Hiroshi Nishiura and Rui Zhao contributed equally to this work.
Involvement of regional neutrophil apoptosis promotion by ribosomal protein S19 oligomers in resolution of experimental acute inflammation
Version of Record online: 15 JAN 2014
© 2013 The Authors. Pathology International © 2013 Japanese Society of Pathology and Wiley Publishing Asia Pty Ltd
Volume 63, Issue 12, pages 581–590, December 2013
How to Cite
Nishiura, H., Zhao, R., Chen, J., Taniguchi, K. and Yamamoto, T. (2013), Involvement of regional neutrophil apoptosis promotion by ribosomal protein S19 oligomers in resolution of experimental acute inflammation. Pathology International, 63: 581–590. doi: 10.1111/pin.12115
Conflict of Interest: The authors declare that they have no conflict of interest.
- Issue online: 15 JAN 2014
- Version of Record online: 15 JAN 2014
- Manuscript Accepted: 31 OCT 2013
- Manuscript Received: 21 AUG 2013
- Ministry of Education, Culture, Sports, Science, and Technology, Japan. Grant Number: 21590441
- carrageenan pleurisy;
- C5a receptor;
- extra-ribosomal function;
- inflammation resolution;
- leukocyte chemotaxis;
- ribosomal protein S19
Isolated peripheral neutrophils spontaneously underwent apoptosis in association with extra-cellular liberation of the monocyte-attracting ribosomal protein S19 (RP S19) oligomers. This apoptosis was prevented by the simultaneous presence of anti-RP S19 antibodies or of a C5a receptor antagonist, but was promoted by supplementing extrinsic RP S19 oligomers. Transformed HL-60 cells to over-produce Gln137Asn-mutant RP S19 were differentiated to neutrophil-like cells. The neutrophil-like cells gained resistance against the spontaneous apoptosis concomitant with the generation of non-functional RP S19 oligomers. When the neutrophil-like cells were intradermally transplanted into mice, the mutant RP S19-producing neutrophils persisted for a long period of time, whereas wild-type RP S19-producing neutrophils underwent apoptosis and were promptly cleared by infiltrated macrophages. When an experimental pleurisy was introduced by injecting carrageenan into the pleural cavity of mice, the inflammation spread slightly to lung parenchyma. When antibodies neutralizing the RP S19 oligomers were simultaneously administrated with carrageenan, the neutrophil infiltration in the lung parenchymal lesion become more severe, occurring as alveolar septal destruction and hemorrhage concomitant with an augmented neutrophil number in the pleural exudate. These results indicate the importance of the RP S19 oligomers and the C5a receptor in neutrophil clearance and acute inflammation resolution.
Ribosomal protein S19 (RP S19) is a component of the ribosome translation machinery. RP S19 appears to be essential for ribosome biogenesis under the steady state cellular condition. During apoptosis, RP S19 is intermolecularly cross-linked between Gln137 and Lys122 by a cellular transglutaminase(s) and extracellularly released.[3-5] Once RP S19 is oligomerized, it is able to ligate the C5a receptor. Cells undergoing apoptosis de novo synthesize the C5a receptor. The RP S19 oligomers promote apoptosis in an autocrine fashion by ligating the C5a receptor on the cells undergoing apoptosis.[5, 6] Additionally, the RP S19 oligomers recruit macrophages in a paracrine fashion as a chemotactic ligand of their C5a receptor and let them engulf the apoptotic cells.[3, 7, 8] Thus the RP S19 oligomers are responsible for synchronizing apoptosis execution and phagocyte recruitment. The RP S19 oligomers also ligate the C5a receptor of neutrophils, but they express an antagonist-induced effect in this case.
