Peritonitis Inflammatory Response in Immune Complex Macrophages

0 downloads 0 Views 296KB Size Report
Immunologists All rights reserved. Copyright © 1999 by ... ICs with Fc RIII and the subsequent release of TNF, initiate the inflammatory response in BALB/c and C57BL/6 mice. The ... initiation of IC peritonitis in the model of the reverse passive.


Cutting Edge: Fc Receptor Type I for IgG on Macrophages and Complement Mediate the Inflammatory Response in Immune Complex Peritonitis1,2 Tanja Heller,* J. Engelbert Gessner,† Reinhold E. Schmidt,† Andreas Klos,* Wilfried Bautsch,* and Jo¨rg Ko¨hl3*

The contributions of Fc receptors (FcRs) for IgG (FcgRs) and complement to immune complex (IC)-mediated peritonitis were evaluated in BALB/c-, C57BL/6-, FcRg chain-, and FcR type III for IgG (FcgRIII)-deficient mice, backcrossed to the C57BL/6 background. In BALB/c mice, but not in C57BL/6 mice, neutrophil migration was markedly attenuated after complement depletion. In mice lacking FcRg chain, neutrophil migration was abolished, whereas it was unaffected in FcgRIII-deficient mice. Huge amounts of TNF-a (TNF) were found in the peritoneal exudate of BALB/c and C57BL/6 mice but were absent in mice lacking FcRg chain or FcgRIII. Surprisingly, a functional inhibition of TNF in BALB/c and C57BL/6 mice had no effect on neutrophil infiltration. These data provide evidence that in IC peritonitis, the activation of FcR type I for IgG on peritoneal macrophages and the activation of the complement cascade, but not the interaction of ICs with FcgRIII and the subsequent release of TNF, initiate the inflammatory response in BALB/c and C57BL/6 mice. The Journal of Immunology, 1999, 162: 5657–5661.

I

mmune complex (IC)4-triggered inflammation was described in an experimental model by Maurice Arthus in 1903 and is now widely used as a reverse passive Arthus reaction in which specific Ab and Ag are passively introduced across a vas-

*Institute of Medical Microbiology, and †Department of Clinical Immunology, Medical School Hannover, Hannover, Germany Received for publication February 9, 1999. Accepted for publication March 9, 1999. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by Grant 01 VM 9305 from the Bundesministerium fu¨r Bildung, Wissenschaft, Forschung and Technologie (to J.K) and by grant from the Sonderforschungsbereich 265 (to J.E.G and R.E.S.). 2 All protocols used in the care and experimentation of mice were in accordance with the German law for the protection of animals and have been approved by the Bezirksregierung (Hannover, Germany). 3 Address correspondence and reprint requests to Dr. Jo¨rg Ko¨hl, Institut fu¨r Medizinische Mikrobiologie, Medizinische Hochschule Hannover, 30623 Hannover, Germany. E-mail address: [email protected]

Abbreviations used in this paper: IC, immune complex; FcR, Fc receptor; FcgR, FcR for IgG; FcgRI, FcR type I for IgG (CD64); FcgRII, FcR type II for IgG (CD32); FcgRIII, FcR type III for IgG (CD16); TNF, TNF-a; CVF, cobra venom factor; PMN, polymorphonuclear neutrophil; C5a, C5a anaphylatoxin.

