Available via license: CC BY-NC-SA 4.0
Content may be subject to copyright.
Available via license: CC BY-NC-SA 4.0
Content may be subject to copyright.
EVIDENCE FOR A GAMMA-INTERFERON RECEPTOR THAT
REGULATES MACROPHAGE TUMORICIDAL ACTIVITY
BY ANTONIO CELADA,* PATRICK W. GRAY, * ERNST RINDERKNECHTfi
AND ROBERT D.
SCHREIBER*
From the *Department of Immunology, Research Institute of Scripps Clinic, La
Jolla, California 92037; and *Department of Molecular Biology, Genentech, Incorporated,
San Francisco, California 94080
Antigen- or mitogen-stimulated T lymphocytes produce factors that can induce
a number of functional and biochemical modifications in macrophage popula-
tions. These modifications include increases in endocytic, biosynthetic, secretory,
and effector cell functions, as well as changes in membrane physiology and
composition (1-4). This process has been called macrophage activation and the
iymphokines that induce these effects are known as macrophage-activating factors
(MAF).'
Over the past three years, work from several laboratories has indicated that
gamma-interferon (IFN-3,), the [FN produced by T lymphocytes, represents one
type of MAF. Purified or partially purified preparations of IFN-5' produced by
normal T cells, T cell hybridomas, or by recombinant DNA technology have
been found to prime macrophages for nonspecific tumoricidal activity (5-11),
induce or enhance intraceilular cytocidal reactions (12, 13), and increase expres-
sion of Ia or DR antigens on the cell surface (14-16). Using monoclonal
antibodies to murine recombinant IFN-% we have recently shown that IFN-'r
represents the major and possibly only lymphokine produced by concanavalin A
(Con A)-stimulated normal murine splenic cells or the 24/G1 murine T cell
hybridoma that can induce tumoricidal activity or Ia expression in macrophagesfl
These observations have prompted an analysis of the interaction of IFN-7 with
macrophage populations. It is the purpose of this report to describe this inter-
action on a functional and biochemical basis and to demonstrate the existence of
a IFN-~, receptor on macrophages. This receptor is capable of binding either
This work was supported by United States Public Health Service Grants CA 34120 and AI 17354
and by grants from Eli Lilly and Co. and the Elsa U. Pardee Foundation. This is publication number
3399IMM from the Research Institute of Scripps Clinic. Correspondence should be addressed to Dr.
Robert D. Schreiber, Department of Immunology, IMMII, Research Institute of Scripps Clinic,
10666 North Torrey Pines Road, La Jolla, CA 92037.
a Abbreviations used in this paper:
Con A, concanavalin A; DME, Dulbecco's modified Eagle's
medium containing 4,500 mg glucose/liter; EPM, elicited peritoneal macrophage; FCS, fetal calf
serum; HKLM, heat-killed
Listeria monocytogenes;
HPLC, high pressure liquid chromatography; IFN,
interferon; IL-2, interleukin 2; Ka, association equilibrium constant; Kd, dissociation equilibrium
constant; MAF, macrophage-activating factor; PBS, phosphate-buffered physiological saline; PEC,
peritoneal exudate cells; PMN, polymorphonuclear leukocyte.
2 Schreiber, R. D., L.J. Hicks, A. Celada, and P. Gray. 1984. Monoclonal antibodies to IFN which
modulate macrophage activation by lymphokines. Manuscript in preparation.
j. ExP. MEn. © The Rockefeller University Press •
0022-1007/84/07/55/20
$1.00 55
Volume 160 July 1984 55-74
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
56
RECEPTOR FOR INTERFERON-3, ON MACROPHAGES
natural or recombinant murine IFN-7 in a specific and saturable manner.
Receptor engagement is required for the initiation of macrophage activation.
Materials and Methods
Media, Supplements and Buffers. All media, supplements, and buffers used in these
experiments were prepared from endotoxin-free stocks and were determined to be free
of endotoxin using the Limulus amebocyte lysate assay (Sigma Chemical Co., St. Louis,
MO). To destroy endotoxin that potentially might have been adherent to glass, all
autoclaved glassware was baked for 3 h at 180°C. RPMI-1640 and Eagle's minimal
essential medium were prepared from powdered stocks (Flow Laboratories, Inglewood,
CA) using USP sterile water (Travenol Laboratories, Inc., Deerfield, IL). Liquid Dulbec-
co's modified Eagle's medium containing 4,500 mg glucose/liter (DME) was purchased
from M.A. Bioproducts, Walkersville, MD. Aseptically drawn fetal calf serum (FCS,
Rehatuin FS) was obtained from Reheis Chemical Co. (Phoenix, AZ) and heat inactivated
(1 h, 56°C) before use. Other media supplements included injectable penicillin G and
streptomycin sulfate (Eli Lilly and Co., Indianapolis, IN), injectable sodium bicarbonate
and sodium heparin (Gibco-Invenex Division, Chagrin Falls, OH), 1 M Hepes buffer
solution, versene, sodium pyruvate, and L-glutamine (M.A. Bioproducts), 10 mM nones-
sential amino acids and trypsin-EDTA (Ix) (Gibco Laboratories, Grand Island, NY),
gentamicin (Scheering Corp., Kenilworth, N J), indomethacin, trizma base, trizma hydro-
chloride, 13-mercaptoethanol, ethanolamine, and saponin (Sigma Chemical Co.), sodium
chloride, sodium mono- and dibasic-phosphate (Fisher Scientific Co., Fairlawn, N J), bovine
serum albumin, Fraction V (United States Biochemical Co., Cleveland, OH), soybean
trypsin inhibitor (Calbiochem-Behring Corp., La Jolla, CA) and concanavalin A (Con A,
Miles-Yeda, Ltd., Rehovot, Israel).
Animals. C3HeB/FeJ, CBA/CaJ, C57BL/6J, DBA/2J, SJL, and C3H/HeJ mice, and
Lewis rats were obtained from the breeding colony at the Research Institute of Scripps
Clinic. A/J, C3D2F1/J, and Swiss (BKL) were obtained from The Jackson Laboratory,
Bar Harbor, ME. Swiss (NCS) mice were obtained from the breeding colony at The
Rockefeller University, New York, NY. Armenian hamsters were obtained from Cam-
bridge Diagnostics, Cambridge, MA and guinea pigs from Crest Caviary, Raymond, CA.
Lymphokine Preparations. Lymphokine-containing supernatants were prepared by Con
A stimulation of cultures of the murine T cell hybridoma 24/G1 or normal murine splenic
cells as described previously (5).
Recombinant Gamma Interferon. E. coli-derived murine IFN-3, was produced as previ-
ously described (17). A preparation (lot number 1551/43) was used that had been purified
to a specific antiviral activity of 7.2 × 10 ~ IRU/mg. IFN was stored at 4°C in 10 mM
Tris-HCl, 0.5 M NaCI buffer, pH 8.0 containing 0.1% 13-mercaptoethanol.
