Elsevier

Immunology Letters

Volume 85, Issue 2, 22 January 2003, Pages 173-180
Immunology Letters

Macrophages in chronic type 2 inflammation have a novel phenotype characterized by the abundant expression of Ym1 and Fizz1 that can be partly replicated in vitro

https://doi.org/10.1016/S0165-2478(02)00225-0Get rights and content

Abstract

Using a murine model of nematode infection, we have discovered macrophages that display a novel phenotype that may be characteristic of macrophages in chronic type 2 inflammation. These nematode-elicited macrophages (NeMφ) are characterized by two unique features: the ability to actively suppress proliferation of a broad range of cell types and the high level expression of two novel macrophage genes, Ym1 and Fizz1. NeMφ also show some similarities with in vitro-derived ‘alternatively activated macrophages’ such as the downregulation of inflammatory cytokines. We therefore investigated how much of the phenotype discovered in vivo could be replicated by activation with Th2 cytokines in vitro. Fizz1 and Ym1 were upregulated by IL-4 and IL-13 in vitro but at a considerably lower level than in NeMφ. In vitro treatment with IL-4 could also partly replicate the ability of NeMφ to block cellular proliferation. As well as the quantitative differences in gene expression and suppressive phenotype, we also observed phenotypic differences in the cell morphology between macrophages activated in vivo and in vitro. Although this study illustrated that macrophages activated in chronic inflammation have distinct features that cannot be readily reproduced in vitro it also demonstrated that some features of the complex NeMφ phenotype can be replicated by treatment of cultured macrophages with Th2 cytokines. In future, we hope to use in vitro analysis to help define the pathways that lead to this distinctive in vivo macrophage phenotype.

Introduction

Macrophages display a wide variety of phenotypes depending on the cytokine environment, tissue localisation and the time point in the inflammatory process. In the event of an infection, the first population of macrophages recruited follow the well-defined classical activation pathway involving Toll receptor ligation and activation by Th1 cytokines such as IFNγ. Classically activated macrophages, in addition to their fundamental role in pathogen clearance, also secrete pro-inflammatory chemokines that further recruit other inflammatory effector cells to the site of infection [1]. Macrophages are also the key players in the resolution of inflammation [2] and under the influence of cytokines such as IL-4 and IL-10 as well as prostaglandins and glucocorticoids, macrophages mediate apoptotic cell uptake, anti-inflammatory processes and wound healing [3], [4], [5], [6], [7]. Importantly, when inflammation is not resolved macrophages remain present as key elements in the chronic inflammatory pathway. However, whether these macrophages function to reduce inflammation and host damage or whether they contribute to host pathology is often uncertain.

We have been interested in addressing this question, particularly in the context of Th2 mediated inflammation such as that found in helminth infection, asthma and atherosclerosis. By using a murine model for nematode infection, we are able to obtain a source of macrophages activated in vivo in chronic Th2 inflammatory conditions [8]. Following peritoneal implant of the filarial nematode Brugia malayi, mice develop a potent Th2 response to the parasite [9] and maintain a long-term stable cellular infiltrate in the peritoneal cavity of which macrophages are the predominant cell type [10]. Using a combination of functional assays and gene expression analysis we have demonstrated that these nematode-elicited macrophages (NeMφ) have a novel phenotype characterized by two striking features: (1) the ability to reversibly suppress the proliferation of T cells and a wide range of tumour cell lines [8] and (2) the dramatic over representation of two gene products, Ym1 and Fizz1, not previously associated with macrophage function [11].

Ym1, a member of the chitinase family, was originally described as an eosinophil chemotactic factor produced by CD8 lymphocytes [12]. A subsequent study has verified the eosinophil chemotactic ability of this molecule, both in vivo and in vitro [13]. We have found that Ym1 represents 10% of the genes expressed by F4/80 purified NeMφ and that the level of expression is 10 000-fold higher than resident peritoneal macrophages [10], [11]. In addition, one of the most striking features of Ym1 is its propensity to produce large crystals in the lungs of mice with chronic lung pathology [14]. A molecule produced in such enormous quantities is unlikely to function solely as a chemotactic factor. One possibility is that Ym1 is an effector molecule whose function may be to encapsulate chitin-bearing pathogens such as yeast, fungi or nematodes. Alternatively, Jin et al. [15] have suggested that it may function to interact with extracellular matrix components consistent with a role for Th2 driven macrophages in wound healing [5].

