|
|
||||||||
Published online before print December 15, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Division of Rheumatology, Duke University Medical Center, and Medical Research Service, Durham VA Hospital, Durham, North Carolina
1 Correspondence: Durham VA Medical Center, Box 151G, 508 Fulton St., Durham, NC 27705. E-mail: dpiset{at}acpub.duke.edu
| ABSTRACT |
|---|
|
|
|---|
Key Words: apoptosis necrosis neutrophils peritonitis
| INTRODUCTION |
|---|
|
|
|---|
/ß production by their action on plasmacytoid dendritic cells [11
, 12
]. Similarly, in mice expressing a rheumatoid factor transgene, such immune complexes can activate B cells by a mechanism involving Toll-like receptor 9 [13
, 14
]. Understanding the origin of blood DNA is therefore important for elucidating the mechanisms of autoimmune and other immune-mediated diseases. In view of the conditions in which its levels are elevated, blood DNA has generally been considered the result of cell death, and apoptosis is implicated as the major mechanism for DNA release. Indeed, DNA in the blood shows low molecular weight and size-laddering, the hallmarks of the apoptotic process [3 , 6 , 15 , 16 ]. Furthermore, in vitro, apoptotic cells can release DNA into the extracellular milieu. In contrast, necrotic cells do not release DNA under in vitro conditions [17 ]. The role of apoptosis in DNA release is strengthened by observations that treatment of mice with an anti-Fas antibody, which induces widespread hepatic apoptosis, causes a large increase in blood DNA [18 ].
The settings in which blood DNA levels are elevated are, in general, complex and involve cell activation and cell death, often occurring concomitantly [2 , 7 , 8 , 19 ]. Thus, in tissue injury from trauma or toxin exposure, inflammation may follow the death-inducing insult. Similarly, in tumors, areas of necrosis and apoptosis may coexist adjacent to surrounding inflammation. These processes are dynamic, and the cellular composition of an inflammatory site as well as the activation state of cells present evolve over time. Furthermore, as these processes progress, the elaboration of cytokines and other inflammatory mediators may provoke subsequent waves of cell death and activation [20 21 22 ].
In a previous study, we used a murine model to elucidate the mechanisms of DNA release, testing whether cell death is sufficient to induce a blood DNA response and whether apoptotic and necrotic cells are similar in their behavior [23 ]. For this purpose, normal mice were administered apoptotic or necrotic Jurkat cells, and plasma DNA was determined using the fluorimetric dye PicoGreen. Results of these studies indicated that, in vivo, apoptotic and necrotic cells can produce blood DNA and that the blood DNA arising from these cells shows laddering with both. Furthermore, we showed that in mice, in which macrophages were eliminated by clodronate treatment, a rise in blood DNA did not occur after administration of the dead cells. Together, these results indicate that the generation of blood DNA depends on macrophages and may not simply reflect the occurrence of cell death.
In view of the close relationship of inflammation and cell death and the role of macrophages in the clearance of dead and dying cells, we have therefore used an in vivo system to investigate whether prior inflammation can influence the generation of blood DNA from dead and dying cells. For this purpose, we induced peritoneal inflammation in normal mice prior to the administration of apoptotic or necrotic cells and then assessed plasma DNA by fluorimetry. In results presented herein, we show that, in mice with peritonitis induced by thioglycollate (TG) or sodium periodate (NaIO4), a plasma DNA response did not occur following administration of Jurkat cells treated with etoposide to induce apoptosis or ethanol to induce necrosis. In contrast, mice treated with peptone (PT) had blood DNA responses similar to those of untreated mice. Together, these results indicate that inflammation can affect the generation of DNA from dead and dying cells, although the nature of this effect may vary depending on the inflammatory stimulus and local cellular events.
| MATERIALS AND METHODS |
|---|
|
|
|---|
For the induction of apoptosis, Jurkat cells growing in RPMI 1640 with 10% fetal bovine serum were treated with etoposide (Sigma Chemical Co.) at 30 ug/ml for 24 h. For the induction of necrosis, cells were treated with 70% ethanol for 10 min. After treatment, cells were washed twice with phosphate-buffered saline (PBS; Gibco-BRL, Grand Island, NY) and suspended in PBS. The cells were administered IP at 108 cells/mouse. Following these treatments, mice were bled at regular times thereafter as indicated in the figures. The blood samples were collected into tubes with 35 µl 0.5 M EDTA, followed immediately by centrifugation. The plasma was removed and stored at 20°C until use. The peritoneal lavage samples were collected in 1 ml cold PBS. The cells were pelleted and stained for flow cytometry analysis. The peritoneal fluids were saved and stored at 20°C until use.