High Pressure Liquid Chromatography (HPLC) Gel Filtration of Recombinant IFN-7. 180
t~l of the purified recombinant IFN-3, was injected into a 7.5 × 600 mm Bio-Sil TSK250
HPLC gel filtration column (Bio-Rad Laboratories, Richmond, CA) connected to a Waters
HPLC system (Waters Associates, Milford, MA). The column was eluted at 23°C with
0.02 M phosphate-buffered physiological saline (PBS), pH 7.2, at a flow rate of 1 ml/min.
Preparation oflodinated IFN-7. The two-peak HPLC fractions that displayed a molec-
ular mass of 32,000 daltons were pooled and used for radiolabeling. 14 #g of IFN in 100
#1 was incubated with 1 mCi ~25I Bolton-Hunter reagent (18) in a "V"-shaped glass reaction
vial (ICN Chemicals, Radioisotope Division, Irvine, CA) for 2 h at 4°C. Protein-associated
and free 125I were separated by centrifugation through tubes containing BioGel P-6 as
described by Fishelson et al. (19). Plastic tubes (10 × 75 mm, Falcon Labware, Oxnard,
CA) pierced at the bottom with a needle were plugged with scrubbed nylon fibers (Fenwal
Laboratories, Deerfield, IL) and packed with BioGel P-6, 100 to 200 mesh (Bio-Rad
Laboratories) equilibrated in PBS. The tube was then placed inside a 12 x 75-mm plastic
tube, centrifuged to dryness for 3 min at 1,000 rpm in an Adams Sero-Fuge (Clay Adams,
Inc., New York, NY) and the inside tube transferred to a new 12 × 75-mm tube. The
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL. 57
labeled IFN was applied to the top of the BioGel P-6 and the tubes were centrifuged as
above. Protein-bound, but not free, '25I was eluted from the gel by this procedure and
collected at the bottom of the 12 X 75 mm tubes. The P-6 tube was washed once with
100 tA of buffer and the two eluted volumes were combined. Recovery of IFN was
essentially 100% as assessed with the quantitative MAF assay. Typical preparations were
labeled to a specific activity of 7.6 ~Ci/ug. ~25I-IFN-~ was stored at 4°C and retained
biological activity for at least 2 wk.
Alpha and Beta Interferons. These murine IFN were purchased from Lee Biomolecular
Laboratories (San Diego, CA). IFN-a had a specific activity of 5 x 105 IRU/mg, and IFN-
3of 1.5 × l07 IRU/mg.
Peritoneal Macrophages. Peritoneal exudate cells (PEC) were harvested as described
previously by lavage from mice that had been injected intraperitoneally 3 d previously
with 1.5 ml of 10% protease peptone (Difco Laboratories, Detroit, MI) (5). Rats, hamsters,
and guinea pigs were injected with 5% casein 5 d before the PEC were harvested.
Macrophage monolayers were prepared by seeding the PEC suspension into flat-bottom
96-well tissue culture plates (Costar, No. 3696, Cambridge, MA) or 6-well tissue culture
plates (Costar, No. 3506). The cells were adhered for 2 h at 37°C and the plates washed
vigorously to remove nonadherent cells.
Mouse Bone Marrow-derived Macrophages. Macrophages derived from bone marrow
cultures were obtained according to the method of Meerpohl et al. (20) with some
modifications. Mice were killed by cervical dislocation and both femurs were dissected
free of adherent tissue. The ends of the bones were cut off and the marrow tissue eluted
by irrigation with PBS. Cells were suspended by vigorous pipetting and washed by
centrifugation, l0 T cells were cultured in a nontissue culture plastic 150-mm petri dish
(Lab-Tek 4030, Miles Laboratories, Inc., Naperville, IL) in 50 ml of DME containing 2
mM L-glutamine, 1 mM Na pyruvate, 50 U/ml penicillin, 50 #g/ml streptomycin, 20%
heat-inactivated horse serum, and 30% L cell-conditioned medium. The cell suspensions
were incubated at 37°C in a humidified 5% CO~ atmosphere. After 6-9 d, macrophages
that were loosely adherent to the dishes were harvested with cold PBS and used.
Human Cells. Human peripheral blood monocytes, lymphocytes, and polymorphonu-
clear leukocytes (PMN) were purified from heparinized, normal venous blood by discon-
tinuous density gradient centrifugation on FicolI-Urografin (21). The upper layer con-
tained a mixture of monocytes and lymphocytes, while 98% of PMN were present in the
lower layer. Erythrocytes were obtained at the bottom of the tube. Monocytes were
separated from lymphocytes by adherence onto autologous serum-treated plastic plates
according to the method of Fischer et al. (22). The purity of all cell preparations was
>98% as assessed by cytocentrifugation and Giemsa-peroxidase staining (23).
Preparation of Membranes from Bone Marrow-derived Macrophages. Membranes were
prepared according to the procedure of Gabel et al. (24). In brief, cells were disrupted
with a sonicator and the cell lysates centrifuged at 850 g for 10 min. The resulting low
speed supernatant was then recentrifuged at 40,000 g for 30 rain. Membrane pellets were
resuspended by homogenization and washed three times. Membrane preparations were
stored at -70°C.
Cultured Cell Lines. The human histiocytic lymphoma cell line (U0sT) and the murine
mastocytoma cell line (P815) were maintained in RPMI supplemented with 2 mM e-
glutamine, 50 U/ml penicillin, 50 #g/ml streptomycin, 1 mM Na pyruvate, 0.075% (wt/
vol) Na bicarbonate, and 10% heat-inactivated FCS. Murine T cell hybridomas TH4.4,
24/G1, and 1.19 and the murine fibrosarcomas L929, TU5, and 1023 were maintained
in DME supplemented with 2 mM L-glutamine 50 U/ml penicillin, 50 ug/ml streptomycin,
1 mM Na pyruvate, and 10% FCS. Two murine macrophage cell lines were also used:
P388D~ was maintained in RPMI and J774 in MEM supplemented with 2 mM L-glutamine,
1 mM pyruvate, 10 mM nonessential amino acids, 0.075% (wt/vol) Na bicarbonate, 50
ug/ml ofgentamicin, and 10% FCS.
Measurement of MAF Activity. MAF was quantitated by measurement of its ability to
induce nonspecific tumorilytic activity in peptone-elicited C3HeB/FeJ macrophages to-
ward P815 mastocytoma cells, as detailed elsewhere (5) with some modifications, Briefly,
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
58 RECEPTOR FOR INTERFERON-'y ON MACROPHAGES
reactions were performed in duplicate or triplicate in 96-well flat-bottom A/2 tissue
culture plates in a total volume of 100 #1 (Costar, No. 3696). Each well contained 1 x 105
adherent macrophages, serial dilutions of MAF, an excess of a second signal (1 x 106
heat-killed Listeria monocytogenes (HKLM)), 10 -6 M indomethacin and 1 x 104111 In-labeled
P815. llqn-oxine chelate was purchased from Mediphysics, Inc. (Bloomfield, NJ) and
labeling was performed as described by Wiltrout et al. (25). Following incubation for 18
h at 37°C in a humidified 5% CO2 atmosphere, 50 #1 of each culture supernatant was
removed and analyzed for ~*In content. 1 U of MAF is defined as that amount which
produces 50% maximal specific ill in release in this assay.
Absorption Studies. Six-well tissue culture plates (Costar, No. 3506) were used for
absorption of MAF. In all the cases, plates were preincubated overnight with medium
containing 20% FCS before use. Six million adherent cells were incubated with 1 ml of
MAF containing supernatant at 37°C in a humidified 5% COu atmosphere. Supernatants
were collected and filtered through a 0.2-tim filter (Millipore Corporation, Bedford, MA)
to remove cellular debris and the remaining activity quantitated using the tumorilytic
assay. As the control for these experiments, 1 ml of MAF-containing supernatant was
incubated without macrophages in the culture plates for the same period of time.
