Aug 4, 1988 - andMembrane-Associated Tumor Necrosis Factor in. Murine Macrophages. STEPHEN W. CHENSUE, MD, PhD,. DANIEL G. REMICK, MD,.
American Journal of Pathology, Vol. 133, No. 3, December 1988 Copyright © American Association of Pathologists
Imnmunoh istochem ical Demonstration of Cytoplasm ic and Membrane-Associated Tumor Necrosis Factor in Murine Macrophages STEPHEN W. CHENSUE, MD, PhD, DANIEL G. REMICK, MD, CHRISTINE SHMYR-FORSCH, BS, THEODORE F. BEALS, MD, and STEVEN L. KUNKEL, PhD
From the Departments of Pathology, Veterans Administration Medical Center and the University ofMichigan Hospitals, Ann Arbor, Michigan
Using a highly specific rabbit antisera directed against murine tumor necrosis factor (TNF), immunohistochemical localization ofthis monokine was performed in cultured mouse peritoneal macrophages. Resident macrophages did not express TNF even after stimulus with lipopolysaccharide (LPS). In contrast, 12% of macrophages elicited with Freund's adjuvant stained positively and up to 60% were positive after LPS stimulation. Analysis of the kinetics of expression revealed that maximal staining occurred from 1-3 hours after stimulus with disappearance of staining by 12 hours. Both a membrane and cytoplasmic pattern of staining
could be demonstrated. The presence of plasma membrane TNF was confirmed by scanning electron microscopy. Northern blot analysis and bioassay revealed that the kinetics of TNF mRNA synthesis corresponded to the appearance of the protein while its disappearance corresponded to the appearance of TNF in the supernate. Thus, TNF synthesis and secretion could be histochemically demonstrated. These findings support the notion that TNF production is a characteristic of activated macrophages and that such cells display membrane-associated TNF at least transiently after stimulation. (AmJ Pathol 1988, 133:564-572)
IT IS WELL ESTABLISHED that macrophages produce a broad spectrum of monokines with varied effects on target tissues. One group of these mediators consists of polypeptides and includes interleukins, interferons, colony stimulating factors, complement factors, and cytolytic factors."2 These mediators have been the subject of intense investigation because it is felt they play roles in the induction and maintenance of immune/inflammatory responses; however, little is known with regard to the relative importance ofthese mediators or how their production is orchestrated during host responses. The present study examined the expression of tumor necrosis factor (TNF) in murine macrophages by immunohistochemical localization. This 17 kd secreted molecule is thought to be involved in cachexia, tumor resistance, and inflammation,3-6 but there is little direct evidence describing its immunophysiologic functions in vivo or in situ. Using a highly specific antiserum directed against murine TNF a, we demon-
strated both cytoplasmic- and membrane-associated TNF in activated macrophages elicited with Freund's complete adjuvant. These findings supported previous observations of membrane and cell-associated TNF.7-9 The appearance and disappearance of immunoreactive material corresponded to specific TNF mRNA synthesis and degradation. The disappearance of cell-associated material also corresponded to the appearance of bioactive TNF in the culture supernate. Thus, the synthesis and secretion of TNF could be histochemically observed. These studies provide the Supported by the Veterans Administration and NIH Grants HL31237, HL31963, and HL35276. Dr. Kunkel is an Established Investigator of the American Heart Association. Accepted for publication August 4, 1988. Address reprint requests to Dr. Stephen Chensue, Department of Pathology, 1335 East Catherine, Medical Science I Building, University of Michigan Medical School, Ann Arbor, MI 48105-602.
564
IMMUNOLOCALIZATION OF TUMOR NECROSIS FACIrOR
Vol. 133 * No. 3
Table 1 -Specificity of Rabbit Anti-Mouse TNF Serum as Determined by Competitive Inhibition Using Recombinant Monokines Percent positively staining cells*
Control
rTNF
rIL-1a
rlL-1,f
65±1
3±4
65±4
64±3
Freund's adjuvant elicited macrophage monolayers were stained after 3 hours of LPS stimulation. Anti-TNF antiserum (1:1000) was added in the presence of 4 ,ug/mI of the indicated monokines.
foundation for localization of TNF in situ by examination of tissue sections of various inflammatory and neoplastic processes.
