expression of interferon-@ during the triggering phase of

THE JOURNAL OF BIOLOGICAL CHEMISTRY 8 1991 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 266, No. 36, Issue of December 25, pp. 24785-24792, 1991 Prrnted in U.S.A.

Expression of Interferon-@ during the Triggering Phase of Macrophage Cytocidal Activation EVIDENCE FOR AN AUTOCRINE/PARACRINE ROLE IN THE REGULATION OF THIS STATE*

(Received for publication, April 15, 1991)

David W. H. RichesSSV and Gregory A. Underwood11 From the $Division of Basic Sciences, Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, Denver, Colorado 80206 and the §Department of Biochemistry, Biophysics and Genetics and the ( 1 Division of Pulmonary Sciences, Department of Medicine, University of Colorado Health Sciences Center, Denver, Colorado 80262

The expression of cytocidal activity is induced by the sequential interaction of macrophages with a priming stimulus, such as interferon (IFN)a, -8, or -y, and a triggering stimulus, such as poly(1.C) or lipopolysaccharide. However, most triggering stimuli are also capable of inducing IFN expression. This suggested to us the possibility that in addition to its role in initially priming macrophages for cytocidal activity, IFN may also be expressed during the triggering stage where it may potentially contribute to the regulation of cytocidal activity. We have explored this question by (i) attempting to dissociate IFN-inducing activity from triggering activity with a variety of structurally related and charge-related polyanions; (ii) determining if macrophages express IFN during the triggering stage; and (iii) questioning if IFN produced during the triggering stage contributes to the regulation of cytocidal activation. Exposure of unprimed macrophages to a triggering concentration of poly(1.C) alone failed to induce IFNj3 expression. However, exposure of IFNj3-primed cells to poly(1. C) dramatically increased the expression of IFNj3 mRNA. Priming with IFNy was likewise found to increase the expression of IFNj3 mRNA in response to a triggering concentration of polyribonucleotides. Three approaches were adopted to ascertain if the increased expression of IFNj3 con- tributed to cytocidal activation. First, macrophages derived from strains of mice which differ in their susceptibility to IFN induction by poly(1.C) were primed with IFNj3, washed, and triggered with poly(1. C). Under these conditions, macrophages derived from stain BlO.A(2R), which are hyporesponsive to poly(1. C ) in terms of IFN induction, also showed a diminished capacity to express Bf, a marker of cytocidal activation. Second, exposure of IFN-primed macrophages to poly(1. C) in the presence of anti-IFNa/b antibody was found to reduce substantially the synthesis of NOz/N03, an alternative marker of macrophage cytocidal activation. Third, exposure of IFN-primed macrophages to the calcium ionophores ionomycin or A23 187, which do not induce the production of IFNj3 during triggering,

* This work was conducted in the F. L. Bryant Laboratory for the Study of Mechanisms of Lung Diseases and was supported by Public Health Service Grants CA 50107 and HL 27353. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertzse- ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

T To whom correspondence and reprint requests should be sent: Dept. of Pediatrics, Neustadt Bldg., Rm. D405, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson St., Denver, GO 80206. Tel.: 303-388-4461.

led to an abbreviated expression of Bf compared with stimuli that induce IFNj3 expression such as poly(1. C). However, the capacity to synthesize Bf in response to A23187 was partially reconstituted when macrophages were triggered with the ionophore in the continuous presence of IFNj3. Collectively, these data show that IFNj3 is expressed during the triggering stage of macrophage cytocidal activation and suggest that it plays an important and previously unsuspected role in the expression of this state.

Macrophage cytocidal activation is a pathway of development which endows macrophages with the capacity to recog- nize and destroy virally and neoplastically transformed cells. Observed in 1971 by Alexander and Evans (1) as a single response of mouse peritoneal macrophages to pg quantities of LPS’ or naturally occurring and synthetic double-stranded RNAs, later work revealed this pathway to comprise two distinct stages (2-4). In the first stage, macrophages become primed upon exposure to a suitable stimulus, the most well characterized being the IFNs (a , p, and y), which serve to increase the sensitivity of the cells to a second, triggering stimulus. Full cytocidal activity is expressed when primed macrophages encounter low (pg-ng) quantities of a triggering stimulus, which can take the form of LPS (2-4), double- stranded RNAs ( 5 ) , heat-killed Listeria rnonocytogenes (6), and cytokines (7,8).

The relationship between the single response originally observed and the two-stage response has been revealed collectively in several laboratories (5 , 9, 10). The observation that most activating stimuli (e.g. double-stranded RNAs, LPS) also induce IFN expression suggested that macrophage-derived IFN may be playing a role in the regulation of cytocidal activation (9). Subsequent studies have shown that an early response of macrophages to pg quantities of triggering stimuli is the expression of IFNa/P ( 5 ) , and, based on the observed inhibition of cytocidal activation by anti-IFNa/P antibodies ( 5 , lo), local production of this cytokine appears to prime macrophages in an autocrine or paracrine fashion thereby rendering the cells sensitive to triggering agents. However, although this initial requirement for IFN expression in response to pg quantities of triggering stimuli can be circum- vented by added IFNs, the stimuli that trigger IFN-primed macrophages to express cytocidal activity nevertheless remain prominent in their ability to induce IFN. This parallelism between triggering activity and IFN-inducing activity sug-

The abbreviations used are: LPS, lipopolysaccharide; IFN, interferon; HBSS, Hanks’ balanced salt solution.

24785

24786 Triggering-associated Induction of IFNP by Macrophages

gested to us the possibility that in addition to its role in initially priming macrophages for cytocidal activity, IFN may also be expressed during the triggering phase, where it may potentially contribute to the regulation of macrophage cytocidal activity. We have explored this question in three ways. First, we attempted to dissociate triggering activity from IFN- inducing activity by exposing unprimed and IFN-primed macrophages to a wide variety of structurally related and charge- related polyanions which differ in their IFN-inducing activity, reasoning that if the hypothesis was incorrect, by priming the cells with IFN, we may circumvent the need for IFN production by the macrophages themselves. Second, we determined if macrophages express IFN during the triggering stage. Third, we explored the possible involvement of IFN in the regulation of the triggering stage by (i) applying genetic differences in IFN production between different mouse strains to determine the capacity of macrophages to express markers of cytocidal activation, and (ii) investigating the effects of anti-IFNa/P antibodies on the expression of these markers. The markers employed were the complement component Bf and the reactive nitrogen intermediates N02/N03. Previous studies have shown their synthesis to be linked to the expression of cytocidal activity (10-15). Our results show that IFNP is expressed during the triggering stage of macrophage cytocidal activation and suggest an important and previously unsuspected role in the positive regulation of this response.