It is well documented that neutrophil apoptosis and subsequent engulfment by macrophages infiltrated later are the most significant phenomena at the resolution stage of acute inflammation.[10-13] However, the molecular mechanism that underlies the macrophage infiltration to phagocytically clear the apoptotic neutrophils is poorly understood. It is also emphasized that neutrophils are programmed to spontaneously undergo apoptosis to prevent undesired tissue destruction at the terminal stage of acute inflammation.[8, 14] However, the underlying molecular mechanisms are again mostly unclear. Recently, promotion of acute inflammation resolution by switching the production of lipid chemical mediators from the pro-inflammatory, such as prostaglandins and leukotrienes, to the anti-inflammatory, such as resolvins and protectins, was demonstrated. Resolvins and protectins initiate and accelerate neutrophil apoptosis in addition to the promotion of anti-inflammatory cytokine release. Getting a hint from this, we made the hypothesis that the RP S19 oligomers would be liberated from neutrophils undergoing apoptosis, and then the oligomers would accelerate neutrophil apoptosis and would recruit phagocytic macrophages, resulting in acute inflammation resolution without undesirable tissue destruction.
Materials and Methods
Chemicals and antibodies
Dulbecco's Modified Eagle Medium (DMEM) and fetal bovine serum (FBS) were obtained from Gibco BRC (Paisley, UK). The ECL Plus Western Blotting Detection System was obtained from Amersham Biosciences KK (Tokyo, Japan), and the multiwell chambers from Neuro Probe (Bethesda, MD, USA). Nucleopore filters were from Nucleopore (Pleasant, CA, USA). Immobilor Transfer Membrane was from Millipore (Billerica, MA, USA). Block Ace was from Dainippon Pharmaceutical (Suita, Japan). All other chemicals were from Nacalai Tesque (Kyoto, Japan) or from Wako Pure Chemicals (Osaka, Japan) unless otherwise specified. A C5a receptor antagonist Ac-Phe-[Orn-Pro-dCha-Trp-Arg] (PMX-53) was prepared according to Woodruff et al. Recombinant C5a and recombinant RP S19 oligomer, and anti-human RP S19 rabbit IgG or anti-human C5a rabbit IgG were prepared as described previously. A rat anti-F4/80 antibody (BM8-biotin) was obtained from Abcam (Kenbridge, UK).
Specific pathogen-free Crij:CD1(ICR) strain mice (15–20 g body weight) were purchased from Charles River (Yokohama, Japan). Animal experiments were performed under the control of the Ethical Committee for Animal Experiments, Kumamoto University School of Medicine. Mice were euthanased while under ether anesthesia.
Transformed HL-60 cells and HL-60-derived neutrophil-like cells
HL-60 cells were obtained from the Riken BioResource Center (Tsukuba, Japan). Transformed sub-lines of HL-60 cells which over-produce wild type RP S19 or Gln137Asn-RP S19, and their mock transformant were prepared using the pCAGGS-IRES neomycin-resistant vector constructs as described previously. The transformed HL-60 cells were differentiated into neutrophil phenotypes in the presence of 1.25% dimethyl sulfoxide (DMSO) for 7 days. The cells (5 × 105 cells/mL) were then re-cultured for another 3 days in the presence of 1.25% DMSO.
Peripheral blood was obtained into a heparinized syringe from healthy volunteers after getting their informed consent under approval of Institutional Review Board of Kumamoto University (approval number 673). Mononuclear cells and neutrophils were isolated from the heparinized blood using Ficoll-Paque Plus (Amersham Biosciences). The mononuclear cell fraction contained approximately 20% monocytes when identified as the macrophage specific esterase positive cells, and the neutrophil fraction contained nearly 100% neutrophils when morphologically identified with Wright-Giemsa staining.
Induction of carrageenan pleurisy
Carrageenan pleurisy was induced according to the method of Murai et al. with some modifications. In brief, 100 μL of 0.5% carrageenan in distilled water was mixed with anti-human C5a rabbit IgG or control IgG (100 μg IgG in 100 μL phosphate-buffered saline (PBS)), and immediately injected into the left pleural cavity of mice using a disposable syringe with a 27-gauge needle.
Phosphatidylserine-exposed cell analysis
Neutrophils were maintained at 5 × 106 cells/10 mL/100 mm diameter dish in the presence of either 10−8 M C5a, 10−8 M RP S19 oligomers, 10−6 M C5a receptor antagonist or control PBS. Aliquots of the cells were washed, resuspended in 1 mL of cold PBS containing 1% FBS and 2.5 mM CaCl2 (FACS medium), and incubated with fluorescein isothiocyanate (FITC)-conjugated annexin V and propidium iodide (PI) (MBL, Nagoya, Japan) for 30 min at 4°C. Intensities of FITC on the cells and PI in the nucleus were analyzed using a FACS Caliber flow cytometer (BD, Tokyo, Japan). The apoptotic cell ratio was expressed as percent of FITC-annexin V-positive cells which exposed phosphatidylserine on the surface.
