Macrophage activation induced by different carbon fiber-epoxy resin composites G . Peluso* lnstitute of Protein Biochemistry and Enzymology-CNR, Arco Felice 80072, Naples, Italy L. Ambrosio, M. Cinquegrani, and L. Nicolais Department of Materials and Production Engineering, University of Naples, and Institute of Composite Materials Technology-CNR, Naples 80125, Italy G . Tajana lnstitute of Normal Human Anatomy, 11 Faculty of Medicine and Surgery, University of Naples, It a ly The activation of cells by interaction with solid surfaces is important in many settings, including the response of tissue to implanted materials. However, few comprehensive studies of both cell migration and activation have been performed so that the connection between these events and immunological activation against foreign material is not well understood. In the present study, synthesis and expression of Ia antigens by peritoneal exudate macrophages after implantation of differ-
ent carbon fiber composites in the rat peritoneal cavity have been investigated in order to determine whether the type of material implanted affected the composition of Ia-bearing cells of the exudate. The results have confirmed the low level of expression of Ia on resident peritoneal macrophages; while we have found that macrophages, harvested after implantation, express a different amount of Ia related to the different cure cycles of the composite materials used.
INTRODUCTION
Integration of biomaterials and devices by rat tissue is a widely used in vivo model to evaluate the biocompatibility of certain materials and therefore has been studied extensively for evaluation of the inflammatory process due to the presence of a foreign body in an organism.’ Although primarily considered a disease of the tissue site of implantation, immunological rejection against a biomaterial may be a systemic disease with a wide variety of extraimplant effects including alterations in lymphocyte and monocyte/ macrophage functions.25To a large extent, these changes in function reflect increased expression or even expression de novo of numerous secretory and surface proteins such as MHC class I1 antigens, Ia.6,7Immune-associated (Ia) antigens or class I1 histocompatibility molecules are essential for macrophages to present antigens to T lymphocytes. In fact, current models of T cell activation place a major emphasis on the role of I region products, and it has *To whom correspondence should be addressed. Journal of Biomedical Materials Research, Vol. 25, 637-649 (1991) CCC 002l-9304/91/050637-13$4.00 0 1991 John Wiley & Sons, Inc.
PELUSO ET AL.
638
been hypothesized that T cells can only recognize antigens in the context of Ia molecules on the macrophage membrane.*f9 The nature of the substrate material with which cells interact has been shown to affect a variety of cellular processes, including recognition, activation, and interleukin production. In the immune system, recognition of foreign substrates is an essential primary step in macrophage clearance and protective functions. Recognition may directly trigger the activation necessary for phagocytosis and consequently may be implicated in reactions to implanted synthetic materials." Commercially available epoxide mixtures are employed as matrices for different composites such as carbon-fiber-reinforced epoxy composites. Thermoset-based composites are processed in a single irreversible operation which transforms a low-molecular-weight liquid into a cross-linked polymer. The cure of a reactive prepolymer involves the transformation of lowmolecular-weight monomers or oligomers from a liquid to rubber and solid state, as a result of the formation of a polymeric network by the chemical reaction of the reactive groups in the system. These reactions are generally achieved under controlled conditions but incomplete degree of cure can be obtained as a consequence of many variables such as aging of the resin in the prepreg, different thicknesses of the laminate, uncontrolled temperature profile in the processing equipment (autoclave, pultruder, oven, etc.). We have investigated the synthesis and expression of Ia antigens by peritoneal exudate macrophages after implantation of different carbon fiber composites in the rat peritoneal cavity to determine whether the type of material implanted affected the composition of Ia-bearing cells of the exudate. METHODS A N D MATERIALS
Materials The prepreg of carbon fiber and epoxy resin was obtained from Fiberite (ICI subsidiary, Greenville, TX). The reported composition'' of C Fiber/ TGDDM-DDS commercial prepreg is summarized in Table I. Three prepreg plies, cut to round shape (diameter 0.7 cm), were laminated and cured in a stainless-steel mold covered with silicon foils. Cure conditions were: (1) 130°C for 3 h for partially cured material (PCM), (2) 130°C for 3 h and 180°C for 2 h for fully cured material (FCM). Table I Chemical Consituents in Carbon Fiber Epoxy Resin Prepreg Constituents TGDDM DDS Diglycidyl orthopthalate epoxy (DGOP) BF3:NHzCzHg (relative to 100 parts of mixture)
Fiberite 934 (wt%) 64 25 11 0.4
MACROPHAGE ACTIVATION
639
DSC analysis
Differential Scanning Calorimetry (DSC) is of ten used for indirect determination of the advancement of the cure of thermosetting systems, with the assumption that the heat evolved is proportional to the extent of reaction." The thermocalorimetric characterization has been carried out in a DSC METTLER 3000 (Mettler Instrumente, Volkestwil, Switzerland) operating in the range of temperature between -20 and 320°C in nitrogen atmosphere and equipped with a liquid nitrogen cooling system. The tests were performed under dynamic conditions on samples of 35-40 mg of prepreg resulting in 14-16 mg of resin. Samples were cured isothermally in an oven by cure cycle that was been reported previously. Then they were scanned in the calorimeter at a heating rate 10°C min-' from which the heat evolved during completion of cross-linking (AH,) was measured. The degree of cure (a)was calculated as a=
AH, - AH, AHt
AH, is the total heat liberated when an uncured material is taken to complete cure and this value is a constant for a particular thermosetting resin.