4

Copyright © 1999 by The American Association of Immunologists



cular barrier (1). It has long been accepted that Ag-Ab complexmediated classical pathway activation of complement is the initial step leading to the release of chemotactic peptides that, in turn, cause neutrophil migration into the inflammatory focus (2). During the last few years, mice with genetically engineered deficiencies of Fc receptors (FcRs) for IgG (FcgRs) became available, giving new insights into the mechanisms involved in the pathogenesis of inflammation in IC diseases (3, 4). Three different types of murine FcgRs exist. The high-affinity FcR type I for IgG (FcgRI) (CD64), which is exclusively expressed on macrophages and neutrophils, binds both monomeric IgG and ICs. It has been demonstrated to mediate Ab-dependent cellular cytotoxicity and phagocytosis in response to cross-linking by Ab (5). The low-affinity FcgRs, FcR type II for IgG (FcgRII) (CD32) and FcR type III for IgG (FcgRIII) (CD16) bind IgG Abs only within Ag-Ab complexes. FcgRII is widely distributed on both lymphoid and myeloid cells and functions as a negative regulator of both Ig-induced B cell activation and IC-mediated activation of mast cells (6). The expression of FcgRIII is restricted to macrophages, neutrophils, NK cells, and mast cells. FcgRI, FcgRIII, and high-affinity FceRI share a signal transduction unit, the g-chain. Studies with FcRg chaindeficient mice revealed that IgG FcRs may be the initial trigger in IC diseases in the skin (7). Supporting this concept, mice with gene target disruption of complement components C3 or C4 as well as the C5-deficient strain DBA/2 showed no impaired Arthus reaction (8). Reconstitution experiments with mast cells and studies with FcgRIII deficient mice demonstrated that FcgRIII on mast cells is crucial in eliciting the inflammatory response in IC disease in the skin (4, 9). We have analyzed the role of FcgRs and complement in the initiation of IC peritonitis in the model of the reverse passive Arthus reaction in different mouse strains (i.e., in BALB/c-, C57BL/6-, FcgRIII-, and FcRg chain-deficient mice). Our studies give evidence that both complement and FcgRs play important roles in the initiation of IC disease. In contrast to the skin, for which the activation of mast cells via FcgRIII is of major importance (4, 9), we found that IC-induced peritonitis in C57BL/6 mice is predominantly triggered by FcgRI on peritoneal macrophages. As it is dependent upon the genetic background, the contribution of complement or FcgRs is diverse. 0022-1767/99/$02.00

5658

CUTTING EDGE

Materials and Methods Animals BALB/c and C57BL/6 mice were purchased from Charles River Laboratories (Wilmington, MA). FcgRIII-deficient animals were generated in our laboratory as described previously (4). A minimum of 12 backcross generations to the background strain C57BL/6 have been made to yield congenic strains. FcRg chain-deficient mice (C57BL/6JMTacfBR-[KO]) backcrossed 12 generations to C57BL/6 mice were obtained from Taconic (Germantown, NY).

Arthus reaction To induce the reverse passive Arthus reaction in the peritoneum, chicken egg albumin (20 mg/kg body weight; Sigma, Deisenhofen, Germany) was injected i.v. subsequently followed by an i.p. injection of rabbit polyclonal IgG rich in Ab to chicken egg albumin (800 mg/mouse; ICN, Eschwege, Germany) exactly as described previously (10). BALB/c mice were killed 2.5, 5, 7.5, 10, 15, 30, 60, 90, 120, 150, 180, 240, 300, 360, and 480 min after the application of IC; C57BL/6 mice were killed 5, 10, 30, 60, 90, 240, 360, and 480 min after injury. The peritoneal cavity was lavaged with 2 ml of ice-cold PBS and 0.1% BSA. Peritoneal cells were pelleted, stained using Diff-Quick (Baxter Dade, Du¨dingen, Switzerland), and subsequently assessed for differential cell count. The supernatant was used for the determination of TNF-a (TNF) concentrations. Where indicated, mice were injected i.v. with 8 mg of cobra venom factor (CVF) on the day before IC challenge and again with 8 mg of CVF 4 h before IC application. CVF was purified from Naja naja venom (Miami Serpentarium Laboratories, Miami, FL) according to a previously described procedure (11). Serum complement levels were determined as described previously (12).

Determination of TNF concentrations in serum and in the peritoneal cavity The TNF concentrations in serum and in the peritoneal lavage fluid were assessed with a mouse TNF ELISA kit (Genzyme, Virotech, Russelsheim, Germany), according to the manufacturer’s instructions. The detection limit of the assay was 31 pg/ml.

Statistical analysis Statistical analysis was performed using the SigmaStat version 2.0 statistical package (Jandel, Erkrath, Germany). First, we tested for a normal distributed population using the Kolmogorov-Smirnov test. To analyze differences between more than two normally distributed groups, a one-way ANOVA was performed. Pairwise comparisons were then performed using Tukey’s test. To analyze differences between two normally distributed groups, an unpaired Student’s t test was used. A p value of ,0.05 was considered to be significant; a p value of ,0.001 was considered to be highly significant.

