ELSEVIER
Journal of Orthopaedic Research 20 (2002) 9-15
Journal of Orthopaedic Research www.elsevier.com/locate/orthres
Differential apoptotic response of J774 macrophages to alumina and ultra- high-molecular-weight polyethylene particles Alain Petit, Isabelle Catelas, John Antoniou, David J. Zukor, Olga L. Huk * Diuision of Orthopaedic Surgery, McGiIl Uniuersity, Lady Ducis Institute for Medicul Reseurch, Tlic Sir Mortitner B. Duais- Jeuidi Giwerul Hospitol, 3755 Ch. C6ie Ste-Cuthevine, Montreui, Que, Canada. H3T l E 2 Received 11 July 2000; accepted 24 April 2001
Abstract
We recently identified apoptosis in in vitro wear particle-stimulated macrophages. The recent explosion of interest in apoptosis lies in the fact that it is under positive and negative regulation through evolutionary conserved biochemical pathways. It may also be possible to modulate macrophage apoptosis in the treatment of periprosthetic osteolysis. The purpose of this study was to compare the macrophage response to identically sized particles of alumina ceramic (AI2O3)and ultra-high-molecular-weight polyethylene (UHMWPE) in terms of TNF-a release and induction of apoptosis. 5774 mouse macrophages were incubated for 0-24 h in the presence of A1203 and UHMWPE particles. TNF-a release was measured by ELISA; Poly(ADP-ribose)polymerase (PARP) and caspase-3 expression was analyzed by Western blot; DNA fragmentation (DNA laddering) was visualized on agarose gel containing ethidium bromide. A1203 particles induced TNF-a release after 4 h incubation with concentrations reaching 483 and 800 pg/ml after 24 h with 125 and 250 particledmacrophage, respectively (control = 161 pg/ml) (P< 0.05 vs. control). The same concentrations of UHMWPE particles induced a much larger and significant TNF-a release after only 1 h incubation, increasing up to 6250 pg/ml after 24 h (P< 0.05 vs. control). Western blot analysis demonstrated that the active caspase-3 fragment ( 1 7 kDa) and the proteolytic PARP fragment (85 kDa) were expressed after 2 h incubation with 125 and 250 A1203particles/macrophage. The active caspase-3 and the PARP fragment had lower expression and appeared after a longer incubation time (8 h) with 125 and 250 UHMWPE particles/macrophage. Finally, DNA fragmentation (DNA laddering) was observed after 16 h with 125 and 250 particles of Alto3 per macrophage whereas no laddering was induced by UHMWPE particles even after 24 h incubation. This study shows that although both A1203 and UHMWPE particles induce TNF-a release, this stimulation was much greater (8-10 times higher) with UHMWPE than A1203 (P < 0.05 vs. control). As well, the induction of apoptosis, as measured by activation of caspase-3, PARP cleavage and DNA laddering, is different for these two particles, being faster and more important with A1203 than UHMWPE. We hypothesize that the ability of A1203 to induce macrophage apoptosis may explain the lower TNF-a release observed with these particles and explain the differences seen in osteolysis patterns of ceramic-ceramic (CC) vs. metal-polyethylene (Mpe) articulations. In conclusion, apoptosis may be a major internal mechanism to decrease macrophage activity and may be a desired therapeutic endpoint. The identification of an apoptosis-related pathway in the macrophage response to ceramic particles provides crucial data for a rational approach in the treatment and/or prevention of periprosthetic osteolysis. 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved.
Introduction Implant wear producing particles leading to osteolysis and subsequent loosening is a critical process that limits the longevity of total hip arthroplasty (THA) [14,35]. Wear particles play an important role in initiating this phenomenon and polyethylene (PE) has been recognized to be the main culprit [3,17,21,32]. Indeed, *Corresponding author. Tel.: +1-514-340-8260 ext. 5287; fax: +15 14-340-7502. E-nzuil address:
[email protected] (O.L. Huk).
many in vivo studies have demonstrated an inflammatory response to PE particles [ 15,201, characterized by macrophage phagocytosis of the particles and leading to the release of mediators that induce bone resorption. In 1969, Pierre Boutin introduced the use of ceramic bearing THAs as an alternative to the classic metalpolyethylene (MPe) articulating interface [2]. The in vivo wear rate of the ceramic-ceramic (CC) joint couple is reported to be up to 200 times less than that of the MPe joint couple [I 11. Although ceramics are considered more bioinert than PE, very little is known about the effect of ceramic particles on macrophage response.