Quantitation of DNA in plasma and peritoneum
Plasma DNA was assessed by a fluorimetric assay as described previously [23
]. Briefly, plasma in various dilutions were mixed at a 1:1 ratio with the dye PicoGreen (Molecular Probes, Eugene, OR), diluted 1:200 in 10 mM Tris, 1 mM EDTA, pH 8 (TE buffer) in a black, 96-well microtiter plate (Costar, Corning Inc., Corning, NY). The DNA concentration was determined from fluorescence measurements using a TECAN GENios microplate fluorescence reader (Salzburg, Austria) with an excitation wavelength at 485 nm and an emission wavelength at 535 nm. Data were collected as relative fluorescence units. The concentration of DNA in plasma was calculated according to a standard curve using double-stranded calf thymus DNA (Sigma Chemical Co.). Peritoneal fluids were diluted 1:10 in TE buffer, and DNA levels were measured as described above.
Flow cytometry analysis
To assess the effects of various agents on immune cell populations, flow cytometry was performed on preparations of peritoneal washout cells. Briefly, mice treated with TG, PT, or NaIO4 were killed by cervical dislocation, and peritoneal cells were harvested at various time-points by lavage with 1 ml cold PBS. The cells were pelleted at 300 g for 5 min and resuspended in hypotonic lysis buffer to remove red blood cells, followed by centrifugation and two washes with PBS/0.5% bovine serum albumin. The cell concentrations were adjusted to 1 x 107/ml in PBS, and 1 x 106 cells were used for staining. Cells were treated with anti-mouse CD16/CD32 antibody (BD PharMingen, San Diego, CA) and then stained with phycoerythrin (PE)-anti-mouse F4/80 (Serotec, Raleigh, NC) or PE-anti-mouse Ly-6G or Ly-6C (Gr1; BD PharMingen), along with Annexin V-fluorescein isothiocyanate (BD PharMingen). Cells were analyzed using a FACScan flow cytometer (Becton Dickinson, Mansfield, MA). Data analysis was done using CellQuest software (Becton Dickinson Immunocytometry Systems, San Jose, CA).
Enzyme assay
Caspase 3 was assayed using a caspase 3 kit (Molecular Probes, E13184). Briefly, plasma samples were diluted 1:100 in PBS, and peritoneal fluids were used neat. Samples were incubated with caspase 3 substrate at room temperature for 30 min. Fluorescence units (FU) were determined using a TECAN GENios microplate fluorescence reader as described above.
| RESULTS |
|---|
|
|
|---|
As shown previously, plasma DNA rises following administration of either type of dead cell with DNA derived predominantly, but not exclusively, from the Jurkat cells [23 ]. Figures 1 2 3 show the results of these experiments; for NaIO4-treated mice, only apoptotic cells were studied. As these data indicate, consistent with previous observations for control mice, the magnitude and time-course of appearance of plasma DNA were similar with apoptotic and necrotic cells, returning to baseline by 24 h after administration. Among the three groups of mice treated to induce peritonitis, however, the appearance of plasma DNA differed. For mice treated with TG or NaIO4, the plasma DNA levels were reduced markedly compared with controls, whereas for PT-treated mice, the levels of DNA were comparable with controls. These results suggest that local inflammation can affect the generation of plasma DNA from dead and dying cells, although the effect will vary depending on how the peritonitis is induced.
|
|
|
Table 1 presents the results of experiments showing the total number of peritoneal exudate cells, the number of neutrophils (Gr1+), and the number of macrophages (F4/80+). As these data indicate, the highest number of peritoneal cells observed occurred in the TG-treated mice at early time-points (6 and 16 h) as well as at later time-points (72 h). At 6 and 16 h, the predominant cells in all groups were neutrophils, and at 72 h, the predominant cells were macrophages. In the TG-treated mice, F4/80+ cells were especially abundant, far exceeding the numbers in the other two treatment groups.
|
These findings suggest that prior to the administration of the apoptotic and necrotic Jurkat cells, the peritoneal cavity had been the site of cell death, and the extent and cell population involved depend on the inducing agent. To further assess this possibility, levels of caspase 3 and DNA were determined in the peritoneal fluid. Caspase 3 and DNA are released from apoptotic cells and provide a marker of apoptosis in biological fluids [17 , 30 , 31 ]. Figure 4 presents results of these determinations for peritoneal fluid. As these data indicate, each of the stimulating agents led to a distinct pattern of DNA and caspase 3 expression. Whereas exudates induced by TG showed DNA and caspase 3, exudates from mice treated with NaIO4 showed only DNA. In contrast, exudates from mice treated with PT showed neither DNA nor caspase 3.