Cell cycle analysis
Aliquots of the HL-60 neutrophil phenotypes were sampled at 0, 12 or 24 h after maturation and immediately fixed in 70% ethanol for 3 h at −20°C. The cells were then stained with PI (20 μg/mL) in the presence of RNase A (20 μg/mL) for 30 min at 37°C. Percentage of the cells at G1 subset, G0/1, S or G2/M phases was respectively analyzed using the FACS cytometer with Multicycle Cell Cycle Software. The cells at G1 subset phase were considered as apoptotic cells.
Leukocyte analyses in pleural exudate
After removal of the liquid fraction of the pleural exudate by centrifugation, red blood cells were ruptured in 0.83% NH4Cl, and the total leukocyte number was counted using a microscopic cell calculator plate or using the FACS cytometer. Neutrophils and monocytes were analyzed by FACS on the bases of the forward scatter and side scatter values.
Hematoxylin and eosin staining
Skin lesions or the lungs were immediately fixed in 10% formalin for preparation of paraffin sections. Paraffin sections with 4 μm thickness were stained with hematoxylin and eosin in the usual way.
Double-stranded DNA fragments in the nucleus were stained with acridine orange and observed using a fluorescence microscope (BX-53; Olympus, Tokyo, Japan) with a Penguin 600CL digital camera (Pixera, Los Gatos, CA, USA) (excitation at 480–490 nm and emission at 520–530 nm). Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) method was also utilized with an ApoMark DNA Fragmentation Detection Kit (Funakoshi, Tokyo, Japan).
Macrophages were stained with the biotinylated anti-mouse F4/80 antibody, and were visualised by the avidin-biotin-peroxidase method with diaminobenzidine. The histologic specimens were observed under a microscope, Provis AX (Olympus) with a digital camera, Penguin 600CL.
Western blotting analysis
Proteins separated by 12% polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate (SDS-PAGE) were transferred to a membrane using a Semi Dry Electroblotter (Sartorious) for 90 min under electric current of 15 V.[22, 23] After treatment with 4% Block Ace, the membrane was incubated with rabbit IgG against RP S19 (100 ng/mL in PBS containing 0.03%Tween 20) for 1 h, and with HRP-conjugated anti-rabbit IgG (20 ng/mL) (Santa Cruz, Santa Cruz, CA, USA) for 30 min at 22°C, in this order. The ECL Plus Western Blotting Detection System displayed the antigen signal.
Monocytes (1 × 106 cells/mL) in DMEM containing 10% FBS were prepared for the multiwell chamber assay using a nucleopore filter with a pore size of 5 μm. After incubation for 90 min, the membrane was separated, fixed with 100% methanol for 1 min, and stained with Giemsa solution for 20 min. The total number of cells that migrated beyond the lower surface of the membrane was counted in five separate high-power microscopic fields. The results are expressed as the number of migrated cells.
Statistical significance was calculated by non-parametric or parametric tests offered in a two-way analysis of variance window (P < 0.05: * and P < 0.01: **).
Spontaneous generation of RP S19 oligomers by isolated peripheral neutrophils
The mature neutrophils are programmed to undergo apoptosis.[8, 14] We examined whether isolated neutrophils generated the RP S19 oligomers in association to the spontaneous apoptosis. We separated peripheral neutrophils, cultured these cells in PBS, and observed appearance of the intracellular and extracellular RP S19 oligomers (represented by dimer in this experiment) using Western blotting method with anti-RP S19 IgG. As shown in Fig. 1a, the neutrophils intracellularly generated the RP S19 oligomers at 12 h, and then extracellularly released the oligomers at 18 h after the isolation. We confirmed the monocyte chemoattraction capacity due to the C5a receptor ligation of the RP S19 oligomers liberated in the supernatant using the chemotaxis chamber method. As shown in Fig. 1b, the culture supernatant at 18 h exhibited monocyte chemotactic capacity which was inhibited by anti-RP S19 IgG or by the C5a receptor antagonist.