EXPERIMENTAL PROCEDURES

Materials

Animals-The following strains of mice were purchased from the Jackson Laboratory, Bar Harbor, M E C3H/HeJ, BALB/c, A/J, C57BL/6 and BlO.A(2R). All mice were females and of ages ranging between 8 and 12 weeks. Upon arrival at the center, the animals were placed in a virus-free barrier room in the Animal Care Facility and were used after 2 weeks.

Reagents-Dulbecco’s modified Eagle’s medium was obtained from Hazelton Laboratories, Denver, PA and was supplemented immedi- ately before use with 2 mM L-glutamine, 100 units/ml penicillin, and 100 pg/ml streptomycin. Fetal bovine serum was purchased from Irvine Scientific, Santa Ana, CA. Twelve-well (22-mm-diameter) tissue culture plates and 100 X 20-mm tissue culture dishes were purchased from Costar, Cambridge, MA. Trans3%-label (>1,000 Ci/ mmol) and [a-32P]deoxycytidine 5”triphosphate (>3,000 Ci/mmol) were obtained from ICN Biomedicals, Inc., Costa Mesa, CA. Goat anti-human Bf IgG was obtained from Atlantic Antibodies, Scarbor- ough, ME. Mouse L-cell derived IFNp (specific activity 1.3 X lo7 units/mg) was purchased from Lee Biomolecular, San Diego, CA. Poly(I.C), poly(G.C), poly(A.U), poly(A.U.U), sulfanilamide, and naphthylethylenediamine hydrochloride were obtained from Sigma. Poly(d1. dC) and poly(dG. dC) were purchased from Pharmacia LKB Biotechnology Inc. All other reagents were of the highest possible purity.

Methods Cells-Mouse bone marrow-derived macrophages were obtained by

culturing femoral and tibial bone marrow from the various strains of mice in Dulbecco’s modified Eagle’s medium supplemented with 2 mM L-glutamine, 100 units/ml penicillin, 100 pg/ml streptomycin, 10% (v/v) heat-inactivated fetal calf serum, and 10% (v/v) L-cell conditioned medium (referred to as BM medium) using the method of Stewart (16) as modified by Riches et al. (10). The bone marrow cells were initially seeded at a density of 0.8 X lo6 nucleated cells/ well in 22-mm-diameter wells for biosynthetic labeling experiments and 1.1 X IO7 cells/dish in 100-mm-diameter plates for RNA analyses. The cultures were maintained under an atmosphere of 10% CO, for 5-7 days to obtain confluent monolayers of macrophages. Macro- phages monolayers were incubated with the various stimuli in BM medium. For most experiments, optimal expression of Bf was attained in response to 1 pg/ml poly(1.C) in the absence of added IFN. The concentration of poly(1.C) employed to trigger the expression of Bf in IFN-primed macrophages was between 1 and 5 ng/ml. In the

absence of added IFN, this concentration of poly(1. C) induced mini- mal expression of Bf and N02/N03. Apart from the studies that utilized different inbred and recombinant strains of mice, all other studies were conducted with bone marrow macrophages from C3H/ HeJ mice.

Analysis of Bf Protein Expression-The expression of Bf was determined by biosynthetic pulse labeling with 100 pCi of [35S] methionine for 1 h followed by immunoprecipitation of the labeled Bf with anti-human Bf antiserum (10, 17). The immunoprecipitates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis through 7.5% gels, and the radioactive bands were localized by fluorography. Each band was then excised, hydrolyzed in 15% hydrogen peroxide at 75 “C for 18 h, and counted in a liquid scintil- lation spectrometer as reported previously (10, 18). Counts/min were corrected to dpm after determination of the degree of quenching within each sample and after accounting for the decay of the 35S between the experiment and the analysis. Incorporation of [35S] methionine into total protein was determined by precipitation of samples of the cell lysates with 10% (w/v) trichloroacetic acid in filter paper discs using the method of Roberts and Peterson (19).

Quantification of NOz/NO,-The accumulation of NO2/NO3 in the culture supernatants of macrophages was quantified in 96-well plates using the Greiss reagent as described by Ding et al. (20). One hundred- p1 aliquots of the culture supernatants were dispensed in duplicate into Falcon 3912 Microtest 111 plates and mixed with 100 pl of a solution composed of 1% (w/v) sulfanilamide, 0.1% (w/v) naphthylethylenediamine hydrochloride, and 2.5% (v/v) concentrated H3P04. A standard curve consisting of 0.1-5.0 nmol of NaN02/100-pl was prepared in BM medium. After incubation at ambient room temper- ature for 10 min, the absorbances of the wells were quantified at 550 nm in a Biotek Instruments enzyme-linked immunosorbent assay plate reader interfaced to an Apple IIe microcomputer. The number of cells/well was determined by lysing the cell monolayers in Zapo- globin and quantifying the number of released nuclei with a model ZM Coulter counter. Concentrations of NO,/NOs were interpolated from the NaN03 standard curve, corrected for the volume of the culture supernatant, and normalized to the number of cells/well. Results are presented as nmol of NOz/N0s/1O6 adherent cells/unit of time (as defined under ‘‘Results”).