Materials and Methods Animals Female, CBA/J (Jackson Laboratories, Bar Harbor, ME) mice were maintained under specific pathogenfree conditions and provided with food and water ad
libitum. Macrophage Isolation and Culture
Peritoneal exudate cells were obtained by aseptic peritoneal lavage at designated intervals after intraperitoneal injection of 0.2 ml of complete Freund's adjuvant (CFA) (Sigma Chemical Co., St. Louis, MO). Resident peritoneal cells were obtained from uninjected, virgin mice. The cells were washed by centrifugation then suspended in RPMI (GIBCO, Grand Island, NY) containing 10% fetal bovine serum (FBS) (Hazelton, Lenexa, KS), 2 mM glutamine, and 100 U penicillin/ 100 utg/ml streptomycin. Cell concentrations were adjusted to achieve optimal cell densities for monokine production, usually 0.5 to 1 X 106/ml. Macrophages were isolated by 1 hour adherence to 35mm culture dishes (Coming Glass Works, Coming, NY) or 8-well slide tissue culture chambers (Lab Tek, Miles Laboratories, Inc., Naperville, IL) at 37 C in a 5% C02, humidified atmosphere. The nonadherent cells were removed by two vigorous washings with 1 ml of warm RPMI. The resident monolayers were >95% and elicited monolayers were >90% macrophages based on nonspecific esterase staining and morphologic criteria. The macrophage monolayers were then overlaid with RPMI-FBS containing graded concentrations of LPS (Escherichia coli 0111 :B4) (Sigma); controls contained no LPS. Supernates were collected at designated intervals after LPS stimulation, centrifuged at 5OOg for 10 minutes to remove particulates, then frozen at -20 C before monokine
565
assay. In some experiments macrophages were isolated on 100-mm tissue culture dishes for RNA isola-
tion. Antisera and Immunolocalization Anti-TNF alpha was produced by immunization of rabbiLs with recombinant murine TNF purified by the method of Beutler et al.'0 The purified antigen consisted of a dominant 17-18 kd band and a faint 38 kd band. The latter was likely a doublet. The antigen (25 ,ug) was administered in multiple intradermal sites with Freund's complete adjuvant followed by a equivalent boost 2 weeks later. The resulting antiserum reacted with the 17-18 kd recombinant TNF in Westem blot analysis. One milliliter of a 1:1000 dilution completely neutralized 2500 units of bioactive material. The antiserum showed high cross reactivity with rat TNF and partial reactivity with human rTNF. In immunolocalization the antiserum lost most reactivity at a 1:10,000 dilution. In competitive inhibition experiments to demonstrate specificity, murine rTNF alpha, but not rIL- 1 alpha or beta (Pfizer Pharmaceuticals, Groton, CT), inhibited 95% of staining (Table 1). Immunolocalization was performed as follows. Macrophage monolayers were fixed for 5 minutes in 4% paraformaldehyde in phosphate-buffered saline (PBS) and rinsed twice with PBS. Before staining they were fixed for 3 minutes in absolute methanol. In some experiments the methanol step was excluded to preserve membrane integrity. The slides were rinsed again with PBS and treated with a 3% hydrogen peroxide solution to inactivate any remaining peroxidase activity. The slides were next treated with a 1:50 dilution of normal goat serum for 10 minutes at 37 C,
U,
-J
100908070-
0 U)
0: L-J I-
605040-
3020-
01 .i
-10*
_j_
I
0
3
_ .
_
_
j_
_
I
6 HOURS
I
9
12
Figure 1-Kinetics of cell-associated TNF expression in cultured macrophages following stimulation with graded concentrations of lipopolysaccharide. Points are mean ± SD. Dashed line, resident peritoneal macrophages; solid lines, 14 day adjuvant elicited peritoneal macrophages.