Proapoptotic effect of RP S19 oligomers on isolated peripheral neutrophils
We next examined whether the RP S19 oligomers liberated by the isolated neutrophils involved in the apoptosis progression of the neutrophils of their own. As shown in Fig. 2a, the simultaneous presence of the anti-RP S19 IgG or the C5a receptor antagonist inhibited the apoptosis of the neutrophils observed at 24 h after the isolation when judged by means of the FITC-annexin V-binding to the cell surface phosphatidylserine.
We then examined the effect of extrinsic RP S19 oligomer administration comparatively with the C5a receptor antagonist or C5a administration on the apoptosis progression by 24 h after the neutrophil isolation. As shown in Fig. 2b, the simultaneous presence of the recombinant RP S19 oligomers supplement augmented the apoptosis when evaluated by the annexin V-binding cell ratio. The C5a receptor antagonist again inhibited the neutrophil apoptosis. We confirm the apoptosis augmentation effect of the supplemented RP S19 oligomers by FACS analysis (Fig. 2c) and by cytochemistry using TUNEL method (Fig. 2d). In this case, a protecting tendency of the supplemented C5a for the neutrophils from apoptosis was observed in addition to the apoptosis promotion by the RP S19 oligomers.
Gain of resistance against spontaneous apoptosis by Gln137Asn-RP S19 mutant production in neutrophil-like cells
The functional RP S19 oligomers possess an inter-molecular isopeptide bond between Lys122 and Gln137. We have previously reported that the Gln137Asn-RP S19 mutant was incapable of generating the functional RP S19 oligomers, thus the extra-ribosomal function of RP S19 was greatly reduced.[2, 21] We, therefore, attempted to prepare Gln137Asn-RP S19-producing neutrophils. We transformed HL-60 cells to excessively produce Gln137Asn-RP S19, and differentiated the cells to neutrophil phenotypes by culturing for 7 days with DMSO. We also prepared wild type-RP S19 high-producing HL-60-derived neutrophil phenotypes to further evaluate the proapoptotic function of the RP S19 oligomers. Mock-transformed HL-60-derived neutrophil phenotypes were a control.
We compared the spontaneous apoptotic rate among the three neutrophil phenotypes by means of viable cell counting and cell cycle analysis. As shown in Fig. 3a, it took 3 days until the majority of Gln137Asn-RP S19-producing neutrophil phenotypes died, whereas the mock-transfectant-derived neutrophil phenotypes died within 1 day. The wild type-RP S19 high-producing neutrophil phenotypes seemed to die even faster than the mock control. We then performed the cell cycle analysis at 12 h and 24 h after the 7-day differentiation culture. The cells at G1 subset phase can be defined as apoptotic cells. As shown in Fig. 3b, it was clear that G0/1 cells decreased but G1 subset cells increased in the wild type RP S19 high-producing neutrophil phenotypes.
To confirm the results, we supplemented the extrinsic RP S19 oligomers just after the 7-day differentiation culture and measured the FITC-annexin V-binding cell ratio 12 h and 24 h later. As shown in Fig. 3c, the extrinsic RP S19 oligomers cancelled the apoptosis resistance of the Gln137Asn-RP S19 producing neutrophil phenotypes.
Retarded clearance of Gln137Asn-RP S19 producing neutrophil phenotypes when transplanted into mouse skin
We examined the functions of the RP S19 oligomers coordinating the apoptosis promotion and the macrophage recruitment in the neutrophil clearance in vivo by means of intra-dermal injection of the neutrophil phenotypes in mice. Typical results are shown in Fig. 4 as macroscopic (4a) and microscopic (4b, c, d) images at 72 h after the intradermal injection. Histologically, we examined apoptotic cells and recruited macrophages by acridine orange staining and immunohistochemical staining with the anti-F4/80 antibody, respectively, in addition to the hematoxylin and eosin staining. In the wild-type RP S19 high-producing neutrophil phenotypes, the mass size was significantly reduced, and both the acridine orange-strongly positive apoptotic cells and the F4/80-positive macrophages are abundant in the mass indicating phagocytic clearance of the former cells by the latter. Making a striking contrast, the mass composed of the Gln137Asn-RP S19-producing neutrophil phenotypes was stable. The stainability to acridine orange is low and F4/80-positive cells are only present outside the mass, indicating a retarded apoptosis of the neutrophil phenotypes and ignorance of the clearance system to the transplanted cells. These results strongly suggest the importance of the RP S19 oligomer-C5a receptor system in the clearance of neutrophils.