RNA Analyses-The expression of Sf, IFNa,, and IFNP transcripts was determined by Northern analysis. The extraction, purification, electrophoresis, and transfer of the RNA to nitrocellulose membranes were carried out as described (10). Briefly, macrophage monolayers were extracted with 4 M guanidine isothiocyanate (21), and the RNA was purified by centrifugation through a cushion of 5.7 M cesium chloride at 100,000 X g for 18 h. Either 15 pg of total RNA or the poly(A+) RNA derived from 100 fig of total RNA by chromatography on oligo(dT)-cellulose (22) was electrophoresed under denaturing conditions through a 1.2% agarose-formaldehyde gel and then trans- ferred to nitrocellulose filters (23). The blots were hybridized with 5 X 106-5 X lo7 dpm of ”P-labeled cDNA probes (24, 25) and autora- diograms prepared by exposure to Kodak XAR-5 film at -70 “C. The IFNO cDNA probe was obtained from Drs. T. Maniatis and T. Enoch, Harvard Medical School, Boston (26) and was initially isolated and characterized in the laboratory of Dr. T. Taniguchi, Japanese Foun- dation for Cancer Research, Tokyo, Japan (27). The mouse IFNa2 cDNA probe was provided by Dr. P. Pitha-Rowe, Johns Hopkins Medical School, Baltimore, and was originally isolated and characterized by Shaw et al. (28). The mouse Bf cDNA probe has been described by Sackstein et al. (29).

RESULTS

IFNs Prime Macrophages for Increased Bf and NOZlNO3 Expression-The capacity of IFNs to prime for the increased expression of Bf and NOz/NOs is illustrated in Fig. 1. Mac- rophage monolayers were primed by incubation with IFNP (100 units/ml) or IFNr (10 units/ml) for 4 h or incubated in BM medium alone (unprimed). These priming conditions had previously been determined to be optimal with respect to concentration and time. The cells were then washed in HBSS to remove the priming stimulus and triggered with increasing concentrations of poly(1-C) (0.1-1,000 ng/ml) for 24 h before quantifying (i) the level of expression of Bf synthesis by biosynthetic pulse labeling with ~-[~~S]methionine followed by quantitative immunoprecipitation, or (ii) N02/N03 accu-

Triggering-associated Induction of IFNB by Macrophages 24787

7500 A. Complement Bf

Poly [kc] (ng/ml)

FIG. 1. Priming by IFNO and IFNy of the synthesis of the complement component Bf (panel A ) and NOz/NO, (panel B ) by mouse macrophage stimulated with the double-stranded polyribonucleotide poly(1-C). Macrophage monolayers were primed with either IFNP (100 units/ml; panel A ) or IFN-y (10 units/ ml, panel B ) for 4 h or were incubated in BM medium alone, washed in HBSS, and exposed to increasing concentrations of poly(1. C) for 24 h before quantifying the level of synthesis of Bf and N02/N03. Open circles, unprimed macrophages; closed circles, IFN-primed macrophages. TCA, trichloroacetic acid.

mulation in the culture supernatants using the Greiss reagent. I t will be seen that exposure of unprimed macrophages to poly(1.C) led to the synthesis of increasing levels of Bf and of N02/N03 in response to poly(1.C). The concentration- response curves for both parameters of macrophage activation were very similar. When IFN-primed macrophages were triggered with poly(1. C), a marked increase in both the sensitivity of the cells to the triggering stimulus as well as the amplitude of the response was observed. For example, the EDso for poly(1. C) in unprimed macrophages was approximately 7 ng/ ml with a maximum amplitude of 3,600 dpm of Bf whereas the EDso in IFNp-primed macrophages was 0.7 ng/ml with a maximum amplitude of 6,500 dpm of Bf. A similar effect of IFNs on poly(1 .C) sensitivity was seen for N02/N03 accumulation. Priming thus increased both the sensitivity of macrophages to the triggering stimulus and the overall magnitude of the response for both parameters of cytocidal activation.

Effect of Polyanions on Bf Expression-Previous studies (9, 30) have noted the striking parallelism between the ability of pg quantities of stimuli to induce cytocidal activation of macrophages and their capacity to induce IFN expression. Although this association is clearly of critical importance to the autocrine priming of macrophages in the absence of added IFNs (5, lo), we have questioned if this association is also of importance to the triggering phase. We initially attempted to dissociate triggering activity from IFN-inducing activity by exposing unprimed and IFN-primed macrophages to a variety of polyanions with known different capacities for IFN induction, reasoning that priming may circumvent the requirement for any production of IFN by the macrophages themselves.

The polyanions studied consisted of four polyribonucleotides which induce IFN expression in the order poly(1.C) >> poly(A.U) > poly(A.U.U) > poly(G.C) (31-34) and four polyanions which are inactive as IFN inducers, namely (poly(d1. dC), poly(dG. dC), maleylated bovine serum albu-

6000

4500

3000

1500

0

Poly [G:C]

Poly [dG:dC] Poly [dl:dC]

t 8.IFN-Prlmed r 1 I I :? 6000 1 m ' 4500

E, Y u

3000

1500

0

' Poly [I:cl Poly [A:U] Poly [A:U:U] Poly [G.C]

Poly [dl:dC] Poly [dG:dC]

Mal43SA Heparin

0 0.1 1 10 100 1000

Polyanion (ng/ml)

FIG. 2. Responsiveness of unprimed (punel A ) and IFNB- primed (panel B ) macrophages to structurally related and charge-related polyanions. Macrophage monolayers were primed with IFNP (100 units/ml, 6 h) or incubated in BM medium alone, washed in HBSS, and exposed to increasing levels of the indicated polyanions for 18 h before quantifying the level of synthesis of Bf. TCA, trichloroacetic acid; Mal-BSA, maleylated bovine serum albumin.

min, and heparin' (32, 35). Macrophage monolayers were primed with IFNP (100 units/ml) for 6 h or were incubated in BM medium alone before stimulating with the polyanions at concentrations ranging between 0.1 and 1,000 ng/ml for 18 h. As will be seen in Fig. 2A each of the four polyribonucleotides was found to stimulate the expression of Bf in unprimed macrophages, with a rank order of effectiveness of that mir- rored their previously reported IFN-inducing activity with poly(1.C) >> poly(A.U) > poly(A.U.U) > poly(G.C), while the non-IFN-inducing polyanions (poly(d1 .dC), poly(dG. dC), maleylated bovine serum albumin, and heparin) also failed to induce Bf expression. Exposure of IFN-primed macrophages to the polyribonucleotide-triggering agents led to an increase in the sensitivity and the amplitude of the response (Fig. 2B). However, the non-IFN-inducing polyanions remained inactive as triggering molecules in IFN-primed cells. Data similar to these were obtained when N02/N03 was quantified in culture supernatants of macrophages exposed to poly(A. U), poly(A. U. U), and poly(G. C). Thus, the effectiveness of stimuli to trigger Bf expression could not be dissociated from their IFN-inducing activity.