Surface analysis X-ray photoelectron spectroscopy (XPS) has been used to analyze the surface properties of both PCM and FCM. With this technique it is possible to obtain quantitative information (surface characteristics) and qualitative information (functional g r o ~ p s ) .XPS ' ~ spectra were obtained on a PHI model 58 XPS-AES spectrometer (Perkin Elmer, Eden Prairie, MN), using the MgK, radiation (1253.6 eV) from a 400 W source. The analysis chamber pressure Pa, without baking. was maintained near 2 x Signal averaging to obtain good spectra was possible due to the connection of the spectrometer to a PDP 11/50 computer. Further data processing (smoothing, background subtraction, integration, deconvolution) was carried out using in-house software on a Sperry 1100/72 mainframe computer. The hydrocarbon Cls peak was used as a reference and set to 284.6 el? The surface composition was calculated from the spectra using the appropriate sensitivity factors.I4
In vivo experimental procedure
Rats Male, Wistar rats, were obtained from Charles River (Como, Italy) and housed in four rats/cage in self-contained laminar flow rooms. The rats weighed approximately 275 g at the start of each experiment. Tissue culture medium: RPMl 1640 supplemented with 10% inactivated Fetal Calf Serum,
640
100 U/mL penicillin, 100 &mL was used.
PELUSO ET AL.
streptomycin sulfate, and 2mM L-glutamine
CelIs To obtain peritoneal washout cells, rats were killed by exsanguination and the abdominal skin was stripped back. Ten milliliters of Hanks' balanced salt solution (HBSS) containing 100 U/mL penicillin, 100 ,ug/mL streptomycin, and 2 U/mL heparin were injected into the peritoneal cavity and immediately removed with a syringe and needle. The cells were washed twice in the same solution (but without heparin), counted, and resuspended in RPMI 1640 complete medium.
Experimental protocols We examined peritoneal exudates induced in various ways to determine whether the type of artificial material used affected the composition of Ia-bearing cells of the exudate. The experiments involved two different protocols: (1) Control rats were operated on the abdomen but not implanted and their exudates were harvested 7 days later, (2) Rats were intraperitoneally implanted with fully or partially cured materials and their exudates were harvested and examined 3 to 7 days later. In most experiments, the peritoneal exudates were harvested from three or four rats and pooled. Most experiments, except as indicated, were done at least three times. Variations in absolute numbers of cells in the exudate were found among groups of rats depending on their age and probably environmental factors as well. Individual experiments were made with rats of the same age and sex, housed in identical conditions.
lmrnunofluorescent staining Cells were stained for immunofluorescence analysis by direct or indirect labeling procedures. For direct labeling, antibodies were conjugated with f luorescein isothiocyanate (FITC); for indirect staining, unconjugated antibodies were used as the first reagent, and FITC labeled xenogeneic polyspecific or IgG class-specific antibodies were used as second reagent. For all indirect staining procedures, controls of only the second reagent without the first were done routinely. The following primary mouse anti-rat monoclonal antibodies (TechnoGenetics, Italy) were used: OX-19 (CD3+T cells); W3/25 (CD4+helpedinducer T cells, macrophages); OX-8 (CD8' cytotoxic/suppressorT cells, Natural Killer cells); OX-12 (K light chains, pan B cells); ED1 (monocytes/macrophages); OX-6 (MHC class 11); OX-39 (IL-2 receptor).
MACROPHAGE ACTIVATION
641
Polyspecific goat anti-mouse antibodies to total immunoglobulins, and to mouse IgG subclasses, were also used. All reagents were used at concentrations previously determined to give maximal staining and the specificities of the antibodies were confirmed using spleen cells from rats of the same strains, or with the "sandwich" reagents used without first-layer antibodies. After the final washes, cells were counterstained with propidium iodide to allow exclusion of dead (brightly stained with propidium iodide) cells from the immunof luorescence analysis.
FACS analysis
Cells were examined in a Becton Dickinson cytofluorograph (FACScan, Mountain View, CA) using the 488-mm emission line of an argon laser. For each stained cell sample, forward scatter (related to cell size), right-angle scatter (related to granularity), green fluorescence (FITC), and red f luorescence (propidium iodide) values were collected for 20,000 cells. Live cells were those without bright propidium iodide staining and with sufficient forward scatter to exclude remaining erythrocytes or other cellular debris. In some experiments, lymphocytes were further distinguished and separately analyzed by virtue of their characteristic combination of forward and rightangle scatter. Control cell samples not stained with specific antibody allowed determination of the proper FACScan gates to distinguish positive from negative cells in each staining procedure. The percentages of f luorescence-positive live cells and the mean fluorescence and scatter values of the fluorescencepositive cells were determined by a Hewlett-Packard computer system. In sandwich staining procedures, the percentage of specifically stained cells was determined by subtracting the percentage of positive cells after control (second layer only) staining from those obtained after staining with both firstand second-layer antibodies. To study the staining intensities of antigens on PEC from differently treated animals, the mean fluorescence channel of the positive cells from implanted rats has been divided by the mean fluorescence channel of the positive cells from operated but not implanted animals, both as determined by the Hewlett-Packard computer attached to the FACS.
RESULTS
Chemical aspects
DSC traces of uncured material and of materials cured by cure cycle #1 and #2 are presented in Figure 1. AHr and AHt values are summarized in Table 11. The degree of cure was a = 1 for materials treated by cure cycle #2 and a = 0.67 for materials treated by cure cycle #1. Previous inve~tigations"~'~*'~ have shown that three reactions dominate the cure behavior of the tetraglycidyldiaminodiphenyl-methane-diamino-
PELUSO ET AL.
642
T
0
a 1
'
.
.
.
1
"
50.
0.
'
-
1
.
.
100.
.. 150.
--
ent carbon fiber composites in the rat peritoneal cavity have been investigated in order to determine whether the type of material implanted affected the composition of Ia-bearing cells of the exudate. The results have confirmed the low level of expression of Ia on resident peritoneal macrophages; while we have found that macrophages, harvested after implantation, express a different amount of Ia related to the different cure cycles of the composite materials used.
INTRODUCTION
Integration of biomaterials and devices by rat tissue is a widely used in vivo model to evaluate the biocompatibility of certain materials and therefore has been studied extensively for evaluation of the inflammatory process due to the presence of a foreign body in an organism.’ Although primarily considered a disease of the tissue site of implantation, immunological rejection against a biomaterial may be a systemic disease with a wide variety of extraimplant effects including alterations in lymphocyte and monocyte/ macrophage functions.25To a large extent, these changes in function reflect increased expression or even expression de novo of numerous secretory and surface proteins such as MHC class I1 antigens, Ia.6,7Immune-associated (Ia) antigens or class I1 histocompatibility molecules are essential for macrophages to present antigens to T lymphocytes. In fact, current models of T cell activation place a major emphasis on the role of I region products, and it has *To whom correspondence should be addressed. Journal of Biomedical Materials Research, Vol. 25, 637-649 (1991) CCC 002l-9304/91/050637-13$4.00 0 1991 John Wiley & Sons, Inc.
PELUSO ET AL.