566
CHENSUE ET AL
AJP o December 1988
I
A-0 Figure 2-Photomicrographic appearance of elicited macrophages stained for cell associated TNF following stimulation with 100 ng/ml LPS. E-12 hours. D-6 hours. F-Nonimmune serum control at 3 hours. Arrows indicate positively staining hours. C-3 hours. B-1 hour. cells. x1000
then decanted and exposed to a 1:1000 dilution of anti-TNF a or a similar dilution of nonimmune rabbit serum. After 10 minutes of incubation at 37 C, the slides were rinsed three times with PBS, then overlaid with biotinylated goat anti-rabbit IgG (1:200) (Vector Laboratories, Burlingame, CA) and incubated an-
other 10 minutes, followed by three additional rinses with PBS. The slides were next treated with peroxidase labeled streptavidin (Sigma), incubated again, rinsed three times, then overlaid with substrate chromogen (3-amino-9-ethylcarbazole) for 5 minutes at 37 C to allow for color development. Mayer's hema-
IMMUNOLOCALIZATION OF TUMOR NECROSIS FACITOR
Vol. 133 9No. 3
567
Figure 3-Pattems of membrane staining in elicited macrophage stained with anti-TNF. Staining was performed as usual except methanol treatment was eliminated to maintain membrane integrity. The cultures were stimulated for 3 hours with 1 sg/ml LPS. X1 000
toxylin was used as a counterstain. The slides were then observed by light microscopy. In some experiments streptavidin bound to 20 nm (Polysciences, Inc., Warrington, PA) or goat anti-rabbit IgG bound to 40 nm gold particles (Janssen Pharmaceutica, Beerse, Belgium) was used to demonstrate membrane associated TNF. Tumor Necrosis Factor Assay The LM fibroblast cell line was used to measure levels of TNF in supernates according to a modification of the procedure ofRuff and Gifford. " Fifty thousand LM cells in 0.1 ml RPMI-FBS were added to each well of a 96-well microtiter dish, then serial log2 dilutions of test supernate were added to each well. Actinomycin D was then added to each well to achieve a final 1C)o -J -J w
0
30 --
0 0 NO LPS
30--
*
S9
* 1 Ag
70-
Ln
6
0
4
z
0
cJ
0
2 01
0o
I
0|! 0
X 3
I
6 HOURS
9
12
Figure 4-Kinetics of membrane TNF expression in elicited macrophages stimulated with LPS. Points represent mean ± SD.
concentration of 1 ug/ml. The dishes were then incubated for 18 hours at 37 C. After incubation the supernates were discarded and the remaining adherent viable cells were stained with 0.1 ml of crystal violet (0.2% in 2% ethanol) for 2 minutes. The microtiter dishes were rinsed four times with water and the absorbance of each well was read at 600 nm with a Titertek Microelisa Autoreader. TNF units were calclulated from a standard curve generated with serially diluted recombinant murine TNF.
Isolation of RNA
Total RNA was isolated by a modification of the methods of Chirgwin et al'2 and Jonas et al."3 Briefly, macrophage monolayers from 60 mm culture dishes were scraped into a solution consisting of 25 mM Tris buffer (pH 8.0) containing 4.2 M guanidine isothiocyanate, 0.5% Sarkosyl and 0.1 M 2-mercaptoethanol. This mixture was homogenized in a glass tissue grinder then mixed with an equal volume of 100 mM Tris buffer (pH 8.0) containing 10 mM EDTA and 1% SDS. This solution was extracted twice with chloroform-phenol (1:1, vol/vol), then twice with chloroform-isoamyl alcohol (24:1, vol/vol). The nucleic acids were precipitated at -20 C overnight then centrifuged into a pellet (1 5,000g for 1 hour at 4 C). The pellet was washed twice in ice cold 80% ethanol and dissolved in a small volume of 10 mM Tris buffer with 1 mM EDTA, pH 7.6. The poly A mRNA was then enriched using oligo dT cellulose (type 77F, Pharmacia, Piscataway, NJ). The yield and purity of RNA was calculated by absorbance ratio at 260 and 280 nm.