IFNP Expression during Triggering-The association between IFN-inducing activity and triggering activity in IFN- primed macrophages suggested the possibility that IFN expression, in addition to its clearly established role in initially priming macrophages, may be playing an important role in the triggering phase of cytocidal activation. We therefore determined the subtypes and conditions of expression of IFN by macrophages exposed to poly(1. C) and compared the tem- poral relationships of this response with the induction of Bf expression. Macrophages monolayers were incubated in medium alone or in medium containing 1 p/ml poly(1. C) for 3, 6, 12, 24, or 48 h, and the levels of IFNa2, IFNp, and Bf transcripts were determined a t each time point by hybridiza- tion of equal amounts of total or poly(A+) RNA with the appropriate 32P-labeled cDNA probe. As shown in Fig. 3, under these conditions, IFNP transcripts were detected in

D. W. H. Riches, unpublished observation.


Hours 3 6 12 24 48 ”-”

Poly [I:C] - + - + - + - + - +

IFNP mRNA

FIG. 3. Autoradiograph of a Northern blot showing the kinetics of IFNB mRNA expression by mouse macrophages and its relationship to Bf mRNA expression after exposure to poly(1. C ) (1 pg/ml) or BPA medium alone. At the indicated times, poly(A’) RNA was isolated, and equal amounts were subjected to 1.2% (w/v) agarose-formaldehyde gel electrophoresis, blotted onto nitrocellulose, and probed with the appropriate “P-labeled cDNA probe.

A. IFNP mRNA B. Bf mRNA

hr 0 1 3 6 9 1 8 0 1 3 6 9 1 8

Unprimed I I I

FIG. 4. Autoradiography of a Northern blot showing the kinetics of expression of IFNB mRNA (panel A ) during the triggering phase of cytocidal activation and its relationship to Bf mRNA expression (panel B ) . Mouse macrophages were primed with IFNp (100 units/ml, 3 h), washed in HBSS, and exposed to a triggering concentration of poly(1 .C) (2 ng/ml) for the indicated times. Fifteen pg of total cellular RNA was subjected to Northern analysis as described under “Methods.”

abundant quantities after a 3-h exposure of the cells to poly(1. C), peaked after 6 h, had declined to low levels by 24 h, and were virtually undetectable after a 48-h exposure to poly( I a

C). In contrast, probing with an IFNa2-specific cDNA probe failed to provide any indication that the transcript for this IFN subtype was expressed in response to challenge with poly(1.C) (data not shown). Bf transcripts were only barely detectable at 3 h and required 24 h of exposure to the stimulus before maximal expression was attained. Neither IFNa2, IFNP, nor Bf transcripts were detected in unstimulated macrophages (Fig. 3). Thus exposure of unprimed macrophages to poly(1. C) led to the restricted expression of IFNP with a time course that preceded that of Bf.

We next questioned if macrophages expressed IFNP under triggering conditions. Macrophage monolayers were primed with 100 units/ml IFNP for 3 h or incubated in medium alone, washed in HBSS, and then triggered with poly(1-C) (2 ng/ ml) for various time intervals up to 18 h before quantifying IFNP mRNA by Northern analysis. As illustrated in Fig. 4A, exposure of unprimed macrophages to poly(1.C) induced a low level of IFNP mRNA expression which was only detectable at the 3-h time point. By contrast, exposure of IFN-primed macrophages to poly(1.C) resulted in a dramatic increase in the rate of induction of the IFNP mRNA, with levels being detectable after 1 h of exposure to poly( I . C). Significantly, the amount of IFNP mRNA detected at the 3-h time point (Fig. 4B) was considerably greater than seen in unprimed cells. The response was transient in nature and had returned toward base-line levels after 6 h of triggering with poly(1. C). As will be seen in Fig. 4,A and B, the kinetics of IFNP mRNA expression also preceded that of Bf mRNA. Bf expression in

unprimed cells was not detected until 6 h after the addition of poly(1. C). By contrast, the expression of Bf in IFN-primed macrophages was detected after only 3h, and the overall amplitude of the response was greatly increased over that seen in unprimed cells.

The ability of IFNP to prime for its own expression by macrophages was not limited to poly(1. C) but was also seen when a triggering concentration of poly(A.U) (100 ng/ml) was employed (Fig. 5A). Furthermore, priming with IFNr (10 units/ml) enhanced the expression of IFNP mRNA in response to a triggering concentration of poly(1-C) (Fig. 5B). Thus IFN priming of macrophages resulted in a transient but greatly enhanced expression of IFNP mRNA when the cells encountered a suitable triggering stimulus.

Effect of the ZFN Phenotype on the Expression of Bf-The ability of IFNs to prime for the expression of IFNP has been observed in other cells types, and hence it was important to determine if this response played a role in the expression of parameters associated with cytocidal activation in macrophages. To explore this issue, two experimental strategies were adopted. First, we investigated the effects of poly(1. C) on the production of Bf in mice which differ in their IFN loci and thus produce varying amounts of IFN in response to inducing stimuli. As previous studies have shown (36), BlO.A(BR) mice are deficient in their ability to express IFN in response to poly(1. C) and hence comprise a low responder phenotype. However, BALB/c, C3H/HeJ, C57BL/6, and A/J mice, although differing at the If-I locus and hence expressing different amounts of IFN in response to Newcastle disease virus (37, 38, 53), display a standard phenotype with respect to IFN induction by poly(1.C) (36). When BM macrophages established from these four strains of mouse were stimulated with increasing concentrations of poly(1. C) (0.3-1,000 ng/ml, 24 h), as expected, no significant differences were observed in the expression of Bf between macrophages derived from C3H/ HeJ, BALB/c, C57BL/6, or A/J mice (Fig. 6). However, macrophages derived from BlO.A(2R) mice showed a markedly diminished production of Bf in response to poly(I1 C) in terms of both the sensitivity of the cells to the stimulus as well as in the overall magnitude of the response (Fig. 6). The

A.