0736-0266/02/$ - see front matter 0 2002 Orthopaedic Research Society. Published by Elsevier Science Ltd. All rights reserved PII: SO7 3 6 - 0 2 6 6 (0 1 )OOO77-8
I0
A. Petit et (11. I .Journal of Orthopedic Ri~searc~h 20 (2002) 9-15
Previous studies performed in our laboratory evaluating interface membranes harvested at revision of CC THAs revealed that, when present in large concentrations, ceramic particles generated in vivo can initiate an inflammatory response in periprosthetic tissue 125,261. We also demonstrated that ceramic particles could induce macrophage TNF-a release [4,5] and apoptosis [6]. The recent explosion of interest in apoptosis lies in the fact that it is under positive and negative regulation through evolutionary conserved biochemical pathways. One of the highlights of apoptotic cell death, as it applies to periprosthetic osteolysis, is that the whole process terminates in the elimination of dead macrophages without the induction of a high inflammatory reaction [8]. The identification of an apoptosis-related pathway in the macrophage response to wear particles may then be a desirable therapeutic endpoint and could provide crucial data for a rational approach in the treatment and/or prevention of periprosthetic osteolysis. The purpose of this study was to compare the macrophage response to identically sized particles of A1201 and UHMWPE in terms of TNF-a release and ability to induce apoptosis.
TNF-n release Concentrations of TNF-a in the supernatants of cell cultures were measured using ELISA kits (Genzyme, Cambridge, MA, USA). These enzyme-linked immunosorbent assay kits are murine specific. The detection limits of the assays are 35-2240 pglml. Kits are specific enough to avoid cross-reactivity of other recombinant cytokines. Protein expripression
Cells (1 x lo6 cells) were homogenized in 150 1-11 of lysis buffer (25 mM tris-HCl (pH 7.5) 1 mM EDTA . 1% SDS - 2% P-mercaptoethanol 10 pllml protease inhibitor cocktail (Sigma-Aldrich Canada, Oakville, ON) and incubated for 30 inin on ice. These cell extracts were spun at SO00 x g for 15 min and the supernatants were collected as the source of total protein. Protein concentration was measured using the Bio-Rad protein assay with bovine serum albumin as standard. Seventy-five micrograms (75 pg) of total proteins for each sample were denatured at IOO"C, loaded on 8% (PARP) and 14% (caspase-3) gels and separated by SDS-PAGE. Proteins were transferred to nitrocellulose membranes using Novex tris-glycine transfer buffer containing 20% methanol. Blotting was performed using anti-PARP diluted I : 1000 (Roche Diagnostics, Laval, Quebec, Canada) and anti-caspase3 diluted at 3 yg/ml (Zymed, San Francisco, CA) as primary antibodies with peroxidase-conjugated anti-rabbit IgG (Zymed) diluted 1 :10 000 as secondary antibody. NEN Renaissance" luminescence reagents were used for detection. Autoradiographies were performed using Kodak X-Omat LS X-ray film. The intensity of bands was quantified using the NIH 1.61B7 software after scanning on UMAX Vista Scan. In the cases that active caspase-3 and proteolytic PARP fragments were present in control cells, the control value was subtracted from the particle-stimulated cells in order to obtain consistent stimulating ratio between experiments. -
-
Materials and methods
D N A ,fragmentation ( D N A laddering)
Pur t icles
D N A was isolated as recently described [6] with minor modifications. Briefly, cells were homogenized in seconds I25 p1 lysis Buffer (50 mM tris-HCl 20 mM EDTA 1% NP-40, pH 7.4) by pipetting up and down for 10 s and centrifuged 5 min at 1600 x g at 4°C. Pellets were reextracted with lysis buffer and centrifuged. Supernatants were pooled and incubated for 2 h at 37°C with 20 yg RNAse A (Amresco, Solon, OH, USA) and overnight at 37°C with 100 yg proteinase K (Amresco, Solon, OH, USA) to remove RNA and proteins, respectively. DNA was extracted twice with an equal volume of phenol/ chloroform mixture and once with chloroformiisoamyl alcohol. After addition of 1/10 volume of 3 M sodium acetate ( p H 5.2) the DNA was precipitated with 2.5 volume ethanol and dissolved in 10 m M tris-HCl - ImM EDTA. p H 8.0. The samples were loaded onto 1.5% agarose gel containing 50 pg/mI ethidium bromide, run at SO volts for about 2 h on Mupid-2 Minigel (Helixx Technologies, Scarborough, Ontario, Canada). D N A was then visualized directly upon illumination with UV light and photographed using Polaroid 667 film (ASA 3000).
