiliations
Free PMC article

Receptor-Interacting Protein Kinases 1 and 3, and Mixed Lineage Kinase Domain-Like Protein Are Activated by Sublytic Complement and Participate in Complement-Dependent Cytotoxicity

Michal Lusthaus et al. Front Immunol. .
Free PMC article

Abstract

The complement system participates in the pathogenesis of many diseases. Complement activation produces several active protein complexes and peptides, including the terminal C5b-9 complexes. It was reported that C5b-9 complexes insert into the plasma membrane and cause membrane perturbation, intracellular calcium surge, metabolic depletion, and osmotic lysis. Previously, we showed that complement-dependent cytotoxicity (CDC) is regulated by JNK and Bid. Here, we demonstrate that three mediators in TNFα-induced necroptosis (regulated necrosis), the receptor-interacting protein kinases, receptor-interacting protein kinase 1 (RIPK1) and receptor-interacting protein kinase 3 (RIPK3), and mixed-lineage kinase domain-like protein (MLKL), are activated by complement and contribute to CDC. Cell treatment with necrostatin-1 (Nec-1), a RIPK1 inhibitor, GSK'872, a RIPK3 inhibitor, or necrosulfonamide and GW806742X, MLKL inhibitors, restrain CDC. These findings were confirmed by using specific siRNAs targeting the synthesis of these proteins. Mouse fibroblasts lacking RIPK3 or MLKL were found to be less sensitive to C5b-9 than were wild-type (WT) fibroblasts. Enhanced CDC was achieved by RIPK1 or RIPK3 overexpression but not by the overexpression of a RHIM-RIPK1 mutant nor by a kinase-dead RIPK3 mutant. Nec-1 reduces the CDC of WT but not of RIPK3-knockout fibroblasts. Cells treated with a sublytic dose of complement exhibit co-localization of RIPK3 with RIPK1 in the cytoplasm and co-localization of RIPK3 and MLKL with C5b-9 at the plasma membrane. Data supporting cooperation among the RIP kinases, MLKL, JNK, and Bid in CDC are presented. These results provide a deeper insight into the cell death process activated by complement and identify potential points of cross talk between complement and other inducers of inflammation and regulated necrosis.

Keywords: C5b-9; complement; mixed lineage kinase domain-like protein; receptor-interacting protein kinase 1; receptor-interacting protein kinase 3; regulated necrosis.