IFNP - + - + - +

IFNP mRNA

-

FIG. 5. Panel A, IFNp priming of IFNp mRNA accumulation during triggering with poly(A.U). Macrophages were primed with IFNp (100 units/ml, 3 h) and triggered with either poly(1.C) (2 ng/ ml), poly(A.U) (100 ng/ml), or BM medium alone for 4 h before harvesting the cells for RNA analysis. Panel B, IFNp mRNA accumulation after priming with IFNy and triggering with poly(1.C). Macrophage monolayers were primed with IFNp (100 units/ml) or IFNy (10 units/ml) for 3 h and triggered with poly(1.C) (2 ng/ml) for 4 h before harvesting the cells for RNA analysis. In each experiment, equal amounts of poly(A’) RNA were electrophoresed through a 1.2% agarose-formaldehyde gel, blotted, and probed with the IFNB cDNA probe.

Triggering-associated Induction of IFNP by Macrophages 24789 4000

11-1 Poly [1:C] E

3000 1 L e 2000 t- At/ 1000 t- hY /

0 1 10 100 1000 Poly [ I C ] (ne/ml)

FIG. 6. Effect of poly(1.C) on the induction of Bf expression in macrophages derived from five strains of mice which differ in their susceptibility to IFN induction in response to poly(1. C). Indicated are the If-1 phenotypes and the poly(1.C) phenotypes derived from published data (35-37). 0, BALB/c; 0, A/J; 0, C3H/ HeJ; A, C57BL/6; W, BIO.A(ZR). Macrophages were exposed to the indicated concentrations of poly(1.C) for 24 h before quantifying the level of synthesis of Bf. TCA, trichloroacetic acid; N.D., not determined.

” 0 0.1 1 10 100

Poly [IC] (nglml) FIG. 7. Effect of IFNj3-priming on the capacity of macro-

phages from C3H/HeJ and BlO.A(PR) mice to express Bf. Macrophage monolayers were primed with IFNp (100 units/ml, 4 h), washed in HBSS, and exposed to the indicated concentrations of poly(1. C) for 18 h before quantifying Bf synthesis. TCA, trichloroacetic acid.

same trends were observed for N02/NOs accumulation in the culture supernatants, indicating that these differences were not restricted to Bf (data not shown). Moreover, the diminished expression of Bf by macrophages derived from BlO.A(ZR) mice was also seen when the cells were primed with IFNP (100 units/ml, 4 h), washed in HBSS, and triggered with increasing concentrations of poly(1. C) (Fig. 7). Indeed, approximately 10-fold greater quantities of poly(1. C) were required to induce a similar response in macrophages derived from BlO.A(2R) mice compared with that seen with cells derived from C3H/HeJ mice. These data thus support the contention that the triggering-associated expression of IFNP plays an important role in the expression of Bf.

Effect of Anti-IFNa/P Antibodies on the Expression of Trig- gering Parameters-As a second approach to determining the potential importance of the triggering-associated expression of IFNB on the synthesis of markers of the cytocidal state, we have determined if anti-IFNa/P antibodies, present during the triggering stage, inhibit the expression of N02/NOs. Mac- rophage monolayers were primed with either 100 units/ml IFNP or 10 units/ml IFNr for 4 h, washed, and then triggered with poly(1. C) (2 ng/ml) for 48 h in the presence or absence of 5,000 neutralizing units of anti-IFNa/P antibody before

L

Poly [kc] 10 nglml

0 5 10 15 20 25 N02/N03 (nmol/106cells/48 hr)

FIG. 8. Effect of anti-IFNalj3 antibody on the production of N02/N03 by mouse macrophages. Panel A, unprimed macrophages were incubated in medium alone or were exposed to a subop- timal concentration of poly(1.C) (10 ng/ml) in the presence or absence of sufficient anti-IFNn/P antibody to neutralize 5,000 units of IFNa/p for 24 h before quantifying NO,/NO:’ accumulation in the culture supernatants. Panel R, macrophages were primed with IFNr (10 units/ml, 4 h), washed in HBSS, and exposed as described above to poly(1.C) in the presence and absence of anti-IFNn/B antibody.

Triggering Sftmulus Poly [kc] A23187 Medium

FIG. 9. IFNs j3 and y prime for the expression of IFNj3 during triggering with poly(1.C) but not by the calcium ionophore A23187. Monolayers olBM macrophages were primed with IFND (100 units/ml), 1FN-y (10 units/ml), or medium alone for 3 h before being washed and triggered with either poly(1.C) (2 ng/ml), A23187 (0.5 p ~ ) , or medium alone for 3 h. The cells were lvsed and analysed for the expression of IFNp mRNA as described under “Methods.”

quantifying the concentration of NOs/NO:’ in the culture supernatants. As will be seen in Fig. 8A, exposure of macrophages to a triggering concentration of poly( I. C) alone resulted in the accumulation of a low level of NOr/NOs in the culture supernatants which, in the continuous presence of anti-IFNa/P antibodies, was totally inhibited. After priming with IFNy, the accumulation of NO2/NO:% in response to poly(1.C) was markedly enhanced. However, when IFN- primed macrophages were triggered with poly(1-C) in the presence of anti-IFNa/P antibody, the accumulation of NO2/ NOn in the culture supernatant of IFNy-primed macrophages was dramatically inhibited (Fig. 8B). Similar results were obtained when macrophages were primed with IFNP (data not shown). The inhibition of N02/NOs accumulation did not appear to be the result of nonspecific effects of antigen- antibody complexes since triggering in the presence of antibodies reactive against two other secretory proteins which are expressed during the triggering phase (Bf and C3) failed to inhibit the accumulation of N02/N03 (data not shown).

Triggering Activity of the Calcium Ionophore A23187”Re- cently conducted studies in our laboratory have shown that the calcium ionophores A23187 and ionomycin trigger the expression of Bf and NO2/NO, in IFN-primed macrophages but not in unprimed cells.:’ Moreover, as will be seen from Fig. 9, in contrast to poly(1. C), triggering of IFNP or 1FN-y- primed macrophages with an optimal concentration of A23187 (0.5 ~ L M ) failed to induce the expression of IFNP mRNA. These findings therefore provided an opportunity to determine the role of the triggering-associated expression of IFNP.