Alumina ceramic (A1203; Alcoa, Bauxite, AK, USA) (1.3 ym) and ultra-high-molecular-weight polyethylene (UHMWPE; Pfizer, Rutherford, NJ, USA) (0.5-2 pin) particles were commercially obtained in the phagocytosable range. These particles have clinical relevance and their sizes are physiologically pertinent [3,11,25]. Particles were endotoxin-free as demonstrated by the Sigma E-toxate assay for detection and semi-quantification of endotoxins. Cell cultiirc.
The 5774 mouse macrophage cell line (ATCC, Rockville, MD, USA) was used in this in vitro model because it has been demonstrated to be morphologically similar to macrophages at the bonexement interface and to respond to cement in a way similar to macrophages at that interface [18]. This cell line is also known to produce IL-I and T N F - r (ATCC specifications). Macrophages were cultured and maintained in RPMI 1640 tissue culture medium (Biomedia Canada, Drummondville, Quebec, Canada) supplemented with 5% foetal bovine serum, 100 U/m1 penicillin. and 100 yglml streptomycin. Cultured cells were exposed to particles in 25 cm' culture flasks (DNA laddering) or 35 x 10 mm dishes (protein expression). The number of particles used per test (25, 125, and 250 particleslmacrophage) was based on previous work in our laboratory which demonstrated that particle phagocytosis was size- and concentration-dependent [4--61. and thereby exerted a volume effect. Particles were suspended in culture medium and vortexed for 10 s at maxiinal speed just before incubation to avoid clumping. They sank down in culture flasks and dishes and were then in direct contact with the monolayer of macrophages. This allowed particle phagocytosis, as we previously demonstrated in culture tubes with A1203 [6] and UHMWPE (personal unpublished data) using flow cytometry. Culture flasks and dishes without particles served as negative controls. Incubations were conducted for 0-24 h at 37°C in a 5% CO2 environment. The resulting conditioned supernatants were collected for ELISA tests and frozen at -80°C.
-
~
Stutistical unalysi;,
Statistical significance was calculated using ANOVA followed by Fisher's PLSD comparison test. P < 0.05 was considered significant. Each result is the mean of 3 4 experiments.
Results
Fig. 1 shows the effect of A1203 and UHMWPE particles on the release of TNF-a by J774 macrophages. Results show that A1203 particles induced a significant TNF-a release after 4 h incubation ( P < 0.05 vs. control; control = 31 pg/ml). This release increased with AI2O3 particle concentrations, reaching 483 and 800 pglml after 24 h with 125 and 250 particles/macrophage, re-
A. Petit et al. I Journal of Orthopuetlic Resenrcli 20 (2002) 9-I5
II
A1203 7000
--. E M
6000
ir
5000 JOOO
c
v
3000 F?on0 I000 n
0 0
J
X
I?
16
20
24
T I M E (Lour)
0
4
8
I2
7,
r
16
20
0 I
24
T I M E (hour)
Fig. 1, Stimulation ofTNF-a release by AlZ03and UHMWPE particles. 5774 macrophages were incubated for 0-24 h in the absence (control) or the presence of particles. TNF-a release was measured by ELISA. Results are the mean i S.E. of four experiments performed in duplicate.