Measurement of Interleukin 2 (IL-2) Activity. IL-2 activity was quantitated using the
method of Smith (26). Serial dilutions of the sample were cultured with 4,000 HT-2 cells,
the IL-2-dependent murine T cell line for 20 h in a total volume of 200 #l. The cells
were then pulsed for 4 h with [3H]thymidine (New England Nuclear, Boston, MA) and
thymidine incorporation determined. The IL-2 standard used in these studies was a gift
from Dr. Jacques Chiller, Lilly Research Laboratories, La Jolla, CA and contained 250
U/ml. The amount of IL-2 present in an unknown sample was determined by titration
against the IL-2 standard.
Preparation and Binding of lFN-7-coated Covaspheres to Macrophages. 100 #1 of a 1.35%
suspension of coumarin (green) fluorescent microspheres (Covaspheres, Covalent Tech-
nology Corp., Redwood City, CA) in PBS were mixed with 25 ~1 of purified recombinant
IFN-'y (2.6 mg/ml) and incubated overnight at 4°C. The microspheres were pelleted by
centrifugation for I0 min at 8,000 g in a Beckman microfuge 12 (Beckman Instruments,
Inc., Palo Alto, CA) and washed one time with PBS containing 1% BSA. Microspheres
were then incubated with 1 ml of 1 M ethanolamine in PBS, pH 7.0 to block any remaining
unreacted sites on the beads. After three washes, microspheres were sonicated for 1 min
and resuspended in 1 ml of RPMI medium and stored at 4°C until use. Microspheres
treated with ethanolamine but not IFN--y were used as controls. For binding studies, 2 x
105 EPM were adhered for 2 h at 37 °C to coverslips (Coverslip No. 1-THK, Bellco Glass,
Inc., Vineland, N J) in 24-well (16-ram diameter) tissue culture plates (Costar, No. 3424).
Coverslips were washed and 20 ul of microspheres was added in 1 ml of medium. After
incubation for 2 h at 4°C, coverslips were washed and examined under a microscope
(Zeiss, Photomicroscope Type II, Munich, West Germany) with ultraviolet and visible
fight.
Binding Studies with ~25I-recombinant IFN-7. Five million bone marrow-derived mac-
rophages125 or the. macrophage cell line P388D1 were incubated with different concentrauons
of I-IFN-7 m a total volume of 150 #1 with RPMI-supplemented medium. After 2 h at
4 ° C, 120 tsl was applied to 250 tA of an oil mixture consisting of six parts dioctylpthalate
and four parts dibutylpthalate (Aldrich Chemical Company, Inc., Milwaukee, WI) in 400
#1 polyethylene microfuge tubes (VWR Scientific, Inc.) and microfuged at 4°C for 1 min
at 9,000 rpm in a Beckman microfuge 12. Tubes were sectioned and the radioactivity of
the pellet and supernatant counted. Specific binding was defined as the difference between
total binding and the nonspecific binding occurring in the presence of a 500-fold excess
of unlabeled IFN-7. Similar studies were carried out using membranes from bone marrow-
derived macrophages except that membranes were incubated with IFN-7 in the following
buffer: 25 mM Hepes, pH 7/0.1 M NaCl/5 mM sodium phosphate/0.34 units per ml
soybean trypsin inhibitor/0.5% saponin. Free and membrane-associated IFN were sepa-
rated by centrifugation through 20% sucrose.
Monoclonal Antibodies. Monoclonal antibodies to recombinant IFN-7 were produced
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL.
59
by fusion of immune Armenian hamster splenocytes with HAT-sensitive murine myeloma
cell lines, as will be described elsewhere. The resulting hybridomas were cloned three
times by limiting dilution. Monoclonal antibodies were purified from tissue culture
supernatants on a column of staphylococcal protein A Sepharose (Pharmacia Fine Chem-
icals, Piscataway, N J).
Results
Demonstration of MAF Absorption.
For these studies, MAF has been defined
as the lymphokine that primes macrophages for expression of nonspecific tu-
morilytic activity toward P815 mastocytoma cells. As a source of MAF, we used
the supernatant from a Con A-stimulated culture of the murine T cell hybridoma
24/G1. This supernatant contained 55,000 U of MAF activity/ml and 833 IRU
IFN-~,/ml. We have previously shown that all the MAF activity in this supernatant
was attributable to IFN-% By titration against purified recombinant murine IFN-
% the supernatant was estimated to contain 190 ng IFN-~/ml.
Fig. 1 demonstrates that MAF activity was removed from the 24/G1 super-
natant following exposure to elicited peritoneal exudate macrophages (EPM). In
this experiment 1 ml of supernatant was diluted 1/100 (550 U/ml) and was
incubated for 4 h at 37°C either in empty wells or in wells containing 6 x 106
adherent EPM. The supernatant from the control well displayed 460 U MAF/
ml as compared with 530 U/ml of unincubated supernatant, indicating that MAF
activity was stable to the 37 °C incubation. However, after exposure to EPM, the
supernatant contained only 92 U/ml of MAF activity. This represents an 80%
loss of MAF activity.
Assessment of Dose, Time, and Temperature Dependency of Absorption.
Removal
of MAF activity from 24/G1 supernatants was dependent on both the number
of cells and the amount of MAF used in the experiment. Elicited or bone
marrow-derived macrophages were equivalent in their ability to bind MAF (Fig.
2). Following 4 h incubation, 50% of the MAF activity was removed by 1.5 x
40" :
J
c
10 3'0 100 300
Reciprocal of MAF Oilution
FIGURE l. Demonstration of MAF absorption by macrophages, 6 x 106 EPM were incubated
for 4 h at 37°C with 1 ml of 24/G1 supernatant diluted 1/100. As control, the diluted
supernatant was incubated without macrophages. Following incubation, the supernatants were
removed and the remaining MAF activity measured using freshly explanted macrophages.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
60 RECEPTOR FOR INTERFERON-~, ON MACROPHAGES
100-
75-
50-
25-
EIm~tad / ~
//o. -~ Dif~ved
l
,/1
Macr0phages
xlO"
FIGURE 2. Dose-dependent absorption of MAF activity by macrophages. Different numbers
of EPM v~ere incubated for 4 b at 37°C with I ml of 1/100 diluted 24/G1 supernatant.
Controls and quantitation of MAF activity were as in Fig. 1.
100-
75"
g
m.
< 50-
E
25-
37 °
4 o
Time
(hours)
FIGURE 3. Temperature dependence of MAF absorption. Absorption and quantitation were
as in Fig. 1.
106 EPM or 3 x 106 bone marrow-derived macrophages and 12 × 106 of either
cell population removed 100% of the 24/G1 derived MAF. The reduction of
MAF activity was also dependent on the initial input of the lymphokine. At MAF
dilutions of 1/25, 1/50, 1/100, 1/150, and 1/200, the amount of activity
removed following treatment with 6 × 10 6 EPM for 4 h at 37°C was 40%, 57%,
73%, 92%, and 100%, respectively.
Absorption was time dependent at 37°C but not at 4°C or 22°C (Fig. 3). At
all three temperatures, exposure of 24/G1 supernatants to 6 × 10 6 EPM for 30
min, led to removal of ~40% of the MAF activity. Assuming that the supernatants
contained 190 ng IFN-3,/ml, an estimate can be made that EPM bound or
consumed ~2,000 molecules of IFN-T/cell. While longer incubation at 4°C or
22°C did not result in additional removal of MAF activity, incubation at 37°C
effected a time-dependent increase in MAF absorption. After 4 h at 37 °C, 85%
of the MAF activity was lost from 24/G1 supernatants and all the activity was
removed when the absorption was carried out for 24 h. Bone marrow-derived
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL.