Results and Discussion Kinetics of neutrophil influx and TNF release The hallmark of the inflammatory response in the reverse passive Arthus reaction in the peritoneum is the neutrophil influx into the peritoneal cavity. We found that neutrophil migration started 2 h after IC challenge, reaching the highest value of ;70% neutrophils after 8 h in both BALB/c and C57BL/6 mice (Fig. 1A). Recent data obtained with mast cell-deficient WBB6-F1-W/Wv (W/Wv) mice suggest that the neutrophil influx is due to the release of preformed TNF from peritoneal mast cells, already detectable 5 min after the application of IC (13). In contrast, we found measurable TNF concentrations in the peritoneal exudate of BALB/c and C57BL/6 mice not earlier than 1 h after IC application, reaching a maximum between 90 and 150 min after initiation of the Arthus reaction. Subsequently, TNF concentrations declined and were no longer detectable after $6 h (Fig. 1B). In the normal littermates of mast cell-deficient W/Wv mice, a completely different kinetic has been described. In addition to the first TNF peak, 5 min after IC challenge a second peak occurred with a maximum at 6 h, which paralleled the polymorphonuclear neutrophil (PMN) influx into the peritoneal cavity. From these data, Zhang et al. concluded that the source of the second TNF peak was the migrated PMNs (13). In fact, we did not observe the first peak in a TNF ELISA, although

FIGURE 1. Kinetics of neutrophil influx (A) and TNF release (B) into the peritoneal cavity or into the serum (C) of BALB/c and C57BL/6 mice. Mice were killed at indicated times (after 2.5, 5, 7.5, 10, 15, 30, 60, 90, 120, 150, 180, 240, 300, 360, and 480 min in BALB/c mice; after 5, 10, 30, 60, 90, 240, 360, and 480 min in C57BL/6 mice) (A and B). Serum samples were determined at 2.5, 15, 30, 90, 120, 150, 240, 360, and 480 min after the application of IC (C). Data are expressed as the mean 6 SD (n 5 3–13).

The Journal of Immunology

5659 cient mice, with the latter strain completely lacking functional FcgRI, FcgRIII, and FceRI (Fig. 2). These data provide evidence that the interaction of IC with FcgRIII is the predominant event triggering TNF release in the mouse strains investigated. Zhang et al. have shown that TNF is released from mast cells in W/Wv mice. However, the molecular mechanisms leading to TNF release have not been investigated (13). In two other reports, FcgRIII was found to mediate edema, hemorrhage, and PMN recruitment in the Arthus reaction in skin. No data have been provided that close the gap between cellular activation and the pathologic sequels (4, 7). Inflammatory response in IC peritonitis is mediated by different pathways in BALB/c or C57BL/6 mice

FIGURE 2. Effect of complement depletion and FcgR deficiencies on TNF release in the peritoneal cavity. At 90 min after IC challenge, TNF concentrations were significantly elevated in C57BL/6 mice compared with control animals that received either the Ab but no Ag (PBS i.v. 5 Ab control) or the Ag but no Ab (PBS i.p. 5 Ag control). In FcgRIII- and FcRg chain-deficient animals, TNF levels were significantly decreased compared with C57BL/6 mice. Results are expressed as the mean 6 SD (n 5 4 – 6). p†, p , 0.001, as assessed by one-way ANOVA. Pairwise comparisons were performed using Tukey’s test.

we analyzed the peritoneal fluid 2.5, 5, 7.5, and 10 min after the application of Ag and Ab, or observe the TNF peak after 6 h. Because the TNF peak in the peritoneal exudate of BALB/c and C57BL/6 mice was found in the absence of neutrophils (i.e., 90 – 150 min after IC challenge) and declined when neutrophils started migration into the peritoneum (.150 min), this cell type is probably not the source of peritoneal TNF in BALB/c and C57BL/6 mice as suggested for mast cell-deficient W/Wv mice and their normal littermates (13). This difference might be due to the genetic background of the WB/Re-W mouse strain present in the W/Wv animals. The most likely candidates for the observed TNF release in BALB/c and C57BL/6 mice are macrophages, which express FcgRI, FcgRII, and FcgRIII (5). Unlike mast cells, macrophages have no stores for TNF but have to synthesize the cytokine de novo (14, 15). TNF was also found in serum with similar kinetics as observed in the peritoneum. Surprisingly, however, TNF was already detectable 30 min after initiation of the Arthus reaction, which is 30 min earlier compared with the TNF levels in the peritoneum. In addition, the maximum concentrations were more than twice as high as in the peritoneum and did not decline to zero as described for the peritoneal fluid (Fig. 1C). These data strongly suggest an IC-mediated activation of circulating monocytes and neutrophils via FcgRs. TNF release in IC peritonitis is mediated through FcgRIII We subsequently evaluated whether the TNF release in BALB/c and C57BL/6 mice depends upon the activation of the complement system, as has been demonstrated in C5-deficient (OSNJ) mice, their normal controls (NSNJ), as well as in mast cell-deficient mice (W/Wv), and their normal littermates (16). Complement activity was abolished by treating the mice with CVF, forming a stable CVFBb complex to cause a brisk and total cleavage of C3, resulting in a complement depletion downstream of C3. The efficacy of the CVF treatment was proven by a complete abrogation of serum hemolytic activity (data not shown). Complement depletion did not affect the TNF release found in the peritoneal cavity of BALB/c and C57BL/6 mice (Fig. 2). However, a marked and identical reduction of TNF was observed in FcgRIII- and FcRg chain-defi-