spectively (P < 0.05 vs. control; control = 161 pg/ml). The same concentrations of UHMWPE particles induced a much larger and significant TNF-a release after only 1 h incubation with 125 particles/macrophage (3750 pg/ml; P < 0.05 vs. control) and reached concentrations as high as 6250 pg/ml after 24 h with 250 particles/ macro phage. Figs. 2 and 3 show the effect of A1203 and UHMWPE particles on the expression of caspase-3. The active fragment of caspase-3 (1 7 kDa) was expressed after only 2 h incubation with 250 A1203 particles/macrophage (1 5'1/0of native caspase-3 expression) and increased with time to reach 131% of native caspase-3 expression after
24 h with the same particle concentration (Figs. 2 and 6). This active fragment had lower with lower particle concentrations reaching only 5% of native caspase-3 expression with 25 particles/macrophage. Caspase-3 active fragment expression was also lower with UHMWPE particles, appearing only after 8 h with 250 particles/ macrophage and reaching 27% of native caspase-3 expression (Figs. 3 and 6). No caspase-3 active fragment was observed with the lowest UHMWPE particle concentration (25 particles/macrophage). Figs. 4 and 5 show the effect of A1203 and UHMWPE particles on the expression of PARP. The proteolytic PARP fragment (85 kDa) was expressed after only 2 h
Fig. 2. Expression of caspase-3 in A1203 particle-stimulated macrophages. 5774 macrophages were incubated for 0-24 h in the absence (control - time 0 h) or the presence of A1203 particles at 25, 125, and 250 particles per macrophage. Caspase-3 expression was visualized by Western blot. Results are representatives of three experiments.
Fig. 3. Expression of caspase-3 in UHMWPE particle-stimulated macrophages. 5774 macrophages were incubated for 0-24 h in the absence (control - time 0 h) or the presence of UHMWPE particles at 25, 125, and 250 particles per macrophage. Caspase-3 expression was visualized by Western blot. Results are representatives of three experiments.
12 150 125 100 75
50 25
0 2
8
16
24
16
24
Time (h)
Fig. 4. Expression of' PARP 111AI20?particle-stimulated macrophages. 5774 macrophages were incubated for 0-24 h in the absence (control time 0 h) o r the presence of A1203particles at 25, 125, and 250 particles per macrophage. PARP expression was visualized by Western blot. Results are representatives of three experiments.
-
2
8 Time (h)
incubation with 250 Al2O3particles/macrophage (12% of native PARP expression) and increased with time to reach 27% of native PARP expression after 24 h with the same particle concentration (Figs. 4 and 6). This active fragment had lower expression with lower particle concentrations (5% of native PARP expression with 25 particles/macrophage). Proteolytic PARP fragment expression was also lower with UHMWPE particles, appearing only after 8 h with 250 particles/macrophage and reaching 14% of native PARP expression (Figs. 5 and 6).
Fig. 6. Expression of caspase-3 and PARP in particle-stimulated macrophages. 5774 macrophages were incubated for 0-24 h in the absence (control) or the presence ofA120, or UHMWPE particles. Caspase-3 and PARP expression were visualized by Western blot and quantify as described in Materials and Methods. Results are the mean S.E. of three experiments.
*
No PARP fragment was observed with the lowest UHMWPE particle concentration (25 particles/macrophage). Figs. 7 and 8 show the effects of A1203 and UHMWPE particles on macrophage DNA fragmentation. DNA fragmentation (DNA laddering) was observed after 16 h with 125 and 250 particles of A1203per macrophage (Fig. 7) whereas no laddering was induced by 25 particles of A1203 per macrophage. UHMWPE particles did not induce DNA fragmentation either at low (25 particles/macrophage) or high (250 particled macrophage) concentrations and incubation times up to 24 h (Fig. 8).
Discussion
Fig. 5. Expression of PARP in UHMWPE particle-stimulated macrophages. 5774 macrophages were incubated for 0-24 h in the absence (control time 0 h ) or the presence of UHMWPE particles at 25, 125, and 250 particles per macrophage. PARP expression was visualized by Western blot. Results are representatives of three experiments. -
Our results show that although both A1203 and UHMWPE particles induce TNF-a release, this stimulation was much greater (8-10 times higher) with UHMWPE than A1203. The induction of apoptosis as measured by activation of caspase-3, PARP cleavage and DNA laddering, was different for these two particle compositions, being faster and more important with A1203than with UHMWPE.
A. Petit rt ul. I Journul of’ Orthopuedic Reseurch 20 (2002) 9-15
13
Time (hour) 0
2
4
8
1624
0 2 4
8
16
24
0 2 4 8 1 6 2 4
M.W.
1300 bp 700 bp 200 bp
(25 parthnacroph. )
(125 part./mrcroph.)
(250 parthacroph.)