Figures

Figure 1
Figure 1
Complement C5b-9 induces receptor-interacting protein kinase 1 (RIPK1)-dependent necrosis. (A) K562, HT-29, or BT474 cells were treated with necrostatin-1 (Nec-1) or with DMSO (0) as control for 1 h at 37°C. Cell death (CD) by antibody (30 min at 4°C) and complement (1 h at 37°C) was performed as described under Section “Materials and Methods.” The experiment with K562 cells was performed with two antibody (Ab) dilutions. The percentage of CD was analyzed by propidium iodide inclusion. Results of three independent experiments are expressed as the mean percentage of CD ± SD. The percentage of CD by Nec-1, antibody, and HIS was 3–7% (negative controls). Statistical analysis showed that Nec-1 significantly inhibited CD (one-way-ANOVA, P < 0.01). (B) K562 cells transfected for 48 h with RIPK1-shRNA, a scrambled (Sc) shRNA, or not transfected (NT) were treated with antibody and complement and the percentage of CD was determined as described in panel (A). Transfection with RIPK1-shRNA significantly inhibited the CD relative to the scrambled shRNA (t-test, P < 0.01). The RIPK1 expression level (relative to actin) in the transfected cells was assessed by Western blotting analysis. A representative blot is shown. (C) K562 cells were transfected with the GFP-tagged plasmids: wild-type RIPK1 (RIPK1-wt), RHIM-ALAA mutant (RIPK1-mut), an empty vector control (C), or were NT. After 52 h, the cells were treated with antibody and complement and the CD was determined as above. Transfection with RIPK1-wt significantly enhanced the CD relative to empty vector and RIPK1-mut (t-test, P < 0.01). Transfection efficiency was determined by analyzing the GFP expression by flow cytometry. A representative histogram is shown. All results represent three independent experiments.
Figure 2
Figure 2
Receptor-interacting protein kinase 3 (RIPK3) contributes to complement-induced cell death (CD). (A) K562 cells were pretreated with GSK’872 or with DMSO as control for 1 h at 37°C and then treated with antibody and complement. The percentage of CD was analyzed by propidium iodide inclusion. Statistical analysis showed that GSK’872 significantly inhibited CD (one-way-ANOVA, P < 0.01). (B) K562 cells were transfected with a RIPK3-shRNA plasmid or an empty vector (C) as control. After 48 h, the cells were treated with antibody and complement and CD was quantified. Transfection with RIPK3-shRNA significantly reduced CD relative to empty vector (t-test, P < 0.01). RIPK3 knockdown was confirmed by Western blotting analysis of cell lysates (shown on the right). (C) K562 cells were transfected with plasmid DNA expressing Flag-tagged wild-type (WT) RIPK3 (RIPK3-wt), RIPK3 kinase-dead mutant (RIPK3-mut), or with its empty vector (C). After 30 h, the cells were treated with antibody and complement and CD was determined. Transfection with RIPK3-wt significantly enhanced CD relative to empty vector and RIPK3-mut (t-test, P < 0.05). To confirm the expression of the flagged proteins, cell lysates were subjected to Western blotting analysis with anti-Flag and anti-actin antibodies (shown on the right). (D) WT and RIPK3 KO mouse embryonic fibroblasts (MEFs) were treated with normal human serum (NHS) (20, 25, or 30%) for 1 h at 37°C. The percentage of CD was determined by trypan blue exclusion. Statistical analysis showed that RIPK3 KO MEF is significantly less sensitive to complement-dependent cytotoxicity than is WT MEF (two-way-ANOVA, P < 0.001). Yet, the two cell types reacted similarly to increasing concentrations of complement (two-way-ANOVA, P = 0.148). Lysates of WT and RIPK3 KO MEFs were analyzed by Western blotting with anti-RIPK3 and anti-actin antibodies (shown on the right). Results are expressed as mean percentage of CD ± SD. Data shown represent at least three independent experiments. NT, not transfected cells.
Figure 3
Figure 3
Complement triggers receptor-interacting protein kinase 1 (RIPK1) and receptor-interacting protein kinase 3 (RIPK3) complex formation. (A) Flag-RIPK3 transfected K562 cells were treated with a sublytic dose of antibody and complement for the indicated times. Cell lysates were immunoprecipitated with anti-RIPK1 antibody. Immunocomplexes were subjected to Western blotting with anti-RIPK1 or anti-Flag antibodies. Aliquots of cell lysates (Input lysate) were also analyzed for the amount of actin. (B) Flag-RIPK3 transfected K562 cells were treated with a sublytic dose of antibody and complement for 10 min at 37°C. Cell lysates were immunoprecipitated with anti-RIPK1 antibody or with goat IgG as control. Controls included cells treated with HIS (heat-inactivated serum) or untreated cells [not transfected (NT)]. RIPK1, RIPK3, and actin were detected as above. (C) Flag-RIPK3 K562 cells were treated with a sublytic dose of antibody and then with complement (or HIS) for 10 min at 37°C. Cells were fixed and permeabilized. RIPK3 