61
macrophages were equivalent to EPM in these experiments. Similar results were
also obtained using other 24/G1 supernatants or supernatants of Con A-stimu-
lated normal splenic cells.
Controls to Indicate that Removal of MAF Activity was a Result of Absorption.
The
ability of macrophages to rapidly (30 rain) remove MAF activity from 24/G1
supernatants at 4°C suggested that the loss of activity was the result of IFN-3,
absorption. To further substantiate this hypothesis three additional experiments
were performed.
To rule out the possibilities that (a) the MAF/IFN-? was degraded or altered
by macrophage-derived factors or (b) treated macrophages produced substances
such as prostaglandins that could interfere with the subsequent quantitation of
MAF activity, a mixing experiment was performed as outlined in Table I.
Supernatants containing untreated MAF or MAF preincubated either in empty
wells or with EPM were mixed in various combinations with medium from EPM-
containing wells. In all cases, the amount of MAF activity that was measured in
the mixtures closely approximated the theoretical values calculated from the
mixture composition. No difference in these results were seen when 10 -s M
indomethacin was present during preparation of the various supernatants. Mac-
rophages incubated with MAF for 2 and 4 h absorbed 46% and 75% of the
activity, respectively, in the absence of indomethacin and 42% and 74% of the
activity was removed in the presence of the cyciooxygenase inhibitor.
In a second series of experiments, paraformaldehyde-fixed or heat-killed
(56°C, 30 rain) EPM were compared to normal EPM for the ability to absorb
MAF activity (Table I1). After 2 h of incubation the untreated cells had removed
61% of the MAF activity, while the fixed cells had removed only 38%. At
subsequent time points the native cells continued to absorb MAF activity, while
the fixed cells did not. Cells heated to 56°C for 30 min were nonviable as
TABLE I
Effect of Mixing Macrophage Supernatants on Quantitation of MAF
Activity
Sample Measured Theoretical
U/ml
Untreated MAF (i/100) 650 --
Absorbed MAF* 96
MAF incubated 4 h (control)* 480
M¢~ incubated with medium ~ 0 --
50% untreated + 50% absorbed 385 373
50% untreated + 50% MAF control 550 565
50% untreated + 50% incubated medium 300 325
50% incubated medium + 50% MAF control 225 240
50% absorbed MAF + 50% MAF control 300 288
Macrophage supernatants were mixed and incubated for 4 h at 37°C
and the MAF activity quantitated.
* 6 x 106 EPM were incubated with 1 ml of 1/100 diluted 24/G1
supernatant 4 h at 37°C.
* I ml of diluted supernatant was incubated in an empty well for 4 h at
37°C.
! 6 x 10 ~ EPM were incubated with 1 ml of medium for 4 h at 37°C.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
62 RECEPTOR FOR INTERFERON-~, ON MACROPHAGES
TABLE [I
Influence of Fixation or Heat Killing on Absorption of MAF Activity by Macrophage
Percent absorption
Treatment Time (h) Cytotoxic activity
2 4 24
% U/ml
None 61 80 100 46,000
1% Paraformaldehyde 38 42 39 <100
Heated(56°C, 30 min) 5 0 6 <100
6 x
10 6
adherent EPM were treated with 1% paraformaldehyde at 4°C for 1 h or heated 30 rain at
56°C. Cells were washed and incubated at 37°C with 1 ml of 24/G1 supernatant diluted 1/100.
r,=
100"
50-
e. ~ IL2 4
0 ½ 4
" 2'4
Time of Incubation
FIGURE 4. Absorption of MAF activity but not IL-2 activity by macrophages. 6 x 10 ~ EPM
were incubated with i ml ofa supernatant containing 25 U IL-2 activity/ml and 550 U MAF
activity/ml. Incubations were performed at 37°C for 2, 4, and 24 h.
evidenced by release of the cytoplasmic marker lactate dehydrogenase and their
permeability to trypan blue. These cells did not express tumoricidal activity
when treated with 24/GI supernatants and also absorbed <10% of the MAF
activity even when the incubation period was prolonged to 24 h.
A third set of experiments was performed to assess the specificity of the
absorption. Macrophages were incubated with a supernatant containing 25 U/
ml IL-2 and 550 U/ml MAF for 2, 4, and 24 h at 37°C. After incubation, the
remaining MAF and IL-2 activities were quantitated (Fig. 4). When compared
to the corresponding control, IL-2 showed 17%, 4%, and 13% absorption after
2, 4, and 24 h, respectively, while for the same periods of incubation, 45%, 83%,
and 100% of MAF was absorbed. Similar results were obtained using 50 or 12.5
U/ml of IL-2. The variation between the values obtained for the IL-2 after
incubation with macrophages was similar to the variation obtained with the
control values, suggesting that IL-2 was not absorbed by macrophages.
Relationship between Time of Absorption and Development of Cytotoxicity.
To
study the relationship between absorption and development of the tumoricidal
response, macrophages were incubated at 37 °C with the 24/G1 supernatant for
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL. 63
periods of time between 1 and 6 h. After each incubation period, the supernatant
was removed and analyzed for MAF activity while the macrophages were washed
and then challenged with P815 in the presence of a second signal (HKLM). Fig.
5 shows that although MAF is absorbed progressively over the period of incu-
bation, a minimum exposure of 4 h is required before the macrophage can
express tumorilytic activity. During this critical period of incubation, 82% of the
MAF has been absorbed. No change was observed in the minimum incubation
time needed to elicit a tumoricidal response when higher concentrations of MAF
were used.
Cell and Species Specificity of Absorption.
Elicited and resident peritoneal exu-
date murine macrophages are known to differ in their responsiveness to IFN-%
In our experimental assay system, only the freshly explanted elicited cell popu-
lation developed IFN-dependent nonspecific tumorilytic activity (Table
III).
This
responsiveness was rapidly lost in culture and was also not expressed by freshly
explanted or cultured resident peritoneal exudate cells. Table III shows that the
defect is not related to IFN-'y binding by the cells, since both the responsive and
100.
i 75-
- 50-
~ q
~ 25-
I
[ I
/
/I
..... "9, ...... ~
2 3 4 ;
Incubation Time (Hours)
5O
,7
-25 _~
c
o=
0
FIGURE 5. Correlation of MAF absorption and development of cytolytic activity. 6 x 106
EPM were incubated for the time indicated with 1/100 diluted 24/G1 supernatant. The
resulting absorbed supernatants were assayed for remaining MAF activity. The macrophages
used for absorption were mixed with 6 x 105 ~ltIn-labeled P815 tumor cells, 6 X l0 T HKLM
(as a second signal) and 10 -6 M indomethacin in a total volume of 1 ml. After 18 h in culture,
0.5 ml of supernatant was removed, the radioactivity counted and the cytolytic activity
calculated (O""O).