568
CHENSUE ET AL
AJP * December 1988
Vol. 133 * No. 3
IMMUNOLOCALIZATION OF TUMOR NECROSIS FACTOR
569
Figure 5-Demonstration of membrane and cytoplasmic TNF expression. Monolayers of elicited macrophages were permeabilized for 15 minutes with 0.03% tnton X-100 the stained for TNF. A-Diffuse staining pattem obtained following permeabilization. 5-Cytoplasmic staining pattern obtained by blocking surface staining with streptavidin bound to 20 nm gold particles. x1000
Northern Blot Analysis Purified mRNA was separated by formaldehyde/ 1% agarose gel electrophoresis, followed by transblotting to nitrocellulose paper. After baking in a vacuum oven, the nitrocellulose was prehybridized, and then hybridized with either a 30 mer oligonucleotide with homology to murine TNF (5'-GTCCCCCTTCTCCAGCTGGAAGACTCCTCC-3) or a 42 mer ohgonucleotide (5'-GGCTGGGGTGTTGAAGGTCTCAAACATGATCTGGGTCATCTT-3') with specificity for murine beta actin.14"5 Both probes were labeled at the 5' end using [y-32P]ATP (ICN, Irvine, CA) and the T4 polynucleotide kinase reaction. 16 After hybridization the blots were washed extensively and subjected to autoradiography using an intensifying screen.
Scanning Electron Microscopy Macrophage monolayers were prepared on 10 mm glass coverslips and fixed 10 minutes with 2% glutaraldehyde in Sorenson's buffer (pH 7.2, 310 mOsM) at room temperature. The cells were then subjected to immunolocalization for TNF using gold-labeled goat anti-rabbit IgG at the secondary antibody step. The monolayers were washed thoroughly then fixed in 1% OS04 for 60 minutes and dehydrated through graded concentrations of ethanol. After critical point drying in liquid C02, the cells were sputter coated with gold and examined in a ISI Super IIIA. Statistics The Student's t-test was used to determine differences between control and experimental groups. Values of P > 0.05 were considered not significant.
Results Immunohistochemical Demonstration of TNF in Macrophages Both resident peritoneal and Freund's adjuvant elicited peritoneal macrophages were examined for the presence of TNF at various times after stimulus with graded doses of LPS. As shown in Figure 1, resident macrophages were negative throughout the study period regardless of the LPS dose. In contrast, the elicited macrophages showed 12% positively staining cells
at the time of stimulus that increased to about 60%
by 3 hours after stimulus. Interestingly, this decreased significantly by 6 hours, and by 12 hours the population was negative. The high dose of LPS (10 ,ug/ml) prolonged the presence of detectable TNF to 12 hours, but otherwise there was little difference in the observed kinetics. It should be noted that the staining is a qualitative procedure and does not reveal quantitative changes in amounts of TNF per cell. Thus, lower doses of LPS may in fact induce less TNF production per cell. At LPS concentrations of 10 ng/ml and less there were fewer numbers of positive cells (data not
shown). Figure 2 shows the photomicrographic appearance of the elicited cells treated with absolute methanol before staining. It is notable that changes in the staining pattern occurred during the incubation. At 1 and 3 hours, the dye was localized adjacent to the nucleus in nearly all of the positively staining cells. By 6 hours, the dye was seen generally as scattered cytoplasmic particles and was more peripherally located in the cell.