G. A. Underwood and D. W. H. Riches, submitted for publication.


We reasoned that since IFNP is not produced during triggering with the calcium ionophores ionomycin and A23187, differences in the amplitude or time course of the expression of Bf or N02/N03 may provide an insight into the role of the IFN produced during triggering. Macrophage monolayers were therefore primed with IFNP and were then washed and exposed to ionomycin (5.0 p ~ ) for 24 or 48 h before quantifying the level of synthesis of Bf. As will be seen in Fig. lOA, under these conditions, the synthesis of Bf is maintained for 48 h in macrophages that were triggered with poly(1.C) but had returned toward base line in cells that were triggered with ionomycin. Identical results were obtained with the A23187. Importantly, as will be seen in Fig. 10, exposure of IFN- primed macrophages to A23187 (1 p ~ ) in the continuous presence of purified IFNP (1,000 units/ml) approximately doubled ( p = 0.003) the level of synthesis of Bf compared with macrophages triggered with A23187 in the absence of added IFN. By contrast, there was no significant effect ( p = 0.512) of purified IFNP (1,000 units/ml) on the expression of Bf in response to triggering with poly(1- C) (2 ng/ml) (Fig. 10B).

DISCUSSION

The question of the level of involvement of IFNs in macrophage cytocidal activation has been wrestled with for over a decade. The possibility that IFNs may stimulate the expression of cytostatic and cytocidal activity originally stemmed from the observations that a variety of polyanionic agents, such as poly(1. C), and bacterial products, such as LPS, could both induce the expression of IFNa/P and activate macrophages (1, 30). Subsequently, it was shown that the addition of purified IFNs (both a/P and y) to macrophage monolayers could directly activate macrophages to express cytocidal activity and, moreover, that incubation of macrophages with poly(1. C) or LPS in the continuous presence of anti-IFNa/P antiglobulin led to an inhibition of the expression of cytocidal activity (9, 40, 41). These findings led to the suggestion that IFNs, by themselves, were capable of inducing this important macrophage response. However, the interpretations of these early studies were drawn into question after the realization that LPS contamination of experimental materials was probably contributing to the expression of cytocidal activity (3,

48 hr [ Poly [IC] lonomycin I

1 B. I

[ +IFNP -IFNP

I I 0 1000 2000 3000 4000 5000

Ef Synthesis (dpml107Tolal TCA ppt. dpm)

FIG. 10. Panel A , synthesis of Bf by IFN-primed macrophages triggered with ionomycin (5 PM) or poly(1-C) (1 ng/ml). At the indicated time points, the cells were biosynthetically labeled with ["'S]methionine and the level of Bf synthesis determined by quantitative immunoprecipitation as described under "Methods." Panel B, reconstitution of the level of synthesis of Bf induced by A23187 during triggering in the presence of purified IFNO. IFN-primed macrophages were triggered with A23187 (1 NM) in the presence (light shading) and absence (dark shading) of purified IFNO (1,000 units/ ml) for 36 h. TCA, trichloroacetic acid.

42). It has subsequently been shown that IFNs are a necessary but not sufficient stimulus for cytocidal activation, and firm evidence has now been accrued to show that IFNs contribute to cytocidal activation by priming macrophages and thereby increasing the sensitivity of the cells to a second triggering stimulus, such as LPS, poly(I.C), or heat-killed L. monocy- togenes (2, 4, 6, 43).

However, although there is no reason to dispute the role that IFNs play in initially priming macrophages to respond to triggering stimuli, the findings reported herein emphasize the striking resemblance between triggering activity and IFN- inducing activity of most of the stimuli that are known to induce the expression of the macrophage cytocidal state. We therefore explored the concept that IFNP may be expressed during the triggering phase and may be contributing to the regulation of cytocidal activity.

Exposure of macrophages to pg quantities of poly(1. C) led to an accumulation of IFNP transcripts that peaked 6 h after stimulation and declined thereafter. IFNa2 transcripts were not found to be expressed under these conditions. These findings are consistent with previously reported studies in terms of the kinetics of induction of IFN activity in mouse peritoneal and BM macrophages during exposure to LPS (44- 47), poly(1-C) (44), and Newcastle disease virus (44, 47) and in terms of the IFN subtype produced. Exposure of unprimed macrophages to a triggering concentration of poly(1.C) was found to induce a low level of expression of IFNP mRNA. However, triggering of IFN-primed macrophages dramatically increased the level of expression of IFNP mRNA as well as shortening the lag time required for induction. The triggering stage-associated expression of IFNB mRNA was transient in nature, being essentially complete by 6 h. However, secreted IFNp protein is relatively stable in culture medium and hence would be expected to be available to the macrophages after the IFNP mRNA levels had declined. Importantly, and of relevance to the potential role of the triggering-associated production of IFNP, IFNy was also found to be an effective priming stimulus of this response under conditions that are optimal for the development of macrophage cytocidal activity.

The ability of IFN to prime for its own production was originally reported by Isaacs and Burke (48) and has since been observed in many cell types (49-51) using both viruses and polyribonucleotides as inducing stimuli. In addition, Hav- ell and Spitalny (45) have shown that crude lymphokine preparations derived from phytohemagglutinin-stimulated mouse spleen cells, which probably contain IFNy, also prime mouse peritoneal macrophages for the increased production of an IFN activity following challenge with endotoxin. Gessani et al. (52) have recently reported similar observations using purified recombinant IFNy. However, these observations have generally been considered in the context of the role of IFN in the induction of antiviral protection. Macrophages do differ from many of the epithelial and mesenchymal cells employed in the antiviral literature by their capacity to express many other complex functions that underlie their role in host defense and protection, including the expression of cytocidal activity. I t was important therefore to determine if the triggering stage-associated production of IFNP contrib- uted to the regulation cytocidal activity or was a more general response of cells to IFNs.