|
| DISCUSSION |
|---|
|
|
|---|
Studies assessing the generation of blood DNA in animals have been of two general types. In the first, an animal is treated with an agent that can induce apoptosis or necrosis in vivo. Agents tested have included lipopolysaccharide (LPS), anti-Fas, and dexamethasone (DEX). It is interesting, that although LPS and anti-Fas cause a blood DNA response, DEX does not [18 ]. These findings suggest that the release of DNA may vary depending on the death stimulus and cell type induced to die. In this regard, systems of this type are potentially confounded by the effects of the inducing agent on the activity of phagocytic cells and clearance of dead and dying cells. For example, LPS may promote the activity of macrophages, and DEX may affect the specific interaction of phagocytes with apoptotic cells [32 , 33 ]. Modeling DNA release by inducing cell death in living animals is also limited by uncertainty in the number and location of the dying cells.
The second general model for modeling the release of DNA into the blood involves the administration of cell treated in vitro to die. In contrast to the administration of an inducer of apoptosis or necrosis, the administration of dead and dying cells provides a more defined system in which the number of dead cells is known more precisely, and the response of the recipient animal can be manipulated. In previous studies exploring this approach, we showed that administration of 108 apoptotic or necrotic Jurkat cells to normal mice leads to the appearance of DNA in the blood in a time- and dose-dependent manner [23 ]. This DNA was observed with two different cell lines made apoptotic or necrotic by a variety of means, suggesting a general property of dead and dying cells. As Jurkat cells are of human origin and were derived from a male, the contribution of this cell type to circulating DNA can be established unequivocally by the demonstration by PCR of Y-chromosomal sequences [23 ].
As shown by many experiments in vivo and in vitro, macrophages play a key role in the clearance of apoptotic and necrotic cells and can engulf these cells with high efficiency [23 , 34 35 36 ]. To explore the role of macrophages in the generation of blood DNA, in previous experiments, we assessed the response to the administration of dead and dying cells of mice in which macrophage function was eliminated by clodronate liposomes. Clodronate is a bisphosphonate, which can induce macrophage apoptosis after uptake and render a mouse deficient in this cell population [37 , 38 ]. Using Jurkat cells as a model, we showed that administration of apoptotic or necrotic cells to a mouse without macrophages fails to produce a blood DNA response. Similar results were obtained in mice treated with silica, which can also eliminate macrophage function [23 ].
Although these findings point to a central role of macrophages in the generation of circulating DNA, they do not define the role of these cells in the DNA release reaction. Most studies on macrophage clearance of dead cells have focused on the uptake of these cells and have not addressed the disposition of their contents. Thus, it is possible that the macrophages take up dead cells, degrade their contents, and expel the remnants into the extracellular milieu. In the absence of macrophages, the dead cells may gradually autolyze or disintegrate in a manner that fails to cause a measurable elevation of circulating DNA. In this scenario, the released DNA is an end-product of the clearance process, with levels rising as the burden of material to be eliminated increases.
An alternative possibility posits that DNA release occurs when the phagocytic capacity of the macrophage is exceeded. In this situation, a phagocytic macrophage filled with dead cells and their breakdown products is induced to undergo apoptosis, releasing its own DNA and that of the engulfed DNA cells. Circumstantial evidence for this possibility is derived from data showing that the blood of mice receiving dead and dying cells contains murine DNA sequences as well as that of the administered human cell. Furthermore, the DNA of murine and human origin shows size-laddering, implying a role for nucleases activated during apoptosis [23 ]. In this scenario, DNA release thus reflects macrophage dysfunction and death rather than heightened physiological function.
The current results provide data relevant to distinguish between these mechanisms. Thus, findings presented herein indicate that inflammation can alter the subsequent clearance of apoptotic and necrotic cells, although the nature of the effect may vary depending on the inflammatory stimulus [39 , 40 ]. Thus, we have tested the effects of three different agents that have been commonly used to induce peritonitis. TG and PT represent complex mixtures comprised of many components [24 , 27 28 29 , 41 ]. With TG, advanced glycation products, which develop as the mixture "ages" may be the key inducing agent. The proinflammatory components in PT have not been well defined. In contrast, NaIO4 is a well-defined chemical that oxidizes surface molecules, including terminal sugars, to produce free aldehydes [25 , 26 , 42 ]. These modifications induce macrophage activation and T cell mitogenesis, and these functional changes are associated with subsequent recruitment of cells into the peritoneum [25 , 26 , 42 ].