was detected by using anti-Flag antibody followed by AF488-labeled secondary antibody. Endogenous RIPK1 was labeled with anti-RIPK1 antibody followed by AF546-labeled secondary antibody. After analysis by confocal microscopy, the RIPK1 and RIPK3 locations were merged by the Co-localization plugin in ImageJ software. (D) Flag-RIPK3 K562 cells were treated for 5 min with a sublytic dose of antibody and complement. Then, they were fixed and permeabilized. Prior to imaging, cells were treated with goat anti-RIPK1 or mouse anti-FLAG antibodies, followed by AF546-labeled donkey anti-goat IgG or AF488-labeled donkey anti-mouse IgG secondary antibodies, respectively. Square ROI was photobleached and images were taken before and after bleaching. Shown are fluorescence resonance energy transfer (FRET) efficiencies normalized to the bleaching percentage of the acceptor. Control (C) is the FRET efficiency outside the bleached regions [in normal human serum (NHS)-treated cells] *P < 0.05 relative to HIS. (E) WT and RIPK3 KO mouse embryonic fibroblasts were pretreated with Nec-1 (50 µM) or with DMSO as control and then treated with NHS for 1 h at 37°C. The percentage of cell death was determined by trypan blue inclusion. Data were represented as mean ± SD of triplicates. **P < 0.01 relative to DMSO. All experiments were repeated at least three times with similar results.
Figure 4
Figure 4
Mixed lineage kinase domain-like protein (MLKL) contributes to complement-induced cell death (CD). (A,B) K562 (A) and HT-29 cells (B) were incubated with necrosulfonamide (NSA) or with DMSO as control for 1 h at 37°C and subsequently with antibody and normal human serum (NHS). Necrotic CD was determined by propidium iodide (PI) inclusion. Results, representing three independent experiments, are expressed as the mean percentage of CD ± SD. Differences between GSK’872 and DMSO (0) treatments were all significant (P < 0.05). (C) K562 cells were transfected with Smart Pool siRNA targeting MLKL or scrambled siRNA as a control, by electroporation. After 48 h, the cells were treated with antibody and complement for 1 h at 37°C. CD was measured by PI staining and expressed as the mean percentage of CD ± SD and the results represent three independent experiments. *P < 0.001 relative to control. Cell lysates, collected 48 h posttransfection, were subjected to Western blotting with anti-MLKL or anti-actin antibodies (shown on the right). (D) WT and MLKL KO mouse embryonic fibroblasts were treated with NHS (15 or 25%) for 1 h at 37°C. The CD percentage was measured by trypan blue inclusion. Results are the mean percentage of death ± SD. *P < 0.001 relative to WT control. Cytoplasmic aliquots were subjected to Western blot analysis of MLKL and actin levels (shown on the right).
Figure 5
Figure 5
Mixed lineage kinase domain-like protein (MLKL) cooperates with RIPK3 during complement-dependent cytotoxicity (CDC) but not with receptor-interacting protein kinase 1 (RIPK1). (A) K562 cells were pretreated with GSK’872 and/or necrosulfonamide (NSA) in the indicated concentrations or with DMSO as control (0), for 1 h at 37°C, and then treated with antibody and complement for 1 h min at 37°C. The percentage of cell death (CD) was measured by propidium iodide (PI) inclusion. Results are expressed as the mean percentage of the inhibition of CD ± SD and represent three independent experiments. *P < 0.01 relative to pretreatment with 0.25 µM GSK’872 without NSA or to 2 µM NSA without GSK’872. [% Inhibition = (% Death of Control − % Death of Treatment)/%Death of Control × 100]. (B) K562 cells were transfected with Smart Pool siRNA targeting MLKL (siRNA) or scrambled siRNA (Sc) as a control by electroporation. After 24 h, the cells were treated with GSK’872 at the indicated concentrations (or DMSO as control, 0) and then with antibody and complement for 1 h at 37°C. CD was measured as above. Statistical analysis showed that CDC of MLKL knocked down cells is significantly less sensitive to GSK’872 than is CDC of cells treated with a scrambled siRNA (two-way-ANOVA, P < 0.001). (C) WT and MLKL KO mouse embryonic fibroblasts (MEFs) were treated with necrostatin-1 (Nec-1s), a selective RIPK1 inhibitor, in increasing concentrations, or with DMSO as control, for 1 h followed by incubation with 10% normal human serum (NHS). The lysis percentage was determined by PI inclusion. Results are presented as the mean percentage of lysis ± SD. Statistical analysis showed that MLKL KO MEF is as sensitive to CDC as WT MEF (two-way-ANOVA, P = 0.538). (D) MLKL-silenced K562 cells were treated with Nec-1s at the indicated concentrations and then with antibody diluted 1:18 or 1:15 for scrambled or siRNA-treated cells, respectively, followed by complement (NHS, 50%). CD was measured by PI staining, expressed as a calculated mean percentage ± SD, representing three independent experiments. Statistical analysis showed that CDC of MLKL knocked down cells is as sensitive to Nec-1 as CDC of cells treated with a scrambled siRNA (two-way-ANOVA, P = 0.104).