First, we quantified the expression of Bf in macrophages derived from strains of mice which differed in their capacity to produce IFN in response to poly(1. C). When unprimed macrophages were exposed to increasing concentrations of poly(1. C), as expected, no differences were seen in the induction of Bf expression between macrophages derived from

Triggering-associated Induction of IFNO by Macrophages 24791

C3H/HeJ, A/J, BALB/c, or C57BL/6 mice. However, the induction of Bf expression in macrophages derived from BlO.A(BR) mice was markedly diminished both in terms of the sensitivity of the cells to poly(1.C) and in the amplitude of the response. Importantly, the diminished ability of macrophages derived from BlO.A(BR) mice to express Bf was maintained when the cells were primed with exogenous IFNP, washed to remove the priming stimulus, and then triggered with poly(1. C). I t should be noted that H-2 linked polymorphisms in Bf functional activity have been reported (54, 55) although these genetic polymorphisms do not influence the antigenic level of Bf (55, 56). These data are thus consistent with the notion that the IFNp produced during the triggering stage of macrophage cytocidal activation actively contributes to the full expression of this state although the possibility that other genetic differences between BlO.A(2R) mice and C3H/HeJ mice may account for these results cannot be ex- cluded. We are currently attempting to resolve this issue since the addition of purified IFNP during the triggering phase was found to incompletely reconstitute the induction of Bf expression.

Second, we determined the effects of anti-IFNa/P antibodies on the expression of the triggered state. Triggering of IFN- primed macrophages with poly(1.C) in the presence of anti- IFNa/P antibodies was found to block the expression of NO,/ NO, substantially. Furthermore, inhibition of NOz/NO3 production by anti-IFNa/P antibodies was equally effective when macrophages were initially primed with IFNP or with IFNy. These data thus further support the view that IFNp plays an active role in the expression of the triggered state. Previously reported experiments (5, 9, 10) which have probed the role of IFNa/P in macrophage cytocidal activation have been conducted by coincubating the cells with anti-IFNa/P antibodies and activating stimuli. Although such studies have provided important information concerning the overall necessity of IFN in cytocidal activation, such an approach has resulted in the important contribution of IFNP to the triggering stage being previously overlooked. Moreover, an additional impli- cation of these findings is that the interpretation of experiments in which macrophages have been primed and triggered in vitro will be complicated by the inadvertent introduction of IFNO.

Cytocidal activation is transiently expressed (4,57), and we suspected that a potential role of the triggering stage-associated IFNp may be to maintain macrophages in a primed state during triggering. An opportunity to investigate this question was provided by our observation that the calcium ionophores ionomycin and A23187 induce the expression Bf and NO2/ NO:] in IFN-primed but not in unprimed macrophage^.^ How- ever, in contrast to poly(1. C), these ionophores do not induce the expression of IFNP during triggering. Thus differences in the amplitude or time course of expression of Bf or N02/N03 during triggering with calcium ionophores could provide an indication of the role of the triggering stage-associated IFNP. Triggering with poly(1. C), under conditions that were previously shown to result in the expression of IFNP during the triggering stage, resulted in a level of Bf synthesis that was maintained over 48 h. However, synthesis of Bf in macrophages that were triggered with ionomycin or A23181, although showing initial similarities to poly(1.C) at 24 h, had returned toward base-line levels by 48 h. Significantly, when IFN-primed macrophages were triggered with A23187 in the presence of added IFNP, the level of synthesis of Bf was partially reconstituted compared with that seen in the absence of added IFNP. Thus, collectively these data strongly suggest

play a key role in maintaining the cells in a primed state during triggering and thus may influence the duration of expression of cytocidal activity. In addition, based on observed cooperation between IFNs (39), it is possible that in the presence of activated T-cells, the triggering stage-associated IFNp may serve to act cooperatively with IFNy, leading to an amplification of cytocidal activity.

Acknowledgments-We would like to thank Linda Remigio for her invaluable technical assistance; Lori Sander for preparing the macrophages; and Nadia de Stackleberg, Barry Silverstein, and Leigh Landskroner for preparing the photographs. We are also indebted to Drs. Tom Maniatis, Tamar Enoch, and Paula Pitha-Rowe for providing the IFNp and IFNa, cDNA probes, respectively, and to Dr. Tom Hamilton for generously providing the maleylated bovine serum albumin used in this study. Finally, we wish to acknowledge Dr. Peter Henson for his encouragement, support, and critical appraisal of this work.

REFERENCES

1. Alexander, P., and Evans, R. (1971) Nature New Biol. 232, 76-

2. Russell, S. W., Doe, W. F., and McIntosh, A. T. (1977) J. E x p . 78

Med. 146, i511Li520 3. Weinberg. J . B.. ChaDman. H. A.. Jr.. and Hibbs. J. B.. Jr. (1978)

I , I ; _

J . Imrr&ol. 12 1, ?2-80

2042

Commun. 131, 395-401

4. Ruco, L. P., and Meltzer, M. S. (1978) J. Zmmunol. 121, 2035-

5. Torres, B. A., and Johnson, H. M. (1985) Biochem. Biophys. Res.

6. Pace, d. L., Russell, S. W., LeBlanc, P. A., and Murasko, D. M.

7. Heidenreich, S., Weyers, M., Gong, J. H., Sprenger, H., Nain,

8. Mace, K. F., Ehrke, M. J., Hori, K., Maccubbin, D. L., and

9. Schultz, R. M., and Chirigos, M. A. (1979) Cell. Immunol. 48,

10. Riches, D. W. H., Henson, P. M., Remigio, L. K., Catterall, J. F.,

11. Stuehr, D. J., and Marletta, M. A. (1985) Proc. Natl. Acad. Sei.

12. Hibbs, J. B., Jr., Taintor, R. R., and Vavrin, Z. (1987) Science

13. Hibbs, J. B., Jr., Taintor, R. R., Vavrin, Z., and Rachlin, E. M. (1988) Biochem. Biophys. Res. Commun. 157, 87-94

14. Stuehr, D. J., and Nathan, C. F. (1989) J. Exp. Med. 169, 1543- 1555

15. Stuehr, D. J., Gross, S. S., Sakuma, I., Levi, R., and Nathan, C. F. (1989) J. Exp. Med. 169, 1011-1020

16. Stewart, C. C. (1981) in Methods for Studying Mononuclear Phagocytes (Adam, D. O., Edelson, P. J., and Koren, H., eds) pp. 5-20, Academic Press, New York

17. Riches, D. W. H., and Henson, P. M. (1986) J. Cell Biol. 102, 1606-1614

18. Cole, F. S., Auerbach, H. S., Goldberger, G., and Colten, H. R. (1985) J . Immunol. 134, 2610-2616

19. Roberts, B. E., and Paterson, B. M. (1973) Proc. Natl. Acad. Sci. U. S. A. 70, 2330-2334

20. Ding, A. H., Nathan, C. F., and Stuehr, D. J. (1988) J. Imrnunol. 141, 2407-2412

21. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J., and Rutter, W. J . (1979) Biochemistry 18, 5294-5299

22. Aviv, H., and Leder, P. (1972) Proc. Natl. Acad. Sei. U. S. A. 69, 1408-1412

23. Thomas, P. (1980) Proc. Natl. Acad. Sei. U. S. A. 77, 5201-5206 24. Feinberg, A. P., and Vogelstein, B. (1983) Anal. Biochem. 132,

25. Feinberg, A. P., and Vogelstein, B. (1984) Anal. Biochem. 137,

26. Enoch T, Zinn, K., and Maniatis, T. (1986) Mol. Cell. Biol. 6,

27. Higashi, Y., Sokawa, Y., Watanabe, Y., Kawade, Y., Ohno, S., Takaoka, C., and Taniguchi, T. (1983) J. Biol. Chem. 258, 9522-9529

(1985) J. Immunol. 134, 977-981

M., and Gemsa, D. (1988) J. Immunol. 140, 1511-1518

Mihich, E. (1988) Cancer Res. 48, 5427-5432

52-58

and Strunk, R. C. (1988) J. Immunol. 141, 180-188

U. S. A. 82,7738-7742

235,473-476

6-13

266-267

801-810

that the triggering stage-associated production o f IFNpmay 28. Shaw, G. D., Boll, W., Taira, H., Mantei, N., Lengyel, P., and

24792 Triggering-associated Induction of IFNO by Macrophages Weissmann, C. (1983) Nucleic Acids Res. 1 1 , 555-573 43. Hibbs, J. B., Weinberg, J. B., and Chapman, H. A. (1979) Adu.

Chem. 258, 14693-14697 44. Fleit, H. B., and Rabinovitch, M. (1981) Cell-lmmunol. 57, 495-

Science 197, 674-676 45. Havell, E. A., and Spitalny, G. L. (1983) J . Reticuloendothel. SOC.

(1967) Proc. Natl. Acad. Sci. U. S. A. 58, 1004-1010

M. (1968) J. Virol. 2, 886-893

Acad. Sci. U. S. A. 71, 182-186

29. Sackstein, R., Colten, H. R., and Woods, D. E. (1983) J. Biol. Exp. Med. Biol. 121B, 433-453

30. Schultz, R. M., Papamatheakis, J. D., and Chirigos, M. A. (1977) 504

31. Field, A. K., Tytell, A. A., Lampson, G. P., and Hilleman, M. R.

32. De Somer, p., De Clercq, E., Billiau, A., Schonne, E., and Claesen, 47. Belardelli, F., Gessani, S., Proietti, E., Locardi, C., Borghi, P.,

33. De Clercq, E., Torrence, P. F., and Witkop, B. (1974) Proc. Natl. Watanabe, Y., Kawade, Y., and Gresser, I. (1987) J . Gen. Virol. 68,2203-2212

34. Terrence, p . F., and De, c . E. (1984) Antiuiral Res. 4,339-350 49, Stewart, W. E,, G ~ ~ ~ ~ ~ , L, B., and Lockart, R. z. (1971) J, viral, 35. Colby, C., and Chamberlin, M. J. (1969) Biochemistry 63, 160-

36. Pugliese, A., Cortese, D., and Forni, G. (1980) Arch. Virol. 65, 51. Fujita, T., and Kohno, s. (1981) virology 112, 62-69 50. Rosztoczy, I. (1976) Acta Virol. (Praha) 20, 472-478

37, D~ M ~ ~ ~ ~ ~ , E., D~ Maeyer.~uignard, J,, and Bailey, D. w, i1975) 52. Gessani, s., Belardelli, F., Pecorelli, A., Puddu, p., and Baglioni,

38, De Maeyer, E,, H ~ ~ ~ ~ , M. c , , D~ Maeyer-Guignard, J., and ~ ~ i l ~ ~ , 53. De Maeyer, E., Jullien, P., Maeyer-Guignard, D.9 and Demant,

39. Pace, J. L. (1988) J. Leukocyte Biol. 44,514-520 54. Roos, M. H., and Demant, P. (1982) Immunogenetics 1 5 , 23-30 40. Schultz, R. M., Chirigos, M. A,, and Heine, u. 1. (1978) Cell. 55. Paolucci, E. s., and Shreffler, D. c. (1983) Immunogenetics 17,

41. Schultz, R. M., and Chirigos, M. A. (1980) Adu. Pharmacol. 56. Garlepp, M. J., Hart, D. A., and Fritzler, M. J. (1989) J. Clin.

42. Pace, J. L., Taffett, S. M., and Russell, S. W. (1981) J. Reticu- 57. Taramelli, D., Bagley, M. B., Holden, H. T., and Varesio, L.

33,369-380 46. Brehm, G., and Krichner, H. (1986) J. Interferon Res. 6,21-28

48. Isaacs, A., and Burke, D. C. (1958) Nature 182, 1073-1074

167 7,792-801

83-87

Immunogenetics 1,438-445

D. W. (1979) Immunogenetics 8,257-263 P. (1975) Immunogenetics 2, 151-160

C. (1989) J. Virol. 63, 2785-2789

Immunol. 35,84-91 67-78

Chemother. 1 7 , 157-193 Lab. Immunol. 28,137-141

loendothel. Soc. 30, 15-21 (1982) Adu. Enp. Med. Biol. 155, 487-492

expression of interferon-@ during the triggering phase of

Documents