Because of properties of these irritants, the nature of the exudates following their administration may differ. Indeed, a variety of studies have compared macrophages arising in these settings with respect to their activation state, functional properties, and production of various cellular proteins. These studies have identified important differences in the features of the resulting exudates [27 28 29 ]. Although the mechanisms of cellular recruitment and activation in the peritoneum are not well understood, our findings suggest that patterns of peritonitis and likely other forms of local inflammation may differ significantly in their impact on the clearance and metabolism of dead and dying cells.
Among possible events that occur in inflammatory sites that may affect the subsequent clearance of dead and dying cells is the occurrence of apoptosis as the inflammatory reaction proceeds. Thus, each of the treatments produced an increased number of annexin+, Gr1+ cells, although these numbers were greatest with TG treatment, which also led to the highest levels of annexin+, F4/80+ cells. Furthermore, with TG-induced peritonitis, the occurrence of apoptosis could be inferred from the presence of DNA and caspase 3 in the peritoneal fluid. In mice with this treatment, a DNA response was not observed following administration of Jurkat cells. Similarly, in mice treated with NaIO4, administration of dead and dying cells did not lead to a blood DNA response. Although peritoneal fluid after NaIO4 administration did not contain caspase 3, it did contain DNA. In contrast, PT-treated mice showed an intact blood DNA response and lacked peritoneal fluid DNA as well as caspase 3.
These findings thus suggest that the effect of inflammation on the magnitude of the blood DNA response resulting from administration of dead and dying cells relates to the presence of DNA in the peritoneal fluid. At present, the mechanisms leading to this DNA and its effect on the functional properties of macrophages are speculative. It is possible, however, that DNA itself or other debris from apoptotic cells may affect the capacity of macrophages to clear dead cells that subsequently arise locally or in the case of these experiments, the administered Jurkat cells. In this regard, the exudates of mice treated with TG and NaIO4, despite their content of DNA, differed in the presence of caspase 3. Although apoptosis can account for simultaneous release of DNA and caspase 3, the mechanisms leading to DNA in the absence of caspase by NaIO4 are unknown. Possibilities include DNA release during cell activation, cellular leakiness secondary to membrane oxidation, or inhibition of caspase 3 by this agent. These possibilities are under investigation.
Studies from a number of investigators have suggested that, in contrast to previous notions, apoptotic cells may have proinflammatory effects and modulate the function of macrophages [43 44 45 46 ]. Among these functions, clearance of apoptotic cells may be modulated, an issue to be considered in experiments. For example, studies about the role of macrophages in the clearance of dead and dying cells have involved mice that have been treated previously with TG to increase macrophage number and facilitate the analysis [47 ]. It is possible that such treatment may have affected important elements of this process, and the effects we have observed on DNA released also affected interactions such as binding and uptake of dead and dying cells.
As shown in studies in human and murine systems, DNA in the blood may have important immune effects that influence the pathogenesis of SLE [48 49 50 ]. These effects include the formation of immune complexes that promote cytokine production and can deposit in the kidney to induce glomerulonephritis [51 , 52 ]. Other nuclear molecules released from dead and dying cells may also have pathogenic effects. For example, the high-mobility group protein has powerful, proinflammatory properties and can function as a cytokine once it has been released from necrotic cells [53 54 55 56 ]. Studies are therefore in progress to determine changes in macrophages that occur during inflammation, including the contact with apoptotic and necrotic cells and their impact on subsequent events in disease pathogenesis.
| ACKNOWLEDGEMENTS |
|---|
Received July 19, 2004; revised November 18, 2004; accepted November 27, 2004.
| REFERENCES |
|---|
|
|
|---|
producing cells in systemic lupus erythematosus Hum. Immunol. 63,1181-1193[CrossRef][Medline]
production in plasmacytoid dendritic cells by immune complexes containing nucleic acid released by necrotic or late apoptotic cells and lupus IgG Arthritis Rheum. 50,1861-1872[CrossRef][Medline]
This article has been cited by other articles:
![]() |
D. S. Pisetsky The Role of Nuclear Macromolecules in Innate Immunity Proceedings of the ATS, July 1, 2007; 4(3): 258 - 262. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |