BIOCHEMICAL
Vol. 182, No. 2, 1992 January 31, 1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS Pages 727-732
THE EFFECT OF INTERLEUKIN-4 ON THE MACROPHAGE RESPIRATORY BURST IS SPECIES DEPENDENT Wayne A. Phillips*, Maryann Croatto, and John A. Hamilton University of Melbourne Department of Medicine Royal Melbourne Hospital, Melbourne 3050, Australia Received
December
16,
1991
Summary. Preexposure of human monocytes to recombinant human interleukin-4 (IL-4) suppressed the respiratory burst response to a number of different stimuli, including phorbol myristate acetate, zymosan, platelet-activating factor and the chemotactic peptide, f-met-leuphe. Under similar conditions, the respiratory burst of murine macrophages was enhanced by preexposure to recombinant murine IL-4. By conducting our studies on cells from different species under similar conditions we have demonstrated that there is a significant disparity in the effects of IL-4 on human and murine macrophages thus providing an explanation for some apparent inconsistencies in the literature and highlighting the need for caution when extrapolating data across species barriers. 0 1992 Academic Press, Inc.
Interleukin-4 (IL-4) was originally described as B cell growth factor or B cell-stimulating factor-l and was believed to be primarily active on B lymphocytes. However, it was soon demonstrated to have effects on a variety of cell types including both B and T lymphocytes, mast cells and macrophages, and was given the designation IL-4 in recognition of the pleiotropic nature of its actions (l-3). Among the numerous accounts of the often diverse biological effects of IL-4 on macrophages are an increasing number of conflicting reports. We (4) and others (5) have demonstrated the ability of IL-4 to prime murine macrophages for an enhanced respiratory burst, although investigators studying the respiratory burst of human monocytes have reported inhibitory effects of IL-4 (6,7). Similarly, several studies have demonstrated a suppressive effect of IL-4 on the production of inflammatory monokines including tumour necrosis factor CY(TNFa)
(8-10); however, there are reports that IL-4 can induce TNFa! synthesis in
macrophages (11,12). Also, some reports indicate an enhancing effect of IL-4 on the tumouricidal (12,13) and microbicidal (14,15) capacity of macrophages while others suggest a suppressive effect (7,16). * To whom correspondence should be addressed.
727
0006-291X/92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
While it is apparent that there are major discrepancies in the literature with regard to the actions of IL-4 on macrophages, the reasons behind these conflicting reports have not been investigated. Although a cursory examination suggests that the major differences in the literature may be the result of differing effects of IL-4 on murine and human macrophages, the possibility that the discrepancies may have arisen from technical differences in the experimental protocols has not been eliminated. Clarification of the basis for differences in the reported effects of IL-4 is essential if we are to fully understand the physiological role of IL-4. We have therefore undertaken investigations on the effect of IL-4 on the respiratory burst of human monocytes and murine macrophages under similar experimental conditions, MATERIALS
AND METHODS
Reagents: Recombinant murine IL-4 purified from Escherichia coli (specific activity r7 X lo6 U/mg, with 1 U/ml giving 50% maximum stimulation of the HT-2 cell line (17)), recombinant human IL-4 purified from E.coli (specific activity 0.5 X 106 U/mg with 1 U giving 50% maximum growth of PHA-activated T cells (18)) and rat anti-human IL-4 monoclonal antibody MP4-25D2-11 (5 rig/ml will block 5 U/ml human IL-4) were generously provided by DNAX Research Institute, Palo Alto, CA. Recombinant human TNFcr was kindly supplied by Boehringer Ingelheim, Vienna, Austria, and recombinant human interferon-y (IFN-7) (2.7 X 10’ U/mg) was supplied by Hoffman-LaRoche, Basel, Switzerland. Other major reagents were purchased from the following sources: the calcium ionophore A23187, the chemotactic peptide f-met-leu-phe (FMLP), horse heart cytochrome c type III, phorbol myristate acetate (PMA) and zymosan A from Sigma Chemical Company, St.Louis, MO; foetal calf serum (FCS) from the Commonwealth Serum Laboratories, Parkville, Victoria, Australia; alpha-minimal essential medium (CY-MEM) from Flow Laboratories, Sydney, Australia; bacterial lipopolysaccharide (LPS) (purified from E.coZi Olll:B4 by the Westphal method) from Difco Laboratories, Detroit, MI; Lymphoprep, from Nycomed, Oslo, Norway; and hexadecyl platelet-activating factor (PAF) was obtained from Novabiochem, Switzerland. All other chemicals and reagents were of analytical grade or higher and obtained from standard commercial sources. PMA, FMLP, and A23187 were stored at -70°C as stock solutions of 10’M in dimethylsulfoxide and were diluted in buffer immediately prior to use. PAF was stored at 20°C in aliquots of 10e3M in phosphate-buffered saline containing 0.25% bovine serum albumin. Zymosan was washed in phosphate-buffered saline (PBS) and boiled for 30 min before use. Cells: Human monocytes were isolated from fresh buffy coat preparations provided by the Red Cross Blood Bank, Melbourne. Thirty ml of buffy coats, diluted 1 in 3 with sterile 0.9% NaCl, were layered over 10 ml pyrogen-tested Lymphoprep (Nycomed, Oslo, Norway) and centrifuged (30 min, 500 g). The upper layer containing the mononuclear cells was collected and washed three times (5 min, 500 g) to remove excess platelets. The cells were then counted and plated into 24-well tissue culture plates at 3 X 106 nucleated cells per well in 1 ml of (r-MEM. After 2hr at 37”C/5%C02, the nonadherent cells were removed by washing three times with 0.5 ml PBS, the medium replaced with 1 ml a-MEM containing 10% FCS, and the adherent cells incubated overnight (37”C,5%COJ. Murine resident peritoneal macrophages were obtained as previously described (19). All cells were isolated and cultured under strict endotoxin-free conditions as previously described (19). Cytokine treatment: Adherent human monocytes or murine resident peritoneal macrophages were washed (3 X 0.5 ml PBS) and the medium replaced with 1 ml CY-MEM containing 10% FCS and appropriate concentrations of cytokine or LPS (as indicated) and incubated for 48 hr (37”C, 5%COJ. 728
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
burst: The respiratory burst of adherent macrophages was assessed by the stimulus-induced reduction of cytochrome c as previously described (19).
Respiratory
RESULTS AND DISCUSSION Both PMA (10e6M) and zymosan (1 mg/ml) are able to induce a substantial respiratory burst in untreated human monocytes (582&-96 nmollmg
protein, n=18,
and 356+87
nmol/mg, n= 16, respectively). Pretreatment of the monocytes with IL-4 for 48 hr suppressed the respiratory burst produced in response to stimulation with either PMA or zymosan in a dose-dependent fashion (Figure l), a result consistent with previous reports (6,7). A similar reduction in the respiratory burst was also observed in IL-4-treated (10 U/ml, 48 hr) monocytes stimulated with the calcium ionophore A23187 (10e6M) (control, nmol/mg, n=lO;
133+39
IL-4-treated, 39&12 nmol/mg, n=5). This inhibitory effect of IL-4 (10
U/ml) could be completely prevented by preincubation of the IL-4 with a neutralising antibody (100 rig/ml) (n=2) or by heating the IL-4 to 100°C for 30 min (n=3). Interestingly, neither PAF (10e7M) or FMLP (10e7M) were able to induce a respiratory burst in the untreated control monocytes. However, it was found that the monocytes could be primed for a response to these agents by pretreatment with TNFol, IFN-7 or LPS for 48 hr (Figure 2). To our
01
"0
IL-4 "
concentration .l 1
10
02
control
TNF
IFN
IL-4
(U/ml)
Figure 1. Dose-dependent effect of IL-4 on the respiratory burst of human monocytes. Human monocytes were pretreated 48 hr with various concentrations of human IL-4. The respiratory burst was assessed by cytochrome c reduction in response to WM PMA (-o-) or 1 mglml zymosan (--o--). Data are expressed as the percentage of the response of control cells pretreated with medium alone. Shown are mean + S.E.M. from three independent experiments. Fieure 2. Priming the monocytes respiratory burst with cytokines. Human monocytes were treated for 48 hr with medium alone (control), TNFa (l@M), IFN-y (100 U/ml), IL-4 (10 U/ml) or LPS (100 rig/ml). The cells were then washed and the respiratory burst assessed in response to stimulation with 10.‘M PAF (open bars) or 10.‘M FMLP (cross-hatched bars). Shown are mean + S.E.M. from at least three independent experiments.
729
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND RIOPHYSICAL
RESEARCH COMMUNICATIONS
knowledge, this is the first report that priming is necessary to enable human monocytes to produce a respiratory burst in response to PAF or FMLP
and this may reflect our strenuous
efforts to minimise any contamination of our cell cultures with bacterial endotoxin, a common laboratory contaminant with potent priming effects on monocytes (20). Unlike TNFq
IFNy,
or LPS, IL-4 was not able to prime human monocytes for a
respiratory burst response to PAF or FMLP (Figure 2). However, IL-4 was found to antagonise the priming effects of TNFa (Figure 3). A similar antagonism of the priming effects of TNFa! on the A23187-induced respiratory burst was also observed (data not shown). Whether this was simply the net result of independent and opposing actions of IL-4 and TNFo or, alternatively, due to IL-4 directly antagonising the action of TNFcr, cannot be discerned by our current data. However, the observation that IL-4 suppressed the respiratory burst in response to all stimuli tested, irrespective of priming, would perhaps favour the former possibility. In contrast to the suppressive effects of IL-4 on human monocytes, we found that murine IL-4 enhances the respiratory burst of murine resident peritoneal macrophages stimulated with PMA
(Figure
4). This finding
is consistent
with the reported
priming
effects of murine
IL-4
on bone marrow-derived macrophages (4) and peptone-elicited peritoneal macrophages (5).
SC .F .g sz
0
40
9.5
20
icij 3%
e
0
0 control
3 Figure
3. Effect
IL-4
of IL-4
TNF
TNF+
IL-4
and TNFc
2Qo L
100 0
4
on the PAF-induced
control
IL-4
respiratory
TNF
burst.
TNF+IL4
Human
monocyteswere pretreated for 48 hr with medium alone (control), IL-4 (10 U/ml), TNFar (10m9M), or TNFa and IL-4 together. The cells were then washed and the respiratory burst assessed in response to IQ’M PAF. Shown are mean + S.E.M. from six independent experiments. Figure 4. Effect of IL-4 and TNFu on the PMA-induced respimtory burst of mu&e macrophages. Murine resident peritoneal macrophages were pretreated for 48 hr with medium alone (control), IL-4 (5 U/ml), TNFo (W’M), or TNFa and IL-4 together. Cells were then washed and the respiratory burst assessed in response to 10e6MPMA. Shown are mean f S.E.M. from five independent experiments. (PcO.05 for each pretreatment compared to control; P
Vol. 182, No. 2, 1992 January 31, 1992
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS Pages 727-732
THE EFFECT OF INTERLEUKIN-4 ON THE MACROPHAGE RESPIRATORY BURST IS SPECIES DEPENDENT Wayne A. Phillips*, Maryann Croatto, and John A. Hamilton University of Melbourne Department of Medicine Royal Melbourne Hospital, Melbourne 3050, Australia Received
December
16,
1991
Summary. Preexposure of human monocytes to recombinant human interleukin-4 (IL-4) suppressed the respiratory burst response to a number of different stimuli, including phorbol myristate acetate, zymosan, platelet-activating factor and the chemotactic peptide, f-met-leuphe. Under similar conditions, the respiratory burst of murine macrophages was enhanced by preexposure to recombinant murine IL-4. By conducting our studies on cells from different species under similar conditions we have demonstrated that there is a significant disparity in the effects of IL-4 on human and murine macrophages thus providing an explanation for some apparent inconsistencies in the literature and highlighting the need for caution when extrapolating data across species barriers. 0 1992 Academic Press, Inc.
Interleukin-4 (IL-4) was originally described as B cell growth factor or B cell-stimulating factor-l and was believed to be primarily active on B lymphocytes. However, it was soon demonstrated to have effects on a variety of cell types including both B and T lymphocytes, mast cells and macrophages, and was given the designation IL-4 in recognition of the pleiotropic nature of its actions (l-3). Among the numerous accounts of the often diverse biological effects of IL-4 on macrophages are an increasing number of conflicting reports. We (4) and others (5) have demonstrated the ability of IL-4 to prime murine macrophages for an enhanced respiratory burst, although investigators studying the respiratory burst of human monocytes have reported inhibitory effects of IL-4 (6,7). Similarly, several studies have demonstrated a suppressive effect of IL-4 on the production of inflammatory monokines including tumour necrosis factor CY(TNFa)
(8-10); however, there are reports that IL-4 can induce TNFa! synthesis in
macrophages (11,12). Also, some reports indicate an enhancing effect of IL-4 on the tumouricidal (12,13) and microbicidal (14,15) capacity of macrophages while others suggest a suppressive effect (7,16). * To whom correspondence should be addressed.
727
0006-291X/92 $1.50 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
182,
No.
2, 1992
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
While it is apparent that there are major discrepancies in the literature with regard to the actions of IL-4 on macrophages, the reasons behind these conflicting reports have not been investigated. Although a cursory examination suggests that the major differences in the literature may be the result of differing effects of IL-4 on murine and human macrophages, the possibility that the discrepancies may have arisen from technical differences in the experimental protocols has not been eliminated. Clarification of the basis for differences in the reported effects of IL-4 is essential if we are to fully understand the physiological role of IL-4. We have therefore undertaken investigations on the effect of IL-4 on the respiratory burst of human monocytes and murine macrophages under similar experimental conditions, MATERIALS
AND METHODS
Reagents: Recombinant murine IL-4 purified from Escherichia coli (specific activity r7 X lo6 U/mg, with 1 U/ml giving 50% maximum stimulation of the HT-2 cell line (17)), recombinant human IL-4 purified from E.coli (specific activity 0.5 X 106 U/mg with 1 U giving 50% maximum growth of PHA-activated T cells (18)) and rat anti-human IL-4 monoclonal antibody MP4-25D2-11 (5 rig/ml will block 5 U/ml human IL-4) were generously provided by DNAX Research Institute, Palo Alto, CA. Recombinant human TNFcr was kindly supplied by Boehringer Ingelheim, Vienna, Austria, and recombinant human interferon-y (IFN-7) (2.7 X 10’ U/mg) was supplied by Hoffman-LaRoche, Basel, Switzerland. Other major reagents were purchased from the following sources: the calcium ionophore A23187, the chemotactic peptide f-met-leu-phe (FMLP), horse heart cytochrome c type III, phorbol myristate acetate (PMA) and zymosan A from Sigma Chemical Company, St.Louis, MO; foetal calf serum (FCS) from the Commonwealth Serum Laboratories, Parkville, Victoria, Australia; alpha-minimal essential medium (CY-MEM) from Flow Laboratories, Sydney, Australia; bacterial lipopolysaccharide (LPS) (purified from E.coZi Olll:B4 by the Westphal method) from Difco Laboratories, Detroit, MI; Lymphoprep, from Nycomed, Oslo, Norway; and hexadecyl platelet-activating factor (PAF) was obtained from Novabiochem, Switzerland. All other chemicals and reagents were of analytical grade or higher and obtained from standard commercial sources. PMA, FMLP, and A23187 were stored at -70°C as stock solutions of 10’M in dimethylsulfoxide and were diluted in buffer immediately prior to use. PAF was stored at 20°C in aliquots of 10e3M in phosphate-buffered saline containing 0.25% bovine serum albumin. Zymosan was washed in phosphate-buffered saline (PBS) and boiled for 30 min before use. Cells: Human monocytes were isolated from fresh buffy coat preparations provided by the Red Cross Blood Bank, Melbourne. Thirty ml of buffy coats, diluted 1 in 3 with sterile 0.9% NaCl, were layered over 10 ml pyrogen-tested Lymphoprep (Nycomed, Oslo, Norway) and centrifuged (30 min, 500 g). The upper layer containing the mononuclear cells was collected and washed three times (5 min, 500 g) to remove excess platelets. The cells were then counted and plated into 24-well tissue culture plates at 3 X 106 nucleated cells per well in 1 ml of (r-MEM. After 2hr at 37”C/5%C02, the nonadherent cells were removed by washing three times with 0.5 ml PBS, the medium replaced with 1 ml a-MEM containing 10% FCS, and the adherent cells incubated overnight (37”C,5%COJ. Murine resident peritoneal macrophages were obtained as previously described (19). All cells were isolated and cultured under strict endotoxin-free conditions as previously described (19). Cytokine treatment: Adherent human monocytes or murine resident peritoneal macrophages were washed (3 X 0.5 ml PBS) and the medium replaced with 1 ml CY-MEM containing 10% FCS and appropriate concentrations of cytokine or LPS (as indicated) and incubated for 48 hr (37”C, 5%COJ. 728
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
burst: The respiratory burst of adherent macrophages was assessed by the stimulus-induced reduction of cytochrome c as previously described (19).
Respiratory
RESULTS AND DISCUSSION Both PMA (10e6M) and zymosan (1 mg/ml) are able to induce a substantial respiratory burst in untreated human monocytes (582&-96 nmollmg
protein, n=18,
and 356+87
nmol/mg, n= 16, respectively). Pretreatment of the monocytes with IL-4 for 48 hr suppressed the respiratory burst produced in response to stimulation with either PMA or zymosan in a dose-dependent fashion (Figure l), a result consistent with previous reports (6,7). A similar reduction in the respiratory burst was also observed in IL-4-treated (10 U/ml, 48 hr) monocytes stimulated with the calcium ionophore A23187 (10e6M) (control, nmol/mg, n=lO;
133+39
IL-4-treated, 39&12 nmol/mg, n=5). This inhibitory effect of IL-4 (10
U/ml) could be completely prevented by preincubation of the IL-4 with a neutralising antibody (100 rig/ml) (n=2) or by heating the IL-4 to 100°C for 30 min (n=3). Interestingly, neither PAF (10e7M) or FMLP (10e7M) were able to induce a respiratory burst in the untreated control monocytes. However, it was found that the monocytes could be primed for a response to these agents by pretreatment with TNFol, IFN-7 or LPS for 48 hr (Figure 2). To our
01
"0
IL-4 "
concentration .l 1
10
02
control
TNF
IFN
IL-4
(U/ml)
Figure 1. Dose-dependent effect of IL-4 on the respiratory burst of human monocytes. Human monocytes were pretreated 48 hr with various concentrations of human IL-4. The respiratory burst was assessed by cytochrome c reduction in response to WM PMA (-o-) or 1 mglml zymosan (--o--). Data are expressed as the percentage of the response of control cells pretreated with medium alone. Shown are mean + S.E.M. from three independent experiments. Fieure 2. Priming the monocytes respiratory burst with cytokines. Human monocytes were treated for 48 hr with medium alone (control), TNFa (l@M), IFN-y (100 U/ml), IL-4 (10 U/ml) or LPS (100 rig/ml). The cells were then washed and the respiratory burst assessed in response to stimulation with 10.‘M PAF (open bars) or 10.‘M FMLP (cross-hatched bars). Shown are mean + S.E.M. from at least three independent experiments.
729
Vol.
182, No. 2, 1992
BIOCHEMICAL
AND RIOPHYSICAL
RESEARCH COMMUNICATIONS
knowledge, this is the first report that priming is necessary to enable human monocytes to produce a respiratory burst in response to PAF or FMLP
and this may reflect our strenuous
efforts to minimise any contamination of our cell cultures with bacterial endotoxin, a common laboratory contaminant with potent priming effects on monocytes (20). Unlike TNFq
IFNy,
or LPS, IL-4 was not able to prime human monocytes for a
respiratory burst response to PAF or FMLP (Figure 2). However, IL-4 was found to antagonise the priming effects of TNFa (Figure 3). A similar antagonism of the priming effects of TNFa! on the A23187-induced respiratory burst was also observed (data not shown). Whether this was simply the net result of independent and opposing actions of IL-4 and TNFo or, alternatively, due to IL-4 directly antagonising the action of TNFcr, cannot be discerned by our current data. However, the observation that IL-4 suppressed the respiratory burst in response to all stimuli tested, irrespective of priming, would perhaps favour the former possibility. In contrast to the suppressive effects of IL-4 on human monocytes, we found that murine IL-4 enhances the respiratory burst of murine resident peritoneal macrophages stimulated with PMA
(Figure
4). This finding
is consistent
with the reported
priming
effects of murine
IL-4
on bone marrow-derived macrophages (4) and peptone-elicited peritoneal macrophages (5).
SC .F .g sz
0
40
9.5
20
icij 3%
e
0
0 control
3 Figure
3. Effect
IL-4
of IL-4
TNF
TNF+
IL-4
and TNFc
2Qo L
100 0
4
on the PAF-induced
control
IL-4
respiratory
TNF
burst.
TNF+IL4
Human
monocyteswere pretreated for 48 hr with medium alone (control), IL-4 (10 U/ml), TNFar (10m9M), or TNFa and IL-4 together. The cells were then washed and the respiratory burst assessed in response to IQ’M PAF. Shown are mean + S.E.M. from six independent experiments. Figure 4. Effect of IL-4 and TNFu on the PMA-induced respimtory burst of mu&e macrophages. Murine resident peritoneal macrophages were pretreated for 48 hr with medium alone (control), IL-4 (5 U/ml), TNFo (W’M), or TNFa and IL-4 together. Cells were then washed and the respiratory burst assessed in response to 10e6MPMA. Shown are mean f S.E.M. from five independent experiments. (PcO.05 for each pretreatment compared to control; P