Augmentation of carrageenan-induced pleurisy and lung parenchymal inflammation in mice by neutralizing RP S19 oligomers locally-generated
We finally used a carrageenan-induced mouse pleurisy model, because it is a simple acute inflammation. To neutralize the RP S19 oligomers, which should be generated by infiltrated neutrophils in the pleural lesion, we simultaneously injected anti-human C5a IgG with carrageenan. The RP S19 oligomers and complement C5-derived C5a fragment share the C5a receptor as ligands. Neither RP S19 monomer nor C5 exhibit a binding capacity to the C5a receptor. Accordingly, the anti-C5a IgG cross-react to the RP S19 oligomers quite well but poorly to the RP S19 monomer or to C5. On the other hand, the anti-RP S19 IgG react to both RP S19 monomer and oligomers. In addition, the RP S19 monomer is normally present in blood plasma as well.[25, 26] It is also oligomerized by activated factor XIII during blood coagulation, and the RP S19 oligomers recruit monocytes/macrophages to phagocytically clear the coagulum. The monomer in circulating plasma would leak into the pleural cavity in association to the vascular permeability enhancement at an early stage of the pleurisy. We had presumed consumption of the anti-RP S19 IgG by the exudated plasma RP S19 monomer, and therefore chose the anti-C5a IgG to neutralize the RP S19 oligomers in vivo. We injected carrageenan and normal rabbit IgG into the left pleural cavity of control mice. We isolated pleural exudate 4 h or 24 h later, and analyzed it for fluid volume and leukocyte numbers. In addition, we examined the lungs histologically.
In the control mice, pleural effusion and neutrophil infiltration were obvious at 4 h, but both of them declined by half at 24 h after the intra-pleural injection. In the ipsilateral left lung, weak edema and neutrophil infiltration were observed at 4 h; these changes decreased to trace the remaining infiltrated mononuclear cells by 24 h. These observations indicated occurrence of a short-lasting acute pleurisy and localized lung inflammation which resolved by 24 h.
In the mice treated with the anti-C5a IgG, more severe pleurisy was observed at 4 h accompanied by a 2.6 fold increase in neutrophils. However, the exudate fluid volume did not significantly increase by the RP S19 oligomer neutralization (Fig. 5a). In the histology of the left lung, more severe edema and neutrophil infiltration around bronchi and large pulmonary arteries were observed (data not shown).
The difference was much more prominent at 24 h. The leukocyte numbers in pleural exudate did not decline but increased, even by 24 h, in the RP S19 oligomer-neutralized group; the accumulated neutrophil number and mononuclear cell number were 31-fold and 11-fold higher than that in the control group, respectively (Fig. 5b). As shown in Fig. 6, the parenchymal inflammatory lesion of the left lung developed enormously in the RP S19 oligomer-neutralized group; the leukocyte infiltration spread from the bronchial wall and vessel wall to peripheral air way regions accompanied by destruction of alveolar septum with severe hemorrhage. The TUNEL staining of the 24 h lung tissue section demonstrates the presence of active neutrophils containing a minute number of apoptotic cells in the RP S19 oligomer-neutralized group. In the control group, apoptotic cells are rarely observed, indicating the apoptotic neutrophils have already cleared in addition to the low severity of the neutrophil infiltration.
In carrageenan-induced pleurisy, the neutrophil infiltration occurs as early as 3 h, which is inhibitable by blockade of chemokine receptors but not of the C5a receptor. Within one day, neutrophil apoptosis and macrophage infiltration began, and the inflammation state resolved in a self-limiting way. The macrophage infiltration at the resolution phase was not affected by the chemokine receptor blockade. In this pleurisy, the prolonged and excessive neutrophil infiltration was currently observed by the neutralization of the RP S19 oligomers. In contrast to the neutrophil number, the fluid volume did not increase. The neutrophil extravasation associates vascular permeability enhancement via cysteinyl leukotriene generation by the neutrophils and endothels in their corporation.[30, 31] The disassociation between the neutrophil number and the fluid volume suggests that the increased neutrophil number would be mainly due to retardation of the neutrophil clearance once infiltrated. This indicates the importance of the RP S19 oligomers in the clearance of infiltrated neutrophils.