638
been hypothesized that T cells can only recognize antigens in the context of Ia molecules on the macrophage membrane.*f9 The nature of the substrate material with which cells interact has been shown to affect a variety of cellular processes, including recognition, activation, and interleukin production. In the immune system, recognition of foreign substrates is an essential primary step in macrophage clearance and protective functions. Recognition may directly trigger the activation necessary for phagocytosis and consequently may be implicated in reactions to implanted synthetic materials." Commercially available epoxide mixtures are employed as matrices for different composites such as carbon-fiber-reinforced epoxy composites. Thermoset-based composites are processed in a single irreversible operation which transforms a low-molecular-weight liquid into a cross-linked polymer. The cure of a reactive prepolymer involves the transformation of lowmolecular-weight monomers or oligomers from a liquid to rubber and solid state, as a result of the formation of a polymeric network by the chemical reaction of the reactive groups in the system. These reactions are generally achieved under controlled conditions but incomplete degree of cure can be obtained as a consequence of many variables such as aging of the resin in the prepreg, different thicknesses of the laminate, uncontrolled temperature profile in the processing equipment (autoclave, pultruder, oven, etc.). We have investigated the synthesis and expression of Ia antigens by peritoneal exudate macrophages after implantation of different carbon fiber composites in the rat peritoneal cavity to determine whether the type of material implanted affected the composition of Ia-bearing cells of the exudate. METHODS A N D MATERIALS
Materials The prepreg of carbon fiber and epoxy resin was obtained from Fiberite (ICI subsidiary, Greenville, TX). The reported composition'' of C Fiber/ TGDDM-DDS commercial prepreg is summarized in Table I. Three prepreg plies, cut to round shape (diameter 0.7 cm), were laminated and cured in a stainless-steel mold covered with silicon foils. Cure conditions were: (1) 130°C for 3 h for partially cured material (PCM), (2) 130°C for 3 h and 180°C for 2 h for fully cured material (FCM). Table I Chemical Consituents in Carbon Fiber Epoxy Resin Prepreg Constituents TGDDM DDS Diglycidyl orthopthalate epoxy (DGOP) BF3:NHzCzHg (relative to 100 parts of mixture)
Fiberite 934 (wt%) 64 25 11 0.4
MACROPHAGE ACTIVATION
639
DSC analysis
Differential Scanning Calorimetry (DSC) is of ten used for indirect determination of the advancement of the cure of thermosetting systems, with the assumption that the heat evolved is proportional to the extent of reaction." The thermocalorimetric characterization has been carried out in a DSC METTLER 3000 (Mettler Instrumente, Volkestwil, Switzerland) operating in the range of temperature between -20 and 320°C in nitrogen atmosphere and equipped with a liquid nitrogen cooling system. The tests were performed under dynamic conditions on samples of 35-40 mg of prepreg resulting in 14-16 mg of resin. Samples were cured isothermally in an oven by cure cycle that was been reported previously. Then they were scanned in the calorimeter at a heating rate 10°C min-' from which the heat evolved during completion of cross-linking (AH,) was measured. The degree of cure (a)was calculated as a=
AH, - AH, AHt
AH, is the total heat liberated when an uncured material is taken to complete cure and this value is a constant for a particular thermosetting resin.
Surface analysis X-ray photoelectron spectroscopy (XPS) has been used to analyze the surface properties of both PCM and FCM. With this technique it is possible to obtain quantitative information (surface characteristics) and qualitative information (functional g r o ~ p s ) .XPS ' ~ spectra were obtained on a PHI model 58 XPS-AES spectrometer (Perkin Elmer, Eden Prairie, MN), using the MgK, radiation (1253.6 eV) from a 400 W source. The analysis chamber pressure Pa, without baking. was maintained near 2 x Signal averaging to obtain good spectra was possible due to the connection of the spectrometer to a PDP 11/50 computer. Further data processing (smoothing, background subtraction, integration, deconvolution) was carried out using in-house software on a Sperry 1100/72 mainframe computer. The hydrocarbon Cls peak was used as a reference and set to 284.6 el? The surface composition was calculated from the spectra using the appropriate sensitivity factors.I4
In vivo experimental procedure
Rats Male, Wistar rats, were obtained from Charles River (Como, Italy) and housed in four rats/cage in self-contained laminar flow rooms. The rats weighed approximately 275 g at the start of each experiment. Tissue culture medium: RPMl 1640 supplemented with 10% inactivated Fetal Calf Serum,
640
100 U/mL penicillin, 100 &mL was used.
PELUSO ET AL.
streptomycin sulfate, and 2mM L-glutamine
CelIs To obtain peritoneal washout cells, rats were killed by exsanguination and the abdominal skin was stripped back. Ten milliliters of Hanks' balanced salt solution (HBSS) containing 100 U/mL penicillin, 100 ,ug/mL streptomycin, and 2 U/mL heparin were injected into the peritoneal cavity and immediately removed with a syringe and needle. The cells were washed twice in the same solution (but without heparin), counted, and resuspended in RPMI 1640 complete medium.