Fig. 7. Effect of A1203 particles on macrophage DNA fragmentation. J774 macrophages were incubated for 0-24 h in the absence (control) or the presence of A1203particles at 25, 125, and 250 particles per macrophage. DNA was analyzed on I.% agarose gel containing ethidium bromide. Results are representatives of three experiments.
Time (hour) 0 2 4 8
1624
0 2 4 8 1 6 2 4
0 2 4 8 1 6 2 4
M.W.
1300 bp 700 bp
200 bp
(25 parthacroph. )
(125 part./macroph.)
(250 parthacroph.)
Fig. 8. Effect of UHMWPE particles on macrophage DNA fragmentation. 5774 macrophages were incubated for 0-24 h in the absence (control) or the presence of UHMWPE particles at 25, 125, and 250 particles per macrophage. DNA was analyzed on 1.5% agarose gel containing ethidium bromide. Results are representatives of three experiments.
Histologically, the aggressive granulomatous infiltrate found at the periprosthetic interface of loose components consists mainly of macrophages [13,16]. These macrophages are a potent source of local cytokine production [lo] and play a significant role in the progression of aseptic loosening of joint arthroplasties [23]. Among these cytokines, TNF-cc is the prototype triggering cytokine initiating a variety of defense mechanisms and immunological responses [33]. A recent study on the presence of cytokine in periprosthetic tissues concluded that TNF-cl, and to a less extend IL-1 and 1L-6, are positively related to the severity of osteolysis around the prosthesis [34]. Moreover, it was demonstrated in bone
organ culture that the addition of anti-TNF-a antibody inhibited the bone resorption stimulated by macrophagepolymethymethacrylate (PMMA) conditioned medium [I]. As well, it was recently shown that mice failing to express TNF-a receptors (in vivo model) were protected from the profound bone resorption associated with PMMA particle implantation [27]. These results suggest that TNF-a plays a central role in the induction of implant osteolysis and it has been proposed that reducing TNF-a may decrease wear particle osteolysis [34]. TNF-a appears to be the most important cytokine produced in vitro by murine 5774 macrophages in response to orthopaedic particles [ 1,4,5,19,27,29] whereas
14
A. Petit et al. I Journal of Orthopaedic Research 20 (2002) 9-15
IL-1 is not induced by these particles. We therefore focused on particle-induced TNF-a release in the present study as an indicator of macrophage stimulation. The difference in TNF-a response between A l 2 0 3 and UHMWPE is larger than what we previously reported for high density polyethylene (HDP) which stimulates TNF-u only two to three times more than A 1 2 0 3 [5]. The less intense TNF-a response observed with HDP in our previous studies may be explained by a difference in particle sizes since the average size of HDP was 4.5 pm compared to 0.5-2.0 pm for the UHMWPE particles used in this study. As well, a difference in structure between the two types of PE can account for the differential response. TNF-a response to UHMWPE observed in our mouse 5774 macrophage model is similar to what was observed in a human monocytic cell line [31]. Moreover, the difference in TNF-a release between A1203and UHMWPE is also observed in this human model [30]. The present study also demonstrated that A1203 particles can stimulate macrophage apoptosis in vitro. Results show that Al2O3 particles induce DNA fragmentation into oligonucleosome length fragments (DNA ladder), a gold standard for the confirmation of apoptosis [22]. We also demonstrated that N2O3 particles stimulate the proteolytic cleavage of PARP, a nuclear protein implicated in DNA repair [9]. The cleavage of this protein into an 85 kDa fragment has been closely associated with apoptosis [12,24]. Moreover, the cleavage of PARP suggests that a caspase pathway is implicated in the macrophage response to ceramic since it is a cellular substrate for caspases [A.This is supported by the present results showing that caspase-3 is activated by Al2O3 particles and our in vivo finding demonstrating the activation of caspase-3 in interface membranes from aseptically loose THAs [28]. Although UHMWPE particles stimulated P A W cleavage, DNA laddering was not observed in our experimental conditions. This can be explained by the fact that PARP cleavage is one of the early reversible events associated with apoptosis and its detection does not assure that the cell will complete the apoptotic cycle with its associated DNA fragmentation visualized as the typical DNA ladder. Moreover, and in keeping with that reported in the literature, the absence of DNA laddering suggests that macrophages can survive despite the presence of active caspase-3 in the cytoplasm [3q. Contrary to the rapid effect of N2O3on PARP cleavage, UHMWPE-induced cleavage is much slower and suggests an indirect effect on the caspase cascade. One of the highlights of apoptotic cell death, as it applies to periprosthetic osteolysis, is that the whole process terminates in the elimination of dead macrophages without the induction of a significant inflammatory reaction [8]. We therefore hypothesize that the ability of A1203particles to induce macrophage a p o p
tosis may explain the lower TNF-a release observed with these particles and explain the differences seen in osteolysis patterns of CC vs. Mpe articulations [26]. The induction of macrophage apoptosis may therefore be a desirable therapeutic endpoint. In conclusion, apoptosis may be a major internal mechanism to decrease macrophage activity and the identification of an apoptosisrelated pathway in the macrophage response to ceramic particles provides groundwork for a rational approach in the treatment andor prevention of periprosthetic osteolysis.
