reoxygenation affects cytokine elaboration by murine peritoneal macrophages

Journal

of Pediatric

Surgery

VOL 31, NO 11

NOVEMBER

1996

The Interval Between a Septic Stimulus and Hypoxia/Reoxygenation Affects Cytokine Elaboration by Murine Peritoneal Macrophages By Gajra Arya and Victor F. Garcia Cincinnati, Ohio l The critically ill patient is commonly exposed to various physiological insults. The authors have previously shown that in vivo hypoxia/ reoxygenation (H/R) alters the pattern of cytokines elaborated by murine peritoneal macrophages given a septic stimulus. In this study they sought to determine whether the interval between a septic stimulus and H/R affected the release of inflammatory mediators by macrophages. Adult CBA-strain mice were injected intraperitoneally with 10 pg of lipopolysaccharide (LPS). On day 0,1,2. 3, 4, or 5 after LPS injection, animals were exposed to 16 hours of hypoxia followed by 2 hours of reoxygenation. Harvested peritoneal macrophages were restimulated in vitro with 2.5 pg/mL LPS or left unstimulated. Culture supernatants collected at 2,4,6,8,10, and 12 hours after LPS injection were assayed for tumor necrosis factor (TNF), prostaglandin E2 (PGEz), and nitric oxide (NO) production. Macrophage-derived mediator production peaked when H/R occurred 3 days following LPS injection (P < .05). These data suggest that the interval between sepsis and subsequent H/R influences the pattern of cytokines elaborated by peritoneal macrophages given a septic stimulus. Copyright a 1996 by W.B. Saunders Company

INDEX ation,

WORDS: inflammatory

Endotoxin pretreatment, mediators.

hypoxia,

reoxygen-

I

N EXPERIMENTAL animals, the effect of LPS closely resembles the state of human septic shock and is commonly associated with reduced microcirculatory blood flow and vasoconstriction.’ In response to LPS, activated macrophages produce cytokines and other inflammatory mediators, including tumor neurosis factor (TNF), prostaglandin E2 (PGE:) and nitric oxide (NO) which contribute to the pathogenesis of septic shock.’ Tissue hypoxia after shock may potentiate the immune response to injury. It has been shown that hypoxia and H/R alter cytokine production by mononuclear celW4 and that pretreatment of animals with endotoxin, cytokines like TNF and interleukin I (IL-l) induce tolerance to a second stimulus.s-7 Our previous work demonstrated increased TNF, but decreased PGE? and NO levels by peritoneal macrophages of mice subjected to H/R 5 days after Jocirna/ofPedratrrcSurgery,

Vol31,

No 11 (November),

1996, pp 1469.1474

endotoxin pretreatment.” The objective of this study was to determine if the interval between endotoxin pretreatment and subsequent H/R affects the amplitude and duration of cytokine elaboration by murine peritoneal macrophages. MATERIALS

AND METHODS

Housing arid In VirgoPriming Adult weighing

male CBA mice 20 to 25 g were

(Jackson housed

I?-hour hght-dark cycle with and water ad libitum. Upon were injected intraperitonealy

Laboratories. Bar Harbor, ME), m rooms maintained at 22°C on a

access to standard laboratory food acclimating for 3 days, the animals (IP) with 10 kg of Escherichm colr

llpopolysaccharide (LPS) 0127:Bg (Dlfco Laboratories. Detroit, MI). This study was approved by the Children’s Hosprtal Research Foundation InstItutional Animal Care and Use Committee.

to Hypoxia

Exposure

Following IP injection with LPS. the animals were drvlded into SIX groups, day 0, 1. 7, 3, 4. or 5. and subjected to 16 hours of hypoxia by placmg them rn a controlled environment chamber. The day 0 animals were placed in the hypoxic following LPS injection. day 1 animals were hours after LPS injection, whereas day

chamber subjected 2 animals

Immediately to hypoxia 24 were made

hypoxic 48 hours after the injectlon with LPS, and so on. The chamber was flushed with a gas mixture contaming

8%

0:.