TNF can induce neutrophil recruitment by a number of different actions (i.e., induction of neutrophil chemoattractants such as IL-8, up-regulation of E-selectin on venular endothelial cells, and the direct up-regulation and activation of integrins on leukocytes) (15– 18). Upon complement activation, neutrophil infiltration can be caused by the potent chemotaxin C5a anaphylatoxin (C5a) and the C5b-9 complex, both of which increase P-selectin on endothelial cells and leukocyte integrins (19). To assess the role of TNF in neutrophil recruitment, TNF activity was blocked by the application of a saturating dose of 100 mg per mouse i.p. of the neutralizing rat anti-mouse TNF mAb V1q just before starting the Arthus reaction. This concentration was found to prevent a lethal shock of mice by 400 mg of LPS/mouse for .5 days (20). The neutralizing capacity of the V1q Ab, obtained from the peritoneal lavage fluid 6 h after initiation of the Arthus reaction, was proven in a L929 TNF assay as described previously (20). To define the contribution of complement, mice were treated with CVF. An i.p. challenge of BALB/c or C57BL/6 mice with ICs resulted in an accumulation of neutrophils in the peritoneal cavity (which are only) 1–2% in untreated animals) of 66.5% 6 11.6% or 65.3 6 8.1% at 8 h, corresponding to 5.1 6 0.9 neutrophils 3 106/mouse. As depicted in Fig. 3, TNF inhibition in BALB/c or C57BL/6 mice resulted in a slight decrease of the neutrophil count to 58.6% 6 8.9% or 52.6% 6 11.6%. However, CVF treatment resulted in completely different effects in BALB/c or C57BL/6 mice (Table I). In the latter strain, PMN margination was only marginally and insignificantly affected (49.9% 6 10.8%). Even after the inhibition of both complement and TNF, the neutrophil number did not change significantly (51.5% 6 11.2%). In contrast, CVF treatment strongly reduced neutrophil influx from 66.5% 6 11.6% to 9.9% 6 5.6% at 8 h after IC application in BALB/c mice (Fig. 3). Thus, the inflammatory response in BALB/c mice but not in C57BL/6 mice can be attributed predominantly to complement, with the most likely candidates being C5a or the C5b-9 complex. Using a specific C5aR antagonist, we were able to address this inhibition to C5a (our unpublished data). These data imply that the genetic background of the different mouse strains is an important factor affecting the inflammatory response in the peritoneal Arthus reaction. Neutrophil recruitment in IC peritonitis of C57BL/6 mice depends mainly upon the activation of FcgRI on peritoneal macrophages Because neither TNF nor complement activation were found to be the key players eliciting neutrophil recruitment in C57BL/6 mice, we tested whether gene-targeted disruption of FcRg chain and/or FcgRIII attenuates neutrophil recruitment (Table I). FcgRIII-deficient mice showed a slight but not significant reduction of neutrophil influx from 65.3% 6 8.1% to 55.5% 6 3.3% (Fig. 4), which was comparable with that seen in C57BL/6 mice in which TNF

5660

CUTTING EDGE

FIGURE 3. Inflammatory response in IC peritonitis is mediated by different pathways in BALB/c and C57BL/6 mice. CVF treatment markedly attenuated the neutrophil influx (9.9% 6 5.6%) in BALB/c mice. Application of the TNF Ab V1q had no effect. In C57BL/6 mice, treatment with CVF, with the TNF Ab, or with a combination of both did not have any significant effect on neutrophil influx. Data are expressed as the mean 6 SD (n 5 4 –11). p, p , 0.001, as calculated by the Student’s t test.