Acknowledgements The authors would like to thank the Orthopaedic Research Fund of the Sir Mortimer B. Davis - Jewish General Hospital for hancial assistance.
References Algan SM, Purdon M, Horowitz SM. Role of tumor necrosis factor alpha in particle-induced bone resorption. J Orthop Res 1996;14:3&5. Boutin P. Arthroplastie totale de hanche par prothkse en alumine frittk. Etude e x m e n t a l e et premieres applications cliniques. Rev Chir Orthop 1972;58:229-46. Campbell P, Ma S, Yeom B, McKellop H, Schmalvied TP, Amstutz HC. Isolation of predominantly submicron-sized UHMWPE wear particles from periprosthetic tissues. J Biomed Mater Res 1995;29127-31. Catelas I, Huk OL, Petit A, Zukor DJ, Marchand R, Yahia L’H. Flow cytometric analysis of macrophage response to ceramic and polyethylene particles: effects of size, concentration, and composition. J Biomed Mater Res 1998;41:600-7. Catelas I, Petit A, Marchand D, Zukor DJ, Yahia L‘H, Huk OL. Cytotoxicity and macrophage cytokine release induced by ceramic and polyethylene particles in vitro. J Bone Jt Surg Br 1999; 81B:51&21. Catelas I, Petit A, Zukor DJ, Marchand R, Yahia L’H, Huk OL. Induction of macrophage apoptosis by ceramic and polyethylene particles in vitro. Biomaterials 1999;2062%30. Cohen GM. Caspases: the executioners of apoptosis. Biochem J 1997;326:1-16. Cohen JJ. Apoptosis. Immunol Today 1993;14126-30. De Murcia G, De Murcia JM. Poly (ADP-ribose)polymerase:a molecular nick-sensor. Trends Biochem 1994;19172-6. Dinarello CA.Interleukin-1 and interleukin-1 antagonism. Blood 1991;77: 1627-52. Dorlot JM, Christel P, Meunier A. Wear analysis of retrieved alumina heads and sockets of hip prostheses. J Biomed Mater Res 1989;23(Suppl A3):299-3 10. Duriez PJ, Shah GM. Cleavageof poly (ADP-ribose)polymeraseA sensitive parameter to study cell death. Biochem Cell Biol 1997;75:337-49. Goldring MB, Goldring SR. Skeletal tissue response to cytokines. Clin Orthop Rel Res 1990;258:245-78. Goodman SB, Chin RC, Chiou SS, Schurman DJ, Woolson ST, Masada MP. A chid-pathologic-biochemical study of the membrane surrounding loosened and nonloosened total hip arthroplasties. Clin Orthop Re1 Res 1989;244:182-7.