35 CO:. and 89% N?. at a Row rate of 5 Limin. until the oxygen content of the atmosphere mslde was reduced to 8°F. The flow rate was then maintained at 1.5 Limin for 16 hours. At the end of this hypoxlc confirmed

period. oxygen by an oxygen

concentration monitor (Catalyst

MD). Immediately after the 16 hours reoxygenated for 2 hours hy removing

Presented the Amencan 1994.

insIde the Research,

of hypoxia. them from

chamber Owmgs

the animals the chamber

was Mills, were and

at the 1993 Annttal Meetq of the Sectwn on Surgeq of Acudenw of Pediatrics. Dallus. Texas. October 21-2.1.

1469

ARYA

1470

placing them in an ambient atmosphere (normoxia). Throughout the experimental period, animals were fasted but were given water ad libitum.

Isolation and Culture of Macrophages Immediately after reoxygenation, the animals were killed by cervical dislocation and their peritoneal exudate cells were harvested by repeatedly lavaging the peritoneal cavity with cold endotoxin-free phosphate-buffered saline (pH, 7.2). The lavage fluid was centrifuged at 5OOgfor 5 minutes, and was resuspended at a final cell concentration of 1.5 x 106/mL in phenol-red-free DMEM (Sigma, St Louis, MO) containing 10% fetal calf serum and 100 U/mL of antibiotics. Cells were plated at 1 mL/well in a 24-well (Falcon) flat bottom culture dish. The macrophages (adhered cells) were then either stimulated in vitro with 2.5 ug/mL of LPS or left unstimulated, and incubated at 37°C in 5% COr.

Assay of Supematants Cell-free culture supernatants were collected from designated wells for each individual time point for measurement of cytokine and prostaglandin activity at 2,4,6,8,10, and 12 hours after in vitro plating. All samples were assayed in duplicate. Figure 1 shows a schematic of the experimental design.

Nitric Oxide Secretion Nitric oxide concentration was measured as NOz- by the Griess reaction9 and expressed in micromoles (uM). Briefly, lOO-)LL aliquots of freshly collected supernatant were incubated with an equalvolume of Griess reagent (1% sulphanilamide/O.l% naphthylethylene diamine dihydrochloride/2.5% HsPO.+) at room temperature for 10 minutes, and the absorbance at 546 nm was determined by a microplate reader (Molecular Devices Corp, Menlo Park, CA). Nor- content was determined by linearly plotting serial dilutions of sodium nitrite as standard, and culture medium alone serving as blank. Supernatants were then frozen at -70°C until further determination of TNF and PGEr.

TNF Evaluation TNF levels were determined by testing the supernatants for WEHI clone 13 cytotoxicity.iO Briefly, 100 uL of serial dilutions of macrophage supernatant were added to WEHI cells plated at 5 x 104/well in the presence of 0.5 ug/mL of actinomycin D in a 96-well (Falcon) microtiter plate. After incubation for 20 hours, the plates were stained with MTT and incubated for an additional 4 hours. After incubation, the cell supernatant was removed and 100 uL/well of isopropranol/HCl mixture was added to the wells. The optical density was read at the dual wavelength of 570/650 nm and plotted as a semilog of color intensity. The assay was standardized by human rTNF (Genzyme Corp, Cambridge, MA) and the units expressed in nanograms per milliliter (ng/mL).

Collect supernatants 2, 4, 6, 6, 10, and

at 12hours

I

Fig 1.

Schematic

Stimulate (2SPgIml or not stimulate

+ I

I

of experimental

design.

LPS) I

AND

GARCIA

PGE2 Detemtination Unextracted supernatants were assayed for PGEz content by EIA (Cayman Chemical Co, Ann Arbor, MI) and the detected units expressed in nanograms per milliliter (ng/mL).