TABLE III
Cytotoxic Activity and MAF Absorption by Elicited and Resident
Peritoneal Macrophages
Percent
Macrophages Cytotoxic
(6 X 10 6) Condition activity absorptiOn*after 4 h
(MAF U/ml) incubation
Elicited Freshly explanted 48,000 78
3-d culture <200 61
Resident Freshly explanted <200 60
3-d culture <200 65
* 1 ml of 24/G1 supernatant diluted 1/100 was used for each absorption
at 37°C.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
64 RECEPTOR FOR INTERFERON-3, ON MACROPHAGES
unresponsive cells absorbed comparable amounts of MAF activity when incu-
bated for 4 h at 37°C (78% and 60-65%, respectively). In other experiments
(data not shown) macrophages from eight different inbred and two outbred
strains of mice were tested for their ability to express IFN-3,-dependent tumori-
cidal activity and to absorb MAF activity from the 24/G1 supernatant. The
inbred mouse strains tested included C3HeB/FeJ (H2k), C3H/HeJ (H2k), CBA/
CaJ (H2k),
a/J (H2a), C57BL/6J (H2b), DBA/2J (H2a), SJL (H2S), and C3D~ F,/J
(H2 k X H2Cl).
Although the various macrophages expressed different levels of
tumorilytic activity, they all absorbed nearly identical amounts of MAF activity
during incubation periods of 2, 4, or 24 h. Thus absorption was not H-2
restricted.
Table IV compares a variety of murine cell lines for their ability to absorb 24/
Gl-derived MAF activity. Murine fibroblasts
(L929),
fibrosarcomas (TU5 and
1023), mastocytomas (P815), and T cell hybridomas (TH4.4, 24/G1 and 1.19)
all absorbed MAF activity at 37 °C in a time-dependent fashion. Two macrophage
cell lines (P388D~ and J774) were identified that were deficient in their ability
to bind natural IFN-3,. Even after 24 h incubation, P388D~ absorbed only 37%
of the MAF activity and J774 only 60%, while all other cell lines and normal
peritoneal exudate macrophages had absorbed 93-100% of the activity. In
addition to their reduced ability to bind IFN-3', P388D~ and J774 also displayed
a quantitative defect in their ability to mount an IFN-dependent nonspecific
tumorilytic response. When 24/G1 supernatants were titrated on control
C3HeB/FeJ macrophages, a value of 43,800 U of MAF activity was obtained.
However, values of only 320 U and 4,300 U were obtained when the assay was
performed with P388Dl or J774, respectively. When examined after 6 h incu-
bation with the hybridoma supernatant, the two cell lines also showed a defective
TABLE IV
Absorption of MAF Activity by Murine Cell Lines
Cell type Time of incubation (h)
2 4 24
EPM control 47 88 100
L929" 22 33 93
TU5* 33 60 100
1023' 35 72 100
P8150 61 93 100
TH4.4 u 37 68 95
24/G1 j 33 97 100
1.191 52 97 100
P388DI ~ 0 10 37
J774 ~ 12 25 60
In all the cases 6 x 106 cells were incubated
ml of 24/G1 supernatant diluted 1/100.
* Fibroblast.
* Fibrosarcoma.
a Mastocytoma.
I T cell hybridoma.
Macrophage cell line.
for 2, 4, and 24 h with 1
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL. 65
spreading reaction. Although 82% of the control cells had spread, only 15% of
P388DI and 22% ofJ774 had undergone morphological changes.
Table V compares cells from a number of different species for their ability to
absorb 24/Gl-derived MAF activity. Absorption appears to display a degree of
species specificity. When peritoneal exudate macrophages from four rodent
species were compared, only murine cells absorbed MAF activity during the first
4 h of incubation. After 24 h of incubation, rat macrophages were also found to
absorb or consume a small amount (37%) of MAF activity. Human polymorpho-
nuclear leukocytes, lymphocytes, and erythrocytes did not absorb murine MAF.
However, human monocytes absorbed 33% of the activity following in vitro
culture for 24 h. This result agrees with the observation that after 1-3 d in
culture, human peripheral blood monocytes acquire the ability to respond to
murine IFN-3, and can express nonspecific tumoricidal activity toward the human
melanoma cell line A357 (Buchmeier, N. A., and R. D. Schreiber, unpublished
observation). In contrast to normal human peripheral blood cells, the human
histiocytic lymphoma cell line Ugs7 was quantitatively equivalent to murine
macrophages in effecting absorption of 24/Gl-derived MAF activity.
Absorption of Recombinant IFN-y by Macrophages.
We have previously found
that the MAF activity displayed by purified recombinant murine IFN-3, was
equivalent, on an antiviral activity basis, to the MAF activity expressed by the T
cell hybridoma-derived IFN-3, (27). This observation prompted a comparison of
the ability of macrophages to absorb natural arid recombinant IFN-3,. For these
experiments, the recombinant 1FN-3, was diluted such that the final solution
contained the same amount of MAF activity as a 1/100 dilution of the 24/G1
supernatant (550 U/ml). After 2 and 4 h incubation with normal EPM the
amount of absorption of the recombinant IFN-3, (32% and 76%, respectively)
was similar to that of the natural IFN (44% and 86%, respectively, Table VI).
Moreover, P388D~ that was deficient in absorbing the natural IFN-3, was also
deficient in absorbing the recombinant material.
In order to directly visualize the interaction of IFN-3, with the macrophage
TABLE V
Species Specificity of MAF Absorption
Time of incubation (h)
2 4 24
Macrophages
Mouse 47 88 100
Rat 0 0 37
Hamster 0 0 0
Guinea Pig ND 0 0
Human cells
Monocytes 0 0 33
Polymorphonuclear leukocytes 0 0 0
Lymphocytes 0 0 8
Erythrocytes 0 0 0
Ugs7 47 94 1 O0
Absorption conditions as in Table IV.
ND,
not done.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
66 RECEPTOR FOR INTERFERON-~. ON MACROPHAGES
TABLE VI
Absorption of Recombinant 1FN-~, by EPM and P388DI
Percent absorption
Recombinant
IFN-~, 24/Gl-IFN-3,
2b 4h 2h 4h
%
Control macrophages 32 76 44 86
P388Dj 0 3 0 8
6 x 10 ~ cells were incubated for 2 or 4 h in the presence of 1 ml of 24/
G1 supernatant or recombinant 1FN-7 diluted such that both preparations
displayed 550 U of MAF activity per ml.
surface, recombinant IFN-3, was bound to fluorescent microspheres and incu-
bated at 4°C with macrophage populations. EPM exposed to microspheres
bearing IFN-7 bound one to nine beads per cell (Fig. 6a). Staining of the cell
population was heterogeneous with 35% of the cells binding less than three
beads. When microspheres were used that did not carry IFN-% but that had
been blocked by incubation with ethanolamine, no binding occurred (Fig. 6b).
Preincubation of EPM with high concentrations of 24/G 1 supernatant, abrogated
binding of IFN-7-covaspheres, indicating that binding was IFN mediated (Fig.
6c). Fig. 6d shows that P388D~ did not bind IFN-7-coated microspheres. This
result thus agrees with the absorption data.
Demonstration of a Specific IFN-7 Receptor on Macrophages: Quantitation of the
Interaction between Recombinant IFN-7 and Macrophages.
The data presented
thus far indicate that natural and recombinant IFN-3, bound to macrophages in
a similar fashion. Because of its availability, the purified recombinant material
was used to quantitate the interaction of IFN-~, with the macrophage surface.