Demonstration of Membrane Associated TNF If the methanol step was eliminated a completely different staining pattern was observed. As seen in Figure 3, the stain revealed cytoplasmic outlines, exposing pseudopodia and lamellapodia, strongly suggesting a membrane staining pattern. The kinetics of the membrane staining pattern was initially identical to the cytoplasmic pattern with about 60% of the cells becoming positive by 1-3 hours (Figure 4). Membrane staining disappeared by 6 hours however, while cytoplasmic staining was still detectable in 30% ofthe cells. Based on the above findings, we surmised that methanol treatment largely stripped away surface membrane TNF but left cytoplasmic intact. To reveal both membrane and cytoplasmic TNF, we attempted to permeablize the cells using a 0.03% Triton-X 100 solution. With this treatment we observed diffuse staining presumably a combination of cytoplasmic and membrane patterns (Figure 5). The membrane component could be completely blocked by treatment with streptavidin-gold just before the streptavidin peroxidase step (Figure 5). The large gold particle had greater difficulty penetrating the membrane and cytoskeleton, yet could still bind to the membrane; this was confirmed by transmission electron micros-
570
CHENSUE ET AL
AJP o December 1988
Figure 6-Demonstration of plasma membrane associated TNF by scanning electron microscopy. Following an initial blocking step with normal goat serum, the monolayers were treated with anti-TNF followed by gold-tagged (40 nm) goat anti-rabbit IgG. The specimens were then subjected to scanning electron microscopy to localize the gold label. A-Anti-TNF serum. B-Nonimmune serum control. x10,000, inset X20,000
copy, which showed only surface binding of the gold particles (data not shown). To confirm the presence of membrane associated TNF, CFA macrophage monolayers were fixed in 2% glutaraldehyde before antibody treatments. After anti-TNF treatment the monolayers were treated with gold tagged goat anti-rabbit IgG then subjected to scanning electron microscopy. As shown in Figure 6, numerous gold particles were seen distributed over the surface membrane of cells treated with anti-TNF but not with nonimmune serum.
Kinetics of TNF mRNA synthesis and TNF Secretion We next wished to determine the relationship of the visualized immunoreactive material to TNF mRNA synthesis and release of bioactive material. Extracts of RNA were prepared from adjuvant elicited macrophages and subjected to Northern blot analysis (Figure 7). Maximum mRNA accumulation occurred at 3 hours then decayed thereafter, paralleling the observed appearance and disappearance of immunoreactive material. The mRNA levels for beta-actin showed no change over the study period. When we measured TNF bioactivity in the supernates of macrophage cultures, activity appeared after 2 hours and reached maximal levels by 4-8 hours. These kinetics corresponded well with the observed disappearance of immunoreactive material from cells (Figure 8). These results provide strong evidence that the synthesis and secretion of TNF was being detected by immunolocalization. Furthermore, they show that
TNF is probably not held in large cytoplasmic stores but is rapidly synthesized and released on demand.
Discussion Macrophages and their products, monokines, are recognized as essential elements in the induction, maintenance and resolution of immune/inflammatory responses. Therefore, a clear understanding of the function, production, and regulation of monokines may allow for the manipulation of aberrant immune responses. The present study demonstrates that TNF, an important macrophage mediator, can be detected immunohistochemically in cultured murine macrophages. This in situ approach allowed for the direct detection of the synthesis and release ofthe monokine. This information extends and lends support to previous observations. Our findings support the notion that TNF is produced by activated macrophage populations, presumably having had a prior exposure to lymphokines or other activating stimuli.'7"18 Nonelicited, resident peritoneal macrophages from CBA/J mice showed no staining for TNF and were incapable of synthesis following LPS stimulus. We also could not demonstrate synthesis of TNF specific mRNA or bioactive material by this population. This finding may in part be strain-related because we have been able to detect small amounts of TNF from the resident population of outbred CD 1 mice. It is clear that primed or elicited macrophages have an augmented capacity to produce TNF. Among elicited macrophages, 12% of the population stained for TNF without stimulus, presumably
producing this mediator in vivo. Moreover, 60% had the capacity to produce TNF on stimulus. Time course studies in our laboratory indicate that the population reaches this level 7 days after CFA injection, suggesting that the macrophages indeed require a period of maturation and/or activation to acquire the capacity to produce TNF. The specific function of TNF in the inflammatory exudate is unknown, although many potential functions have been suggested by in vitro studies such as activation of neutrophils,'8 stimulation of fibroblasts,20 induction of interleukin1,21 colony stimulation factors,22 and la antigen expression.23 Our demonstration of membrane-associated TNF is in accord with previous observations showing a
TNF 28S-~~~~~~~. 18S . . .~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~. . . . .