Statistical Analysis Each group had a sample size of 10 or more, and the data are expressed as mean f SEM. Data were analyzed by the Student’s t test and ANOVA of repeated measures: P values of less than .05 were considered to be significant. RESULTS

TNF Release

Stimulated macrophages produced peak amounts of TNF when H/R took place 3 days after endotoxin pretreatment (Fig 2A). Unstimulated macrophages produced maximal TNF levels when H/R occurred 2 days after LPS injection (Fig 2B). TNF production by both stimulated and unstimulated macrophages declined after peaking and decreased throughout the period of in vitro incubation. TNF production on each day was significantly different from the previous days (P < .Ol). At all time points, stimulated macrophages produced significantly more TNF than unstimulated macrophages (P < .Ol). PGE2 Production

The pattern of PGE2 production by both stimulated (Fig 3A) and unstimulated (Fig 3B) macrophages was similar, ie, levels peaked on day 3 and plateaued thereafter. No statistical differences in PGE2 levels were noted between days 2,3,4, and 5, or between days 0 and 1. PGE,! levels on days 0 and 1 were significantly lower than that on all the other days (P < .05). Additionally, the amount of PGE2 produced by stimulated macrophages was significantly greater than that by unstimulated macrophages at all time points. PGEz levels increased throughout the period of in vitro incubation and were most elevated at 12 hours following incubation. NO Production

Both stimulated (Fig 4A) and unstimulated (Fig 4B) macrophages produced peak amounts of NO when H/R took place 3 days after endotoxin pretreatment. NO production on each day was significantly different from the previous days (P < .Ol). NO production by stimulated macrophages was significantly higher than unstimulated macrophages only after 8 hours of in vitro incubation. NO levels increased with duration of in vitro incubation and were highest at 12 hours after incubation.

SEPSIS AND

HYPOXIA/REOXYGENATION

1471

q n q q q q

2hr 4hr 6hr 8hr 10hr 12hr

8 6 q 2h n 4h q 6h n 8h fl

q

2

Fig 2. (A) TNF production by stimulated macrophages. (B) TNF production by unstimulated macrophages.

B0

DISCUSSION

These data suggest that the time interval between a septic stimulus and subsequent H/R affects cytokine release by peritoneal macrophages. This effect is demonstrated by a time-dependent change in the amplitude of TNF, PGE*, and NO production by macrophages. In the case of stimulated macrophages, peak production of all inflammatory mediators was noted when H/R occurred 3 days after the injection with LPS. In unstimulated macrophages a bimodal peak was seen for the production of TNF and NO, the levels of both mediators decreasing on day 1. Peak production of TNF occurred on day 2 and NO levels peaked on day 3. In both stimulated and unstimulated macrophages, TNF and NO levels declined after peaking, whereas PGEz production plateaued thereafter. TNF levels decreased throughout the period of in vitro incubation and were maximum at 2 hours after

2

3

4

10h 12h

5

DAYS

incubation. In contrast to TNF production, PGEz and NO levels increased throughout and were highest at 12 hours after incubation. This pattern of mediator production remained the same for stimulated as well as unstimulated macrophages. Amounts of TNF and PGE2 produced at each time point by the stimulated macrophages was significantly higher than those produced by the unstimulated macrophages. NO production by stimulated macrophages was significantly higher but only after 8 hours of in vitro incubation. The significant differences in the amounts of mediators produced suggests that despite a profound H/R insult, pretreated macrophages are able to respond to a second LPS stimulus. In previous studies we found that when H/R followed 5 days of pretreatment with either LPS or saline, peritoneal macrophages produced significantly more TNF but less NO than the normoxic control animals. However, LPS-pretreated animals

their cytotoxicity,r3-I5 suggesting that a negative feedback regulation may exist. TNF, a product of activated macrophages, has been found not only to mediate cell cytotoxicity I6 but also to act as an endogenous pyrogen” and stimulate PGE2 synthesis.18Macrophage-derived PGEz has been implicated as an endogenous regulator that shuts off macrophage activation via suppression of TNF synthesis.14,15,19-21

The results reported here show that PGE, can affect macrophage cytotoxicity with regard to the release of TNF. Levels of TNF were shown to increase gradually over days and then decline, in comparison with PGEz levels, which increased and plateaued. This finding suggests that activated macrophages produce TNF that induces the synthesis of PGE2, which in turn, down-regulates the release of TNF. With increased and sustained PGEz production, TNF production wanes. These findings are in

SEPSIS

AND

HYPOXIA/REOXYGENATION

1473

q n q q

2h 4h 6h 8h

III]

10h 12h

q

A

0

1

2

3

4

5

DAYS

n 4h q 6h q 8h q 10h n 12h

5

Fig 4. (A) NO production by stimulated macrophages. (B) NO production by unstimulated macrophages.