was blocked (Fig. 3). This result provides additional evidence that TNF does not mediate PMN migration in IC-mediated peritonitis in C57BL/6 mice, because the TNF release is mainly caused by the interaction of IC with FcgRIII (Fig. 2). Additional treatment of FcgRIII-deficient mice with CVF, however, reduced neutrophil accumulation significantly to 31.5% 6 3.5%. Surprisingly, neutrophil influx dropped to 11.0 6 7.5 in mice lacking FcRg chain and was completely abrogated after the depletion of complement in these animals (Fig. 4). These data strongly suggest that the activation of macrophages via FcgRI is the major mechanism that promotes the inflammatory response in Ab-mediated peritonitis in mice with a C57BL/6 background, because FcgRI is exclusively expressed on macrophages and neutrophils. In addition, a second pathway exists, which is complement-dependent. The interaction of ICs with FcgRIII on peritoneal mast cells and the subsequent release of TNF play only a minor role with respect to neutrophil recruitment in this particular model of IC disease. In conclusion, we have used FcRg chain- and FcgRIII-deficient mice in parallel to differentiate between FcgRI- and FcgRIII-mediated effects. Because no FcgRI-deficient animals have yet been generated and no neutralizing Abs against murine FcgRI are available, this is currently the only approach to evaluate the contribu-

tion of FcgRI to IC disease in vivo. Up to now, phagocytosis and Ab-dependent cellular cytotoxicity have been described as the main functional properties of FcgRI. In addition, its involvement in Ag presentation of IgG opsonized Ags has been discussed (21). However, the finding that the interaction of ICs with FcgRI is important in Ab-mediated tissue disease is novel and unexpected. The data obtained with FcRg chain or FcgRIII-deficient mice in the Arthus reaction model in the skin suggested that inflammatory sequels depend exclusively upon the interaction of ICs with FcgRIII on mast cells (4, 9). Our data, which are summarized in Table I, support a concept in which both complement and FcgRs play an important role in the initiation of the inflammatory cascade in IC-mediated diseases. In addition, our data provide evidence that the pathogenesis of the inflammatory response is tissue- and strain-specific and thus cannot be generalized as suggested previously (22). In BALB/c mice, the inflammatory reaction is predominantly mediated by complement. In contrast, FcgRI is crucial in C57BL/6 mice, whereas complement plays a synergistic role. TNF, although released in huge amounts into the peritoneal cavity through the interaction of immune complexes via FcgRIII, certainly does not contribute to neutrophil accumulation in BALB/c and C57BL/6 mice, as is the case in W/Wv mice.

Table I. Summary of the contributions of complement and FcRgs to neutrophil recruitment and TNF release into the peritoneal cavitya BALB/c Reduction of

Neutrophil margination TNF release

1C-depletion

111 2

FcRg chain2/2

C57BL/6

2 2

FcgRIII2/2

1 C-depletion

Ø C-depletion

1 C-depletion

Ø C-depletion

111 ND

11 111

1 ND

2 111

a 111, abrogation of neutrophil margination or TNF release; 11, strong significant reduction of neutrophil margination; 1, significant reduction of neutrophils; 2, no significant reduction of neutrophil margination or TNF release; 0, without complement depletion.

The Journal of Immunology

FIGURE 4. Neutrophil recruitment in IC peritonitis depends upon FcgRI on peritoneal macrophages and complement. Mice were sacrificed 8 h after IC application. Neutrophil influx was not significantly attenuated in FcgRIII-deficient mice (FcgRIII2/2) compared with C57BL/6 mice. Application of CVF to the FcgRIII-deficient mice significantly reduced neutrophil migration. FcRg chain-deficient mice (FcRg chain2/2) showed a significant reduction of neutrophil accumulation that was further diminished when CVF was applied. Data are expressed as the mean 6 SD (n 5 4 – 8). p/†/‡, p , 0.001, as determined by ANOVA. Pairwise comparisons were performed using Tukey’s test.

With respect to therapeutic implications, the data imply that both FcgRI and FcgRIII as well as complement have to be considered in therapeutic approaches aimed to attenuate tissue destruction in IC disease in humans.

Acknowledgments We thank D. Ma¨nnel for providing the TNF Ab V1q and for testing the TNF neutralizing capacity of the Ab in the cytotoxicity assay. We thank B. P. Morgan and R. S. Ames for discussion and comments. In addition, the continuous support of D. Bitter-Suermann is gratefully acknowledged.

References 1. Arthus, M. 1903. Injections re´pete´es de serum de cheval chez le lapin. C. R. Soc. Biol. 55:817. 2. Fearon, D. T., and W. W. Wong. 1983. Complement ligand-receptor interactions that mediate biological responses. Annu. Rev. Immunol. 1:243.