The inflammation expanded to the ipsilateral lung, indicating carrageenan would have permeated into lung parenchyma. However, the neutrophil infiltration seemed to be autonomously regulated, probably because the irritant was only a polysaccharide. The resolution process took place as the self-limiting mechanism. In striking contrast, the neutrophil clearance at the lung lesion seemed not to be achieved when the RP S19 oligomers were immunologically neutralized. Destruction of structural components of alveolar septum and the hemorrhage of peripheral airway areas resulted in concomitance with the persistent neutrophil infiltration. The major clearance mechanism for infiltrated neutrophils is their own apoptosis.[10-13] Indeed, only a few apoptotic cells were seen in the parenchymal lesion.
As shown in the series of current experiments in vitro, apoptosis-initiated neutrophils liberated the RP S19 oligomers which have the capacity for both neutrophil apoptosis promotion and monocyte chemoattraction. From these experimental results in vivo and in vitro, we propose that the synchronization of the neutrophil apoptosis and the macrophages recruitment by the RP S19 oligomers must be involved at the resolution stage of acute inflammation. It must be significant to prevent abused tissue destruction by the activated neutrophils.
When the Gln137Asn-RP S19-producing neutrophil phenotype were intradermally transplanted in the mice, destruction of surrounding tissue was not obvious, although the neutrophil-like cells persisted there. This was probably due to lower functions of the neutrophil phenotypes than the primary neutrophils as previously reported for the different cytotoxic capacities.
The C5a receptor is the receptor of the RP S19 oligomers on neutrophils which are undergoing apoptosis. The neutrophil C5a receptor involves in the chemotactic migration of neutrophils as the receptor of C5a at the initiation phase of acute inflammation when the complement activation takes place. Therefore, the C5a receptor plays a dual role in acute inflammation, both to initiate and to terminate. In contrast to the RP S19 oligomers, we observed the life-span prolongation effect of C5a on neutrophils as reported previously. The C5a receptor is a G protein-coupled receptor. The previous study demonstrated that the delay of neutrophil apoptosis was mediated through the phosphatidyl inositol 3-kinase-signaling pathway via the G protein. The G protein signal is negatively regulated by the regulator of G protein signaling (RGS). We previously reported that the RP S19 oligomers and C5a up-regulated and down-regulated RGS3 gene expression, respectively, in apoptosis-initiated HL-60 and AsPC-1 cells. This would be also the mechanism to cause the pro- and anti-apoptotic dual effects of the C5a receptor on neutrophils. It should be experimentally confirmed in future. Similarly, opposite experimental data obtained in the C5a receptor deficient mice have been a paradox; C5a receptor deficient mice showed increment and decrement neutrophil influxes in Pseudomonas aeruginosa-induced bronchopneumonia and in anti-ovalbumin IgG-induced reverse passive Arthus reaction, respectively.[36, 37] We assume that the neutrophil C5a receptor is used mainly at the resolution phase as the receptor of the RP S19 oligomers in the former model but is used at the initiation phase as the receptor of C5a in the latter. However, this is a speculation which should be experimentally tested. In the present study, the C5a receptor exhibits another dual function between neutrophils and macrophages even though ligated by the same RP S19 oligomers. The gene structure of the C5a receptor indicates no splice valiant of its mRNA; therefore, the uniform C5a receptor should be present on neutrophils and on macrophages. It is expected that the dual function should depend upon difference of the intra-cellular signal transduction system between neutrophils and macrophages. An experimental study to elucidate the difference is ongoing in our laboratory.
We thank Ms. T. Kubo for her technical assistance in the histological preparations. This work was supported by a Grant-in Aid for Scientific Research C (KAKENHI 21590441 to T. Y.) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.
- 25A plasma protein indistinguishable from ribosomal protein S19: Conversion to a monocyte chemotactic factor by a factor XIIIa-catalyzed reaction on activated platelet membrane phosphatidylserine in association with blood coagulation. Am J Pathol 2010; 176: 1542–1551., , et al.