Recombinant IFN-7 was first subjected to HPLC gel filtration to remove any
degraded material that may have formed during storage. As detected by ultra-
violet absorption at 280 nM, two molecular weight species eluted from the
column with symmetrical elution profiles (data not shown). The first peak
displayed an apparent molecular weight of 32,000, while the second peak
corresponded to an Mr of 5,000-7,000. Only the material in the 32,000-dalton
peak expressed MAF and antiviral activities and reacted with monoclonal anti-
bodies to IFN-%
Fig. 7 shows the results of a binding experiment that used bone marrow-
derived macrophages and 125I-labeled HPLC-purified recombinant IFN-7. Spe-
cific binding has been defined as the binding that is inhibited in the presence of
a 500-fold excess of unlabeled IFN-7. In these experiments -80% of the total
binding was specific. By Scatchard plot analysis, the cells were found to carry,
per cell, 15,200 receptors that bound ligand in a homogeneous, noncooperative
manner and displayed an affinity (Ka) of 1.18 X 10 s M ~1. The average of three
such experiments produced mean values of 12,000 receptors per cell and a Ka
of 0.9 x l0 s M -1. Analysis of IFN-3, binding to P388Da indicated that the cell
line carried only 760 receptors per cell, which is 6.3% that displayed by bone
marrow-derived macrophages. Similar experiments were performed on mem-
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
C3
t-
>
,q
>
FIGURE 6. Binding of fluorescent microspheres coated with recombinant
1FN-~' to macrophages. 2 × l0 s EPM adherent on coverslips were incubated
with microspheres for 2 h at 4 °C. (a) EPM with microspheres bearing IFN-
7. (b) Microspheres coated with ethanolamine. (c) EPM were first incubated
at 4°C with a 1/100 dilution of 24/G1 supernatant and then with IFN-3,
bearing microspheres for 2 h at 4°C. (d) P388D~ incubated with IFN-7
microspheres.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
68
RECEPTOR FOR INTERFERON-')" ON MACROPHAGES
2-
_~ °
°° /: °2t
~
1-
~. ,It 0l , ,
~"
r.~ I¢ 0 50 100
B (fmoleS]
0 100 260 3[)0
IFN Offered (ng)
FIGURE 7. Quantitation of IFN-3, binding to murine macrophages. 5
X
l0 e bone marrow-
derived macrophages were incubated at 4°C for 2 h with different amounts of radiolabeled
HPLC purified, recombinant IFN-~'. Cell associated and free ~I-IFN were separated by
centrifugation over pthalate oil as outlined in Materials and Methods. Specific binding was
defined as the difference between total binding and the nonspecific binding occurring in the
presence of a 500-fold excess of unlabeled IFN-3'. The insert represents Scatchard plot analysis
of the data. B, bound IFN-3, and F, free IFN-3,.
4.0
2.0
R
m 8000
IRB
2o mH
IRU
416
LRU
BI
CONTROL HYBRIDOMA-DERIVED IFNy MEDIUM lENa
Competitor
8000
IRU
IFN~
FIGURE 8. Specificity experiments for recombinant ~2sI-IFN-7 binding to macrophage. Bone
marrow-derived macrophages were preincubated for 30 min at 4°C with designated amounts
of competitor in a total reaction volume of 100 pl. 20 #1 of l~Sl-recombinant IFN-~ (290 IRU)
was then added to each tube and incubation continued for 2 h at 4°C. Specific binding and
separation of bound and free ~251-1FN-3, as in Fig. 7. For these studies the T cell hybridoma
supernatant or a sham supernatant were concentrated 10-fold before use by ultrafihration.
brane preparations of bone marrow-derived macrophages. These studies showed
that 17 #g of membranes (weight recovered from 107 cells) specifically bound a
maximum of 5 ng of IFN-3'.
Fig. 8 demonstrates the specificity of the binding reaction. Preincubation of
the bone marrow macrophages with different amounts of 24/Gl-derived IFN-y
inhibited the uptake of 125I-recombinant IFN-3' in a dose-dependent fashion.
Addition of 208 IRU of natural IFN-y to the reaction mixture caused a 62%
reduction in binding of a comparable amount of the recombinant material (290
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL.
69
IRU), indicating that the two products bound to the cells with comparable
affinity. Crude preparations of IFN-/3 displayed a weak ability to inhibit binding
of recombinant IFN-3, when added in large excess (8,000 IRU), while IFN-a
displayed no inhibitory activity at all.
Relationship between Receptor Binding and Macrophage Activation.
Two mono-
cional antibodies to IFN-~, (H2 and H21) that differentially modulate IFN-3,-
dependent biological responses in macrophages were used to demonstrate the
involvement of the IFN-'y receptor in macrophage activation. H21 has been
found to inhibit the ability of IFN-3' to induce nonspecific tumorilytic activity or
Ia expression in macrophages, while H2 was found to enhance these activities. 2
Fig. 9 shows that inhibition or enhancement of IFN-induced macrophage func-
tions correlates with the interaction of IFN-~ with the IFN-'), receptor. Depending
on the input of IFN-"y, an excess of H21 inhibited 100% of the cellular uptake
of IFN-3,, while H2 enhanced binding as much as 3.3-fold. No alteration of
binding was observed when normal hamster IgG was substituted for the mono-
clonals.
Discussion
This report documents the existence of a IFN-~r-receptor on murine macro-
phages and indicates that receptor engagement is a necessary first step in the
induction of nonspecific tumorilytic activity in these cells. Macrophages bound
either purified recombinant IFN-~ produced in bacteria or natural IFN-'y derived
from either a murine T cell hybridoma or normal murine splenic cells. Binding
was specific, saturable, of high affinity, and resulted in the induction of a
biological response in appropriate macrophage populations. These parameters
thus indicate that IFN-~ was binding to a specific cell surface receptor.
Although substantial quantities of purified natural murine IFN-'y were not
available for these studies, the interaction could be demonstrated by quantitating
the macrophage-dependent absorption of I FN-~ from supernatants of stimulated
T lymphocytes. We chose to quantitate MAF activity as an indicator of IFN-'y,
6-
~3-
z
H2
• = T
50 100 150
tFN Offered (rig)
FIGURE 9. Effect of monoclonal antibodies on IFN-3, uptake by macrophages. 5 X 106 bone
marrow-derived macrophages were incubated for 2 h at 4°C with the indicated amounts of
12sI-IFN-3, in the presence of medium (0), normal hamster IgG (O), monoclonal anti-IFN, H2
(A) or monoclonal anti-IFN, H21 (&). Specific binding and separation of cell-associated '2sI-
IFN-3, as in Fig. 7.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
70 RECEPTOR FOR INTERFERON-q, ON MACROPHAGES
since the macrophage was the target cell for this activity and since the MAF assay
was more quantitative and 10 times more sensitive than the antiviral activity
assay. For most of the studies, a supernatant of the 24/G1 T cell hybridoma was
used. Previous studies have shown that the MAF activity produced by 24/G1
was largely due to IFN-3, (6, 8, 27). Recently, using monoclonal antibodies to
recombinant IFN-'y we have found that IFN-"r represented the only MAF in
these supernatants and in Con A-stimulated supernatants of normal murine
splenic cells, z The absorption experiments showed that the ability of macrophages
to remove MAF activity from culture supernatants was dependent on cell
number, temperature, time of incubation, and displayed species specificity. The
data also indicated that the activity was removed by cellular absorption and not
by extracellular degradation or production of macrophage-derived factors that
interfered with MAF quantitation. This conclusion was supported by the obser-
vations that (a) absorption occurred at 4 o C, (b) paraformaldehyde-fixed murine
macrophages absorbed MAF activity, (c) absorption of 24/G1 MAF was effected
by murine macrophages but not hamster- or guinea pig-derived cells, (d) absorp-
tion was specific since treated supernatants showed a selective reduction in MAF
activity but not I L-2 activity, and (e) mixing 24/G 1 supernatants with macrophage
culture supernatants did not affect the levels of MAF activity.