5661 3. Takai, T., M. Li, D. Sylvestre, R. Clynes, and J. V. Ravetch. 1994. FcRg chain deletion results in pleiotrophic effector cell defects. Cell 76:519. 4. Hazenbos, W. L., J. E. Gessner, F. M. Hofhuis, H. Kuipers, D. Meyer, I. A. Heijnen, R. E. Schmidt, M. Sandor, P. J. Capel, M. Daeron, J. G. van de Winkel, and J. S. Verbeek. 1996. Impaired IgG-dependent anaphylaxis and Arthus reaction in FcgRIII (CD16)-deficient mice. Immunity 5:181. 5. Ravetch, J. V., and J. P. Kinet. 1991. Fc receptors. Annu. Rev. Immunol. 9:457. 6. Takai, T., M. Ono, M. Hikida, H. Ohmori, and J. V. Ravetch. 1996. Augmented humoral and anaphylactic responses in FcgRII-deficient mice. Nature 379:346. 7. Sylvestre, D. L., and J. V. Ravetch. 1994. Fc receptors initiate the Arthus reaction: redefining the inflammatory cascade. Science 265:1095. 8. Sylvestre, D., R. Clynes, M. Ma, H. Warren, M. C. Carroll, and J. V. Ravetch. 1996. Immunoglobulin G-mediated inflammatory responses develop normally in complement-deficient mice. J. Exp. Med. 184:2385. 9. Sylvestre, D. L., and J. V. Ravetch. 1996. A dominant role for mast cell Fc receptors in the Arthus reaction. Immunity 5:387. 10. Ho¨pken, U. E., B. Lu, N. P. Gerard, and C. Gerard. 1997. Impaired inflammatory responses in the reverse arthus reaction through genetic deletion of the C5a receptor. J. Exp. Med. 186:749. 11. Bitter-Suermann, D., M. P. Dierich, W. Ko¨nig, and U. Hadding. 1972. Bypassactivation of the complement system starting with C3. Immunology 23:267. 12. van Dijk, H., P. M. Rademaker, and J. M. Willers. 1980. Estimation of classical pathway of mouse complement activity by use of sensitized rabbit erythrocytes. J. Immunol. Methods 39:257. 13. Zhang, Y., B. F. Ramos, and B. A. Jakschik. 1992. Neutrophil recruitment by tumor necrosis factor from mast cells in immune complex peritonitis. Science 258:1957. 14. Gordon, J. R., and S. J. Galli. 1990. Mast cells as a source of both preformed and immunologically inducible TNF-a/cachectin. Nature 346:274. 15. Zhang, Y., B. F. Ramos, and B. A. Jakschik. 1991. Augmentation of reverse arthus reaction by mast cells in mice. J. Clin. Invest. 88:841. 16. Ramos, B. F., Y. Zhang, and B. A. Jakschik. 1994. Neutrophil elicitation in the reverse passive Arthus reaction: complement-dependent and -independent mast cell involvement. J. Immunol. 152:1380. 17. Malaviya, R., T. Ikeda, E. Ross, and S. N. Abraham. 1996. Mast cell modulation of neutrophil influx and bacterial clearance at sites of infection through TNF-a. Nature 381:77. 18. Wershil, B. K., Z. S. Wang, J. R. Gordon, and S. J. Galli. 1991. Recruitment of neutrophils during IgE-dependent cutaneous late phase reactions in the mouse is mast cell-dependent: partial inhibition of the reaction with antiserum against tumor necrosis factor-a. J. Clin. Invest. 87:446. 19. Ko¨hl, J., and D. Bitter-Suermann. 1993. Anaphylatoxins. In Complement in Health and Disease, 2nd Ed. K. Whaley, M. Loos, and J. M. Weiler, eds. Kluwer Academic Publishers, Dordrecht, p. 295. 20. Echtenacher, B., W. Falk, D. N. Ma¨nnel, and P. H. Krammer. 1990. Requirement of endogenous tumor necrosis factor/cachectin for recovery from experimental peritonitis. J. Immunol. 145:3762. 21. Heijnen, I. A., M. J. van Vugt, N. A. Fanger, R. F. Graziano, T. P. de Wit, F. M. Hofhuis, P. M. Guyre, P. J. Capel, J. S. Verbeek, and J. G. van de Winkel. 1996. Antigen targeting to myeloid-specific human FcgRI/CD64 triggers enhanced antibody responses in transgenic mice. J. Clin. Invest. 97:331. 22. Ravetch, J. V., and R. A. Clynes. 1998. Divergent roles for Fc receptors and complement in vivo. Annu. Rev. Immunol. 16:421.