Figure 6
Figure 6
C5b-9 deposition triggers mixed lineage kinase domain-like protein (MLKL) translocation to the plasma membrane and C5b-9-MLKL interaction. (A) K562 cells were treated with a sublytic dose of antibody and with complement [normal human serum (NHS)] or HIS for 10 min at 37°C. After fixation and permeabilization, MLKL was stained with anti-MLKL antibody and AF546-labeled secondary antibody. Representative confocal microscope images are presented. NT, non-treated cells. (B) K562 cells were treated with antibody and complement as above. In addition to MLKL labeling, C5b-9 was detected with an anti-C5b-9 antibody (AE-11) and with AF488-labeled secondary antibody. MLKL and C5b-9 locations were merged by the Co-localization plugin in ImageJ software. Representative cells (from three independent experiments) show the co-localization of MLKL with C5b-9 in the plasma membrane region. (C–E) K562 cells were treated with complement (NHS) as above and cell lysates were prepared. Controls included HIS and C8-depleted human serum (C8D). (C) Cell lysates were subjected to immunoprecipitation with anti-C9 antibody followed by Western blot analysis with anti-C9 or anti-MLKL antibodies. (D) MLKL was immunoprecipitated from cell lysates with an anti-MLKL antibody. MLKL was detected by Western blotting. MLKL and actin were also quantified in whole-cell lysates (Lysates). (E) Proteins immunoprecipitated from cell lysates with anti-MLKL antibodies, as in (D), were eluted from the beads and the amount of C5b-9 was quantified by ELISA. Results are expressed as adsorption at 450 nm (A450). *P < 0.05 relative to C8D or HIS treated cells (t-test).
Figure 7
Figure 7
C5b-9 complex co-localizes with receptor-interacting protein kinase 1 (RIPK1) and receptor-interacting protein kinase 3 (RIPK3). (A,B) K562 cells expressing Flag-RIPK3 were treated with a sublytic dose of antibody and with C9D-normal human serum (NHS) supplemented with AF546-labeled human C9 (A) or with NHS (B) for 10 min at 37°C. (C,D) Same cells as in panel (A,B), respectively, were then washed and incubated for an additional 20 min in HBSS at 37°C. RIPK3 was detected with anti-Flag antibody and secondary AF488-labeled antibody. C5b-9 was detected with anti-C5b-9 antibody (AE-11) and with secondary AF488-labeled antibody. RIPK1 was labeled with anti-RIPK1 antibody and with AF546-labeled secondary antibody. Representative cells from four independent experiments are shown. Merging of RIPK1 and RIPK3 with C5b-9 locations was performed by the co-localization plugin in ImageJ software.
Figure 8
Figure 8
Bid and JNK are involved in receptor-interacting protein kinase 1/receptor-interacting protein kinase 3 (RIPK3)/mixed lineage kinase domain-like protein (MLKL) signaling, leading to complement-induced necrosis. General protocol: cells were pretreated with the indicated inhibitor for 1 h at 37°C and then treated with NHS for 1 h, at 37°C. The percentage of necrotic cell death (CD) and the percentage of CD inhibition by the inhibitor, relative to DMSO, were calculated (after data conversion to y/1 − y) as explained in Section “Materials and Methods.” *P < 0.01 relative to DMSO (A,C,D,F) or between WT and KO cells (B,D) (t-test). (A) WT, JNK1, or JNK2 mouse embryonic fibroblasts (MEFs) were pretreated with necrostatin-1 (Nec-1) (50 µM) or DMSO as control. (B) WT and RIPK3 KO MEFs were pretreated with SP600125 or DMSO as control. (C) WT and MLKL KO MEFs were treated with 10 µM SP600125 or DMSO. (D) WT or Bid KO MEFs were treated with Nec-1 or DMSO. (E) WT or Bid KO MEFs were pretreated with GSK’872 at 2 or 5 µM or with DMSO as control. (F) WT or Bid KO MEFs were treated with 1 µM GW806742X or DMSO. All results represent at least three independent experiments.

Similar articles

See all similar articles

Cited by 5 articles

References

    1. Walport MJ. Complement. First of two parts. N Engl J Med (2001) 344:1058–66.10.1056/NEJM200104053441406 - DOI - PubMed
    1. Ricklin D, Hajishengallis G, Yang K, Lambris JD. Complement: a key system for immune surveillance and homeostasis. Nat Immunol (2010) 11:785–97.10.1038/ni.1923 - DOI - PMC - PubMed
    1. Merle NS, Noe R, Halbwachs-Mecarelli L, Fremeaux-Bacchi V, Roumenina LT. Complement system part II: role in immunity. Front Immunol (2015) 6:257.10.3389/fimmu.2015.00257 - DOI - PMC - PubMed
    1. Muller-Eberhard HJ. The membrane attack complex of complement. Annu Rev Immunol (1986) 4:503–28.10.1146/annurev.iy.04.040186.002443 - DOI - PubMed
    1. Morgan BP, Luzio JP, Campbell AK. Intracellular Ca2+ and cell injury: a paradoxical role of Ca2+ in complement membrane attack. Cell Calcium (1986) 7:399–411.10.1016/0143-4160(86)90042-4 - DOI - PubMed

Publication types

MeSH terms

Substances

Feedback