18S 0 13 6 912
TIME (hours) Figure 7-Kinetics of TNF and beta-actin mRNA expression in elicited macrophages. Macrophage monolayers were stimulated with 1 pg/mI LPS, then RNA was extracted and subjected to Northern blot analysis at the indicated times.
571
IMMUNOLOCALIZATION OF TUMOR NECROSIS FACTOR
Vol. 133 * No.3
40
c120X100-
-A
,
o 60
A DAY 14
/
40-
20z
O--
6-
0
A
O-O 2
4
0 6
10 8 HOURS
0 RES
12
14
16
Figure 8-Kinetics of TNF bioactivity released into supernates following LPS stimulation. Day 14 adjuvantelicited and resident peritoneal macrophage cultures were stimulated with 1 ug/ml LPS, then supernates were collected at subsequent intervals and tested for TNF activity. Unstimulated cultures did not produce TNF. Points represent the mean of three determinations.
membrane bound TNF activity in fixed macrophage and monocyte preparations.7'8 Our kinetic analysis, showing that membrane TNF disappears before cytoplasmic, is intriguing in view of recent studies suggesting that membrane TNF is bound to TNF receptors and is not an integral membrane protein.9 It is tempting to speculate that our findings reflect shedding or internalization of TNF receptors and represents a feedback signal to regulate TNF synthesis as described for other polypeptide hormones.2"26 We are currently testing this hypothesis and initial studies appear to support this model because removal of secreted TNF significantly augments and prolonges TNF synthesis. TNF synthesis also appears to be affected by prostaglandins and the accumulation of these compounds in the supernate likely also contributes to TNF regulation.27 The ability to directly observe monokine production in situ is a powerful tool in the study of these mediators. We have shown that TNF is rapidly synthesized and released over a 6-hour period. It does not appear to be stored in significant amounts and may regulate its own production. Initial studies in our laboratory have shown that this method is applicable to frozen tissue sections and we have demonstrated TNF within foci of granulomatous inflammation. Such experiments should allow for a greater understanding of the spectrum of tissue responses that involve TNF
production.
References 1. Powanda MC, Oppenheim JJ, Kluger MJ, Dinarello CA: International workshop on monokines and other
572
CHENSUE ET AL
non-lymphocytic cytokines: Abstracts. J Leuk Biol 1987,42:534-620 2. Oppenheim JJ, Kovacs EJ, Matsushima K, Durum SK: There is more than one interleukin- 1. Immunol Today 1986, 7:45-56 3. Le J, Vilcek J: Tumor necrosis factor and interleukin 1: Cytokines with multiple overlapping biological functions. Lab Invest 1987, 56:234-248 4. Old LJ: Tumor necrosis factor. Science 1985, 230:630632 5. Beutler B, Cerami A: Cachectin and tumor necrosis factor as two sides of the same biological coin. Nature 1986, 320:384-388 6. Urban JL, Shepard M, Rothstein JL, Sugarman BJ, Schreiber H: Tumor necrosis factor: A potent effector molecule for tumor cell killing by activated macrophages. Proc Natl Acad Sci USA 1986, 83:5233-5237 7. Decker T, Lohmann-Matthes M, Gifford GE: Cell-associated tumor necrosis factor (TNF) as a killing mechanism of activated cytotoxic macrophages. J Immunol 1987, 138:957-962 8. Espevik T, Nissen-Meyer J: Tumour necrosis factorlike activity on paraformaldehyde-fixed monocyte monolayers. Immunology 1987,61:443-448 9. Bakouche 0, Ichinose Y, Heicappell R, Fidler IJ, Lachman