While these results strongly suggested the existence of an IFN-3, receptor on
murine macrophages, they had to be validated with quantitative uptake experi-
ments using radiolabeled IFN-y. This was accomplished with ~2~I-labeled purified
recombinant IFN-3'. Scatchard plot analysis of the binding data showed that the
binding of IFN-7 to macrophages at 4°C was noncooperative (linear Scatchard
plots), saturable (~12,000 receptors/cell), specific (binding was inhibitable by
unlabeled IFN-3'), and of moderately high affinity
(K a = 0.9 X 10 8 M-I),
Binding
could be demonstrated to membrane preparations as well as to intact cells.
Several lines of evidence indicated that the interaction of the receptor with
natural IFN-3, was comparable to that observed with the recombinant material.
Both types of IFN-y expressed equal MAF activities when compared on an
antiviral activity basis (8, 27). Hybridoma-derived and recombinant IFN-3' were
absorbed in an identical fashion by normal macrophages and were not absorbed
by the P388D1 cell line. Subsequent quantitation of ~25I-recombinant IFN-3,
binding to this cell line indicated that P388D1 carried only 6.3% as many
receptors as were expressed on normal macrophages. The number of receptors
on normal macrophages determined by the absorption experiments with natural
IFN-3, (2,000-5,000 receptors/cell) was comparable to the number obtained by
measurement of l~5I-recombinant IFN-y binding (12,000 receptors/cell). Finally,
unlabeled, natural IFN-3, could quantitatively compete with the labeled recom-
binant material for receptor binding. This result indicated that the two IFN-3'
preparations bound to the receptor with similar affinities.
The monoclonal antibody studies indicate that binding of IFN-y to its cell
surface receptor is a necessary step for the induction of a biological response in
macrophages. Modulation of the binding reaction correlated with appropriate
increases or decreases in the eventual induction of tumoricidal activity. It is most
likely that the inhibition produced by H21 reflected its ability to directly or
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL. 71
sterically block regions of IFN-~, that interact with the receptor. The enhance-
ment observed with H2 probably reflects the formation of soluble immune
complexes that were multimeric with respect to IFN-~, and that formed multi-
point attachments to the macrophages through IFN-~, receptors. These possibil-
ities are currently under examination.
The results presented in this communication are consistent with observations
made in other laboratories. The ability of macrophages to remove MAF activity
from lymphokine preparations was previously reported by Yamamoto and To-
kunaga (28). However, this earlier study used a MAF of unknown biochemical
identity and failed to establish absorption as the mechanism for MAF removal.
A number of other studies have indicated that MAF-dependent induction of
tumoricidal activity is not H2 restricted (29). Our experiments support this
conclusion, since macrophages from mice with different H2 backgrounds ab-
sorbed MAF activity to comparable degrees.
The species specificity of the binding agrees with other results obtained
independently. Although murine IFN-y did not bind to hamster or guinea pig
macrophages, it did bind to cultured normal human monocytes and to the human
histiocytic lymphoma cell line U937. Both natural and recombinant murine IFN-
y have been found to activate normal human peripheral blood monocyte-derived
macrophages for nonspecific tumoricidal activity (Buchmeier, N. A., and R. D.
Schreiber, unpublished data). Moreover, unlabeled recombinant
murine
IFN-3,
can inhibit binding of
~25I-human
recombinant IFN~" to human mononuclear
phagocytes (Celada, A., and R. D. Schreiber, manuscript in preparation). These
results thus clearly differentiate IFN-~,-dependent MAF activity from IFN-y-
antiviral activity on fibroblasts. The latter displays strict species specificity.
To date only one other series of studies exists that suggests the existence of a
cellular IFN-y receptor. Anderson et al (30, 31) have demonstrated binding of
purified natural human IFN-y to human GM-258 fibroblasts. These fibroblasts
expressed 8,000-20,000 receptors/cell and bound ligand with a Kd of 2-6 ×
10 -9 M (or a K~ of 1.7-5 X l0 s M-l). These parameters are similar to the values
obtained in our study. We have also documented the presence of an IFN-'y
receptor on the murine fibroblast cell line L9~9 as well as several other cell types
using the MAF absorption technique. However, it is not yet known whether the
receptors on the different cell types are identical or distinct. More structural and
immunochemical data about the IFN-'r receptor is needed before this question
can be answered. Work is currently underway to isolate and characterize the
macrophage ]FN-'y receptor.
The identification of a IFN-y receptor on macrophages that participates in the
induction of nonspecific tumoricidal activity is the first step at understanding
macrophage activation at the molecular level. Work is currently in progress to
determine the fate of the receptor-bound ligand and whether receptor engage-
ment is sufficient to initiate a tumoricidal response. The quantitation of this
receptor-ligand interaction should also provide a means of defining the mecha-
nisms of action of other factors that have been reported to have MAF activity
but not antiviral activity (32-34).
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
72 RECEPTOR FOR INTERFERON-~' ON MACROPHAGES
Summary
Gamma-interferon (IFN-~) is the macrophage-activating factor (MAF) pro-
duced by normal murine splenic cells and the murine T cell hybridoma 24/G 1
that induces nonspecific tumoricidal activity in macrophages. Incubation of 24/
G1 supernatants diluted to 8.3 IRU IFN--y/ml with 6 x 106 elicited peritoneal
macrophages or bone marrow-derived macrophages for 4 h at 37°C, resulted
in removal of 80% of the MAF activity from the lymphokine preparation. Loss
of activity appeared to result from absorption and not consumption because (a)
40% of the activity was removed after exposure to macrophage for 30 min at
4 ° C, (b) no reduction of MAF activity was detected when the 24/G 1 supernatant
was incubated with macrophage culture supernatants, and (c) macrophage-treated
supernatants showed a selective loss of MAF activity but not interleukin 2 (IL,2)
activity. Absorption was dependent on the input of either IFN-~, or macrophages
and was time dependent at 37 °C but not at 4 o C. With four rodent species tested,
absorption of murine IFN-3, displayed species specificity. However, cultured
human peripheral blood monocytes and the human histiocytic lymphoma cell
line U937 were able to absorb the murine lymphokine. Although the majority of
murine cell lines tested absorbed 24/G1 MAF activity, two murine macrophage
cell lines, P388D~ and J774, were identified which absorbed significantly reduced
amounts of natural IFN-7. Purified murine recombinant IFN-3, was absorbed by
elicited macrophages but not by P388D~. Normal macrophages but not P388D~
bound fluoresceinated microspheres coated with recombinant IFN-y and binding
was inhibited by pretreatment of the normal cells with 24/G1 supernatants.
Scatchard plot analysis showed that 12,000 molecules of soluble 125I-recombinant
IFN-7 bound per bone marrow macrophage with a Ka of 0.9 x 10 s M -l. Binding
was quantitatively inhibitable by natural IFN-3 ~ but not by murine IFN~. IFN-/3
competed only weakly. Monoclonal antibodies against IFN-2¢ either inhibited or
enhanced MAF activity by blocking or increasing IFN-7 binding to macrophages,
respectively. These results indicate that IFN-7 reacts with a receptor on macro-
phage in a specific and saturable manner and this interaction initiates macrophage
activation.
The authors are grateful for the skilled and dedicated technical assistance of LoriJ. Hicks
and Virginia M. Keivens. We also wish to thank Dr. Jacques Chiller of the Eli Lilly
Research Laboratories, La Jolla, California for his help with the IL-2 measurements and
for his constructive criticism.
Received for publication 19 March 1984.