Experimental protocols We examined peritoneal exudates induced in various ways to determine whether the type of artificial material used affected the composition of Ia-bearing cells of the exudate. The experiments involved two different protocols: (1) Control rats were operated on the abdomen but not implanted and their exudates were harvested 7 days later, (2) Rats were intraperitoneally implanted with fully or partially cured materials and their exudates were harvested and examined 3 to 7 days later. In most experiments, the peritoneal exudates were harvested from three or four rats and pooled. Most experiments, except as indicated, were done at least three times. Variations in absolute numbers of cells in the exudate were found among groups of rats depending on their age and probably environmental factors as well. Individual experiments were made with rats of the same age and sex, housed in identical conditions.
lmrnunofluorescent staining Cells were stained for immunofluorescence analysis by direct or indirect labeling procedures. For direct labeling, antibodies were conjugated with f luorescein isothiocyanate (FITC); for indirect staining, unconjugated antibodies were used as the first reagent, and FITC labeled xenogeneic polyspecific or IgG class-specific antibodies were used as second reagent. For all indirect staining procedures, controls of only the second reagent without the first were done routinely. The following primary mouse anti-rat monoclonal antibodies (TechnoGenetics, Italy) were used: OX-19 (CD3+T cells); W3/25 (CD4+helpedinducer T cells, macrophages); OX-8 (CD8' cytotoxic/suppressorT cells, Natural Killer cells); OX-12 (K light chains, pan B cells); ED1 (monocytes/macrophages); OX-6 (MHC class 11); OX-39 (IL-2 receptor).
MACROPHAGE ACTIVATION
641
Polyspecific goat anti-mouse antibodies to total immunoglobulins, and to mouse IgG subclasses, were also used. All reagents were used at concentrations previously determined to give maximal staining and the specificities of the antibodies were confirmed using spleen cells from rats of the same strains, or with the "sandwich" reagents used without first-layer antibodies. After the final washes, cells were counterstained with propidium iodide to allow exclusion of dead (brightly stained with propidium iodide) cells from the immunof luorescence analysis.
FACS analysis
Cells were examined in a Becton Dickinson cytofluorograph (FACScan, Mountain View, CA) using the 488-mm emission line of an argon laser. For each stained cell sample, forward scatter (related to cell size), right-angle scatter (related to granularity), green fluorescence (FITC), and red f luorescence (propidium iodide) values were collected for 20,000 cells. Live cells were those without bright propidium iodide staining and with sufficient forward scatter to exclude remaining erythrocytes or other cellular debris. In some experiments, lymphocytes were further distinguished and separately analyzed by virtue of their characteristic combination of forward and rightangle scatter. Control cell samples not stained with specific antibody allowed determination of the proper FACScan gates to distinguish positive from negative cells in each staining procedure. The percentages of f luorescence-positive live cells and the mean fluorescence and scatter values of the fluorescencepositive cells were determined by a Hewlett-Packard computer system. In sandwich staining procedures, the percentage of specifically stained cells was determined by subtracting the percentage of positive cells after control (second layer only) staining from those obtained after staining with both firstand second-layer antibodies. To study the staining intensities of antigens on PEC from differently treated animals, the mean fluorescence channel of the positive cells from implanted rats has been divided by the mean fluorescence channel of the positive cells from operated but not implanted animals, both as determined by the Hewlett-Packard computer attached to the FACS.
RESULTS
Chemical aspects
DSC traces of uncured material and of materials cured by cure cycle #1 and #2 are presented in Figure 1. AHr and AHt values are summarized in Table 11. The degree of cure was a = 1 for materials treated by cure cycle #2 and a = 0.67 for materials treated by cure cycle #1. Previous inve~tigations"~'~*'~ have shown that three reactions dominate the cure behavior of the tetraglycidyldiaminodiphenyl-methane-diamino-
PELUSO ET AL.
642
T
0
a 1
'
.
.
.
1
"
50.
0.
'
-
1
.
.
100.
.. 150.
--