LB: Plasma membrane associated tumor necrosis factor: A non-integral membrane protein possibly bound to its own receptor. J Immunol 1988, 140:11421147 10. Beutler B, Mahoney J, Trang NL, Pekala P, Cerami A: Purification of cachetin, a lipoprotein lipase-suppressin hormone secreted by endotoxin-induced RAW 264.7 cells. J Exp Med 1985, 161:984-995 11. Ruff M, Gifford GE: Purification and physiochemical characterization of rabbit tumor necrosis factor. J Immunol 1980, 125:1671-1677 12. Chirgwin JM, Przybyca AE, MacDonald RJ, Rutter WJ: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochem 1979, 18:5294-5299 13. Jonas E, Sargent TD, Davis IB: Epidermal keratin gene expressed in embryos of Xenopus laevis. Proc Natl Acad Sci USA 1985, 82:5413-5417 14. Pennica D, Hayflick JS, Bringman TS, Palladino MA, Goeddel DV: Cloning and expression in Escherichia coli of the cDNA for murine tumor necrosis factor. Proc Natl Acad Sci USA 1985, 82:6060-6064 15. Tokunaga K, Taniguchi H, Yoda Y, Sakiyama: Nucleotide sequence of a full-length cDNA for mouse cy-
AJP * December 1988
toskeletal beta actin mRNA. Nucleic Acid Res 1986, 14:2829 16. Maniatas T, Fritsch EF, Sambrook J: Molecular Cloning: A Laboratory Manual. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory, 1982 17. Collart MA, Belin D, Vassalli JD, De Kossodo S, Vassalli P: Gamma interferon enhances macrophage transcription of the tumor necrosis factor/cachectin, interleukin 1 and urokinase genes, which are controlled by short-lived repressors. J Exp Med 1986, 164:2113-2118 18. Gifford GE, Lohmann-Matthes ML: Gamma interferon priming of mouse and human macrophages for induction of tumor necrosis factor production by bacterial lipopolysaccaride. JNCI 1987, 78:12 1-124 19. Shalaby MR, Aggarwal BB, Rinderknecht E, Svedersky LP, Finkle BS, Palladino MA: Activation of human polymorphonuclear neutrophil functions by interferon-y and tumor necrosis factor. J Immunol 1985, 135: 2069-2073 20. Vilcek J, Palombella VJ, Hendrikson-DeStephano D, Swenson C, Feinman R, Hirai M, Tsujimoto M: Fibroblast growth enhancing activity of tumor necrosis factor and its relationship to other polypeptide growth factors. J Exp Med 1986, 163:632-643 21. Bachwich PR, Chensue SW, Larrick JW, Kunkel SL: Tumor necrosis factor stimulates interleukin 1 and prostaglandin E2 production in resting macrophages. Biochem Biophys Res Comm 1986, 136:94- 10 1 22. Vogel SN, Douches SD, Kaufman EN, Neta R: Induction of colony stimulating factor in vivo by recombinant interleukin 1 g and recombinant tumor necrosis factor-a. J Immunol 1987, 138:2143-2148 23. Chang RJ, He Lee S: Effects of interferon-y and tumor necrosis factor-a on the expression of an Ia antigen on a murine macrophage cell line. J Immunol 1986, 137: 2853-2856 24. Dower SK, Urdal DL: The interleukin- 1 receptor. Immunol Today 1987, 8:46-51 25. Unglaub R, Maxeiner B, Thoma B, Pfizenmaier K, Scheurich P: Downregulation of tumor necrosis factor (TNF) sensitivity via modulation of the TNF binding capacity by protein kinase C activators. J Exp Med 1987, 166:1788-179 26. Gorden P, Carpentier JL, Fan JY, Orci L: Receptor mediated endocytosis of polypeptide hormones: Mechanism and significance. Metabolism 1982, 7:664-669 27. Kunkel SL, Wiggins RC, Chensue SW, Larrick J: Regulation of macrophage tumor necrosis factor production by prostaglandin E2. Biochem Biophys Res Comm 1986, 137:404-410