References
1. North, R.J. 1978. The concept of activated macrophage. J. Immunol. 121:806.
2. Karnovsky, M. L., and J. K. Lazdins. 1978. Biochemical criteria for activated
macrophages. J. lmmunol. 121:809.
3. Cohn, Z. A. 1978. The activation of mononuclear phagocytes: fact, fancy and future.
J. Immunol. 121:813.
4. Rocklin, R. E., K. Bendtzen, and D. Greineder. 1980. Mediators of immunity:
lymphokines and monokines. Adv. lmmunol. 29:56.
5. Schreiber, R. D., A. Altman, and D. H. Katz. 1982. Identification of a T cell
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
CELADA ET AL. 73
hybridoma which produces large quantities of macrophage-activating factor.
J. Exp.
Med.
156:677.
6. Schreiber, R. D., J. L. Pace, S. W. Russell, A. Altman, and D. H. Katz. 1983.
Macrophage activating factor produced by a T cell hybridoma: physiochemical and
biosynthetic resemblance to ),-interferon.
J. Immunol.
131:826.
7. Pace, J. L., S. W. Russell, B. A. Torres, H. M. Johnson, and P. W. Gray. 1983.
Recombinant mouse ), interferon induces the priming step in macrophage activation
for tumor cell killing.
J. Immunol.
130:2011.
8. Pace, J. L., S. W. Russell, R. D. Schreiber, A. AItman, and D. H. Katz. 1983.
Macrophage activation: priming activity from a T-cell hybridoma is attributable to
interferon-),.
Proc. Natl. Acad. Sci. USA.
80:3782.
9. Schultz, R. M., and W. J. Kleinschmidt. 1983. Functional identity between murine
),-interferon and macrophage activating factor.
Nature (Lond.).
305:239.
10. Roberts, W. K., and A. Vasil. 1982. Evidence for the identity of murine gamma
interferon and macrophage activating factor.
J. Interferon Res.
4:519.
11. Svedersky, L. P., C. V. Benton, W. H. Berger, E. Rinderknecht, R. N. Harkins, and
M. A. Palladino. 1984. Biological and antigenic similarities of murine interferon-),
and macrophage-activating factor.
J. Exp. Med.
159:812.
12. Nathan, C. F., H. W. Murray, M. E. Wiebe, and B. Y. Rubin. 1983. Identification of
interferon-), as the lymphokine that activates human macrophage oxidative metabo-
lism and antimicrobial activity.
J. Exp. Med.
158:670.
13. Nacy, C. A., S. L. James, w. R. Benjamin, J. J. Farrar, W. T. Hockmeyer, and M. S.
Meltzer. 1983. Activation of macrophages for microbicidal and tumoricidal effector
functions by soluble factors from EL-4, a continuous T cell line.
Infect. Immun.
40:820.
14. Steeg, P. S., R. N. Moore, H. M.Johnson, andJ.J. Oppenheim. 1982. Regulation of
murine macrophage Ia antigen expression by a lymphokine with immune interferon
activity.
J. Exp. Med.
156:1780.
15. Basham, T. Y., and T. C. Mer!gan. 1983. Recombinant interferon-), increases HLA-
DR synthesis and expression.
J. Immunol.
130:1492.
16. King, D. P., and P. P. Jones. 1983. Induction of Ia and H2 antigens on a macrophage
cell line by murine interferon.
J. Immunol.
131:315.
17. Gray, P. W., and D. V. Goeddel. 1983. Cloning and expression of murine immune
interferon cDNA.
Proc. Natl. Acad. Sci. USA.
80:5842.
18. Bolton, A. E., and W. M. Hunter. 1973. The labelling of proteins to high specific
radioactivity by conjugation to a ~I containing a chelating agent. Application to the
radioimmunoassay.
Biochem. J.
133:529.
19. Fishelson, Z., M. K. Pangburn, and H. J. Mfiiler-Eberhard. 1984. Characterization
of the initial C3 convertase of the alternative pathway of human complement. J.
Immunol.
132:1430.
20. Meerpohl, H. G., M. L. Lohmann-Matthes, and H. Fisher. 1976. Studies on the
activation of mouse bone marrow-derived macrophages by the macrophage cytotox-
icity factor (MCF).
Eur. J. Immunol.
6:213,
21. Aguado, M. T., N. Pujol, E. Rubiol, M. Tura, and A. Celada. 1980. Separation of
granulocytes from peripheral blood in a single step using discontinuous density
gradients of Ficoil-Urografin. A comparative study with separation by dextran. J.
Immunol. Methods.
32:41.
22. Fischer, D. G., W.J. Hubbard, and H. S. Koren. 1981. Tumor cell killing by freshly
isolated peripheral blood monocytes.
Cell. Immunol.
58:426.
23. Kaplow, L. S. 1965. Simplified myeloperoxidase staining using benzidine dihydro-
chloride.
Blood.
26:215.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
74 RECEPTOR FOR INTERFERON-',/ ON MACROPHAGES
24. Gabel, C. A., D. E. Goldberg, and S. Kornfeld. 1983. Identification and characteri-
zation of cells deficient in the mannose 6-phosphate receptor: evidence for an
alternative pathway for lysosomal enzyme targeting.
Proc. Natl. Acad. Sci. USA.
80:775.
25. Wiltrout, R. H., D. Taramelli, and H. T. Holden. 1981. Measurement ofmacrophage-
mediated cytotoxicity against adherent and nonadherent target cells by release of
~Indium-oxine.
J. Immunol. Methods.
43:319.
26. Smith, K. A. 1980. T-cell growth factor,
lmmunol. Rev.
51:337.
27. Schreiber, R. D. 1984. Identification of gamma interferon as a murine macrophage
activating factor for tumor cytotoxicity.
Contemp. Top. lmmunobiol.
13:171.
28. Yamamoto, S., and T. Tokunaga. 1981. o-mannose as a component of the macro-
phage surface receptor for macrophage-activating (MAF) in mice.
Cell Immunol.
61:319.
29. Fidler, I.J. 1975. Activation
in vitro
of mouse macrophages by syngeneic, allogeneic
or xenogeneic lymphocyte supernatants.
J. Natl. Cancer Inst.
55:1159.
30. Anderson, P., Y. K. Yip, and J. Vilc~k. 1980. Specific binding of l~SI-human inter-
feron-3" to high affinity receptors on human fibroblasts.
J. Biol. Chem.
257:11301.
31. Anderson, P., Y. K. Yip, andJ. Vilc~k. 1983. Human interferon-3, is internalized and
degraded by cultured fibroblast.J.
Biol. Chem.
258:6497.
32. Meltzer, M. S., W. R. Benjamin, and J. J. Farrar. 1982. Macrophage activation for
tumor cytotoxicity: induction of macrophage tumoricidal activity by lymphokines
from EL-4, a continuous T cell line.J.
Immunol.
129:2802.
33. Kniep, E. M., W. Domzig, M. L. Lohmann-Matthes, and B. Kickh6fen. 1981. Partial
purification and chemical characterization of macrophage cytotoxicity factor (MCF,
MAF) and its separation from migration inhibitory factor (MIF).J.
Immunol.
127:417.
34. Erickson, K. L., L. Cicurel, E. Gruys, and I.J. Fidler. 1982. Murine T-cell hybridomas
that produce lymphokine with macrophage activating factor activity as a constitutive
product.
Cell. Immunol.
72:195.
on July 12, 2011jem.rupress.orgDownloaded from
Published July 1, 1984