0.0

B

0

1

accordance with the concept that PGEz limits expression of TNF by negative feedback inhibition.19-21 The mechanism for H/R-mediated increased TNF production is unclear. It may be mediated by the release of oxygen-free radicals, which in an autocrine and paracrine fashion, promote the production of inflammatory cytokines.3 Macrophages activated by LPS, pro-inflammatory cytokines, or interferons also produce the cytotoxic radical, N0.22-24 These activators can work synergistically2Js,‘6 or as sole agents2s27,28 for increased secretion of NO. In this study, NO production, by both stimulated and unstimulated macrophages, paralleled the pattern of TNF production and not the pattern of PGE?. This is consistent with the ability of anti-TNF antibodies to suppress NO production, suggesting that endogenous TNF is involved in the induction of N0.24.‘9 Our observation is in contrast to that of Gaillard et aF7who showed a direct correlation between amount of PGE? and

2

DAYS

3

4

5

nitrite levels. However, Zhang et al”” provide evidence that pretreatment of macrophages with LPS enhanced NO production but down-regulated TNF production, suggesting that there may be mechanistic differences between priming and stimulating macrophages to produce TNF. TNF, a principal mediator of sepsis, induces the release of proinflammatory cytokines, which cause tissue injury.31 Clinical trials have shown that treatment of severely septic patients with anti-TNF monoclonal antibody may be beneficial.32 However, the most efficacious dose and time of administration of anti-TNF antibody remain undefined. Our study suggests that there is a specific interval, after an initial septic stimulus, during which H/R can result in significant alterations in mediator release. This finding may help determine the time when administration of anti-TNF antibody would be most efficacious in the treatment of such sequential insults.

ARYA AND GARCIA

1474

REFERENCES 1. Schumaker PT, Samael RW: Oxygen delivery and uptake by peripheral tissues: Physiology and pathophysiology. Crit Care Clin 5:255-269, 1989 2. Ding AH,

Nathan CF. Stuehr DJ: Release of reactive nitrogen intermediates and reactive oxygen intermediates from mouse peritoneal macrophages. Comparison of activating cytokines and evidence for independent production. J Immunol 141: 2407-2412,198s 3. Ghezzi P, Dinarello CA, Bianchi M, et al: Hypoxia increases production of interleukin-1 and tumor necrosis factor by human mononuclear cells. Cytokine 3:189-194, 1991 4. Koga S, Ogawa S, Kuwabara K, et al: Synthesis and release of interleukin-1 by reoxygenated human mononuclear phagocytes. J Clin Invest 90:1007-1015, 1992 5. Li MH, Seatter SC, Manthei R, et al: Macrophage endotoxin tolerance: Effect of TNF or endotoxin pretreatment. J Surg Res 57:85-92, 1994 6. Vogel SN,

Kaufman EN, Tate MD, et al: Recombinant interleukin-la and recombinant tumor necrosis factor (Y synergize in vivo to induce early endotoxin tolerance and associated hematopoietic changes. Infect Immun 56:2650-2657,198s 7. West MA, Li MH, Seatter SC, et al: Pre-exposure to hypoxia or septic stimuli differentially regulates endotoxin release of tumor necrosis factor, interleukin-6, interleukin-1, prostaglandin Ez, nitric oxide, and superoxide by macrophages. J Trauma 37:82-90, 1994 8. Arya G, Garcia VF: Hypoxiaireoxygenation affects endotoxin tolerance. J Surg Res 59:13-16, 1995 9. Green LC, Wagner DA, Glowgowski J, et al: Analysis of nitrate, nitrite and ( isN) nitrate in biological fluids. Anal Biochem 126:131-138. 1982 10. Espevik T, Nissen-Meyer J: A highly sensitive cell line, WEHI 164 clonel3, for measuring cytotoxic factor/tumor necrosis factor from human monocytes. J Immunol Methods 95:99-105, 1986 11. Kleinschmidt WJ, Schultz RM: Similarities of murine gamma interferon and the lymphokine that renders macrophages cytotoxic. J Interferon Res 2:291-299,1982 12. Grabstein KH, Urdal DL, Tushinski RJ, et al: Induction of macrophage tumoricidal activity by granulocyte-macrophage colonystimulating factor. Science 232:506-508, 1986 13. Poste G, Kirsh R: Rapid decay of tumoricidal activity and loss of responsiveness to lymphokines in inflammatory macrophages. Cancer Res 39:2582-2590,1979 14. Taffet SM, Russel SW: Macrophage-mediated tumor cell killing: regulation of expression of cytolytic activity by prostaglandin E. J Immunol 126:424-427,198O 15. Taffet SM, Pace JL. Russel SW: Lymphokine maintains macrophage activation for tumor cell killing by interfering with the negative regulatory effect of prostaglandin Ez. J Immunol 127:121124,198l 16. Mannel DN, Meltzer MD, Mergenhagen SE: Generation and characterization of a lipopolysaccharide-induced and serumderived cytotoxic factor for tumor cells. Infect Immun 28:204-211, 1980