A. Petit et al. I Journal of Orthopaedic Research 20 (2002) 9-15 [15] Goodman S, Wang JS, Regula D, Aspenberg P. T-lymphocytes are not necessary for particulate polyethylene-induced macrophage recruitment. Histologic studies of the rat tibia. Acta Orthop Scand 1994;65:15740. [16] Haynes DR, Rogers SD, Hay S, Pearcy MJ, Howie DW. The differences in toxicity and release of bone-resorbing mediators induced by titanium and cobalt-chromium-alloy wear particles. J Bone Jt Surg Am 1993;75:825-34. [17] Hernandez JR, Keating EM, Faris PM, Meding JB, Ritter MA. Polyethylene wear in uncemented acetabular components. J Bone Jt Surg Br 1994;76-B:2634. [IS] Horowitz SM, Doty SB, Lane JM, Burstein AH. Studies of the mechanisms by which the mechanical failure of polymethylmethacrylate leads to bone resorption. J Bone Jt Surg Am 1993;75A:802-13. [19] Horowitz SM, Luchetti WT, Gonzales JB, Ritchie CK. The effects of cobalt chromium upon macrophages. J Biomed Mater Res 1998;41:468-73. [20] Howie DW, Manthey B, Hay S, Vernon-Roberts B. The synovial response to intraarticular injection of polyethylene wear particles. Clin Orthop Re1 Res 1993;292352-7. [21] Jasty M, Bragdon C, Jiranek W, Chandler H, Maloney W, Harris WH. Etiology of osteolysis around porous-coated cementless total hip arthroplasties. Clin Orthop Re1 Res 1994;308:111-26. [22] Junying Y. Evolutionary conservation of a genetic pathway of programmed cell death. J Cell Biochem 1996;60:411. [23] Kadoya Y, Revel1 PA, Al-Saffar N, Kobayashi A, Scott G, Freeman MA. Bone formation and bone resorption in failed total joint arthroplasties: histomorphometric analysis with histochemical and immunohistochemical techniques. J Orthop Res 1996; 14:473-82. [24] Kaufmann SH, Desnoyers S, Ottaviana Y, Davidson NE, Poirier GG. Specific proteolytic cleavage of poly (ADP-ribose) polymerase: an early marker.of chemotherapy-induced apoptois. Cancer Res 1993;53:397685. [25] Lerouge S, Huk 0, Yahia LH, Sedel L. Characterizationof in vivo wear debris from ceramic-ceramic total hip arthroplasties. J Biomed Mater Res 1996;32:627-33.
1s
[26] Lerouge S, Huk 0, Yahia L’H, Witvoet J, Sedel L. Ceramicceramic and metal-polyethylene total hip replacements. Comparison of pseudomembranes after loosening. J Bone Jt Surg Br 1997;79-B:135-9. [27l Merkel KD, Erdmann JM, McHugh KP, A b u - h e r Y, Ross FP, Teitelbaum SL. Tumor necrosis factor-alpha mediates orthopaedic implant osteolysis. Am J Pathol 1999;154203-10. [28] Petit A, Zukor DJ, Huk OL. Identification of apoptosis-related proteins in pseudomembranes of aseptically loose total hip arthroplasty. J Bone Jt Surg Br 1999;81-B(Suppl I - Orthopaedic Transactions):99. [29] Prabhu A, Shelburne CE, Gibbons DF. Cellular proliferation and cytokine responses of murine macrophage cell line J774A.1 to polymethylmethacrylate and cobalt-chrome alloys particles. J Biomed Mater Res 1998;42:655-63. [30] Rader CP, Baumann B, Sterner T, Rolf 0, Hendrich C, Schutze N, Jakob F. TNF response of macrophages to wear particles, and its modification by drugs. Biomediz Tech 1999;44:135-41. [31] Rader CP, Sterner T, Jakob F, Schutze N, Eulert J. Cytokine response of human macrophage-like cells after contact with polyethylene and pure titanium particles. J Arthroplasty 1999; 14840-8. [32] Santavirta S, Nordstrom D, Metsarinne K, Konttinen YT. Biocompatibility of polyethylene and host response to loosening of cementless total hip replacement. Clin Orthop Re1 Res 1993;297:100-10. [33] Stashenko P, Dewhirst FE, Peros WJ, Kent RL, Ago JM. Synergistic interactions between interleukin 1, tumor necrosis factor, and lymphotoxin in bone resorption. J Immunol 1987; 138:1464-8. [34] Stea S, Visentin M, Granchi D, Ciapetti G, Donati ME, Sudanese A, Zanotti C, Toni A. Cytokines and osteolysis around total hip prostheses. Cytokine 2000; 12:1575-9. [35] Willert HG. Reactions of the articular capsule to wear products of artificial joint prostheses. J Biomed Mater Res 1977;ll: 15764. [36] Zeuner A, Eramo A, Peschle C, De R. Caspase activation without death. Cell Death Differ 1999;6:1075-80.