17. Dinarello CA, Cannon JG, Wolff SM, et al: Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin 1. J Exp Med 163:1433-1450,1986 18. Dayer JM, Beutler B, Cerami A: Cachectinitumor necrosis factor stimulates collagenase and prostaglandin EZ production by human synovial cells and dermal fibroblasts. J Exp Med 162:21632168,1985

19. Schultz RM, Pavlidis NA, Stylos WA, et al: Regulation of macrophage tumoricidal function: a role of prostaglandins of the E series. Science 202:320-321,197s 20. Goodwin JS, Webb DR: Regulation of the immune response by prostaglandins. Clin Immunol 1mmunopatho115:106-122,198O 21. Renz H, Gong JH, Schmidt A, et al: Release of tumor necrosis factor-o from macrophages: Enhancement and Suppression are dose-dependently regulated by prostaglandin Ez and cyclic nucleotides. J Immunol 141:2388-2393,1988 22. Stuehr DJ, Marletta MA: Mammalian nitrate biosynthesis: mouse macrophages produce nitrite and nitrate in response to Eschenchia coli lipopolysaccharide. Proc Nat1 Acad Sci U S A 821773%7742,1985 23. Stuehr DJ,

Marletta MA: Induction of nitrite/nitrate synthesis in murine macrophages by BCG infection, lymphokines, or interferon-y. J Immunol139:518-525,1987 24. Rockett KA, Awburn MM, Aggarwal BB, et al: In vivo induction of nitrite and nitrate by tumor necrosis factor, lymphotoxin, and interleukin-1: Possible roles in malaria. Infect Immun 60:3725-3730,1992 25. Stenger

S, Solbach W, Rollinghoff, et al: Cytokine interactions in macrophages is induced by the synergistic action of 1FN-r and IL-4 and accounts for the antiparasitic effect mediated by IFNy and IL-4. Eur J Immunol21:1669-1675,199l 26. Gaillard T, Mulsch A, Busse R, et al: Regulation of nitric oxide production by stimulated rat kupffer cells. Pathobiology 59:280-283, 1991 27. Gaillard

T, Mulsch A, Klein H, et al: Regulation by prostaglandin Ez of cytokine-elicited nitric oxide synthesis in rat liver macrophages. Biol Chem 372897-9021992 28. Lamas S, Michel T, Brenner BM, et al: Nitric oxide synthesis in endothelial cells: evidence for a pathway inducible by TNF-a. Am J Physiol261:C634-641,199l 29. Drapier JC, Wietzerbin J, Hibbs JB: Interferon-y and tumor necrosis factor induce the L-arginine-dependent cytotoxic effector mechanism in murine macrophages. Eur J Immunol 18:1587-1592, 1988 30. Zhang X, Morrison DC: Lipopolysaccharide-induced selective priming effects on tumor necrosis factor cx and nitric oxide production in mouse peritoneal macrophages. J Exp Med 177:511516,1993 31. Beutler B, Cerami AC: The common meditor of shock, cachexia and tumor necrosis. Adv Immunol42:213-231,1988 32. Fisher CJ, Opal SM, Dhainaut J, et al: Influence of an anti-tumor necrosis factor monoclonal antibody on cytokine levels in patients with sepsis. Crit Care Med 21:318-327,1993