We found Fizz1 through analysis of a subtractive library designed to define IL-4 dependent macrophage genes where it represented over 35% of the subtracted genes and 2% of the total NeMφ mRNA [11]. Fizz1 was first identified in the lavage fluid of mice with experimentally induced asthma [16]. Subsequently, it has been recognized as a member of a family of secreted cysteine-rich molecules with similarity to resistin, a molecule produced by adipocytes which induces resistance to insulin [17]. Neither of these studies identified a role for macrophages or IL-4 in Fizz1 production. A role for Fizz1 in the inhibition of nerve growth factor (NGF) has been proposed [16] but its function in Th2 mediated settings such as allergy or helminth infection remains to be elucidated.

Because Ym1 and Fizz1 have been identified in several non-helminth settings and particularly in association with the lung, it suggests that the NeMφ phenotype we have described may be characteristic of macrophages associated with Th2 chronic inflammation and not restricted to either nematode infection or the peritoneal cavity. The phenotype of NeMφ, including suppression, downregulation of pro-inflammatory cytokines and the high level of expression of novel genes, is highly dependent on IL-4 in vivo [8], [11], [18]. We thus considered them to have an ‘alternative activation’ phenotype, a term coined by Gordon et al. to distinguish macrophages activated by IL-4 or IL-13 from classically activated macrophages [19], [20]. Alternatively activated macrophages (AAMφ) have subsequently been expanded to include the effects of IL-10 and glucocorticoids [21]. Until very recently, most studies of alternative activation have been carried out by in vitro treatment of cultured macrophages with Th2 cytokines and glucocorticoids [4], [19], [22], [23], [24]. NeMφ share some features of macrophages activated by Th2 cytokines in vitro such as the downregulation of inflammatory chemokines [11] and increased class II MHC expression (unpublished observation). We thus investigated whether any of the IL-4 dependent features of NeMφ we identified in vivo could be replicated in vitro. We demonstrate here that the NeMφ phenotype can be partly replicated in vitro, providing a powerful adjunct to in vivo studies for the elucidation of the function of these cells, which are central players in many chronic inflammatory situations.

Section snippets

Generation of macrophages

For all experiments, mice used were 6–8-week-old male C57BL/6 or BALB/c.

In vivo-derived NeMφ: B. malayi adult parasites were removed from the peritoneal cavity of infected jirds purchased from TRS Laboratories (Athens, GA) or maintained in house. Mice were surgically implanted intraperitoneally (i.p.) with six live adult female B. malayi. Three weeks later the mice were euthanized by cardiac puncture and peritoneal exudate cells (PEC) were harvested by thorough washing of the peritoneal cavity

Macrophage alternative activation marker Arginase1 is induced by in vitro treatment with IL-4 and in NeMφ

To date, the only well characterized marker of AAMφ is Arginase1, shown originally in vitro [26], [27]. The hepatic isoform of Arginase (Arginase1) is the counterpart of inducible nitric oxide synthase (iNOS), which mediates the classical pathway of inflammatory macrophages, catalysing the conversion of l-arginine to nitric oxide [28]. Arginase1 competes with iNOS and mediates the alternative metabolic pathway of l-arginine to l-ornithine and urea [29]. We have previously shown that NeMφ

Discussion

We have used NeMφ as a powerful model for the study of alternate activation of macrophages in vivo particularly in relation to macrophages activated in type 2 chronic inflammatory settings such as asthma or nematode infection. In this study we determined the relevance of using in vitro treatment with type 2 cytokines as a means to investigate the function of these cells. In vitro activation of thioglycollate-elicited and J774 macrophages with IL-4 showed induction of the alternative activation

Acknowledgements

This work was supported by the Wellcome Trust. We would like to thank Yvonne Harcus for exceptional technical support and Ian Dransfield for helpful discussion.

References (36)

  • M. Owhashi et al.

    J. Biol. Chem.

    (2000)
  • L. Guo et al.

    J. Biol. Chem.

    (2000)
  • H.M. Jin et al.

    Genomics

    (1998)
  • S. Goerdt et al.

    Immunity

    (1999)
  • D.C. Webb et al.

    J. Biol. Chem.

    (2001)
  • D.M. Underhill et al.

    Annu. Rev. Immunol.

    (2002)
  • C. Haslett et al.

    Philos. Trans. R. Soc. Lond. B. Biol. Sci.

    (1994)
  • K.M. Giles et al.

    Br. J. Haematol.

    (2000)
  • A. Gratchev et al.

    Scand. J. Immunol.

    (2001)
  • M. Hesse et al.

    J. Immunol.

    (2001)
  • Y. Liu et al.

    J. Immunol.

    (1999)
  • A.G. Rossi et al.

    J. Immunol.

    (1998)
  • P. Loke et al.

    Eur. J. Immunol.

    (2000)
  • R.A. Lawrence et al.

    J. Immunol.

    (1994)
  • F.H. Falcone et al.

    J. Immunol.

    (2001)
  • P. Loke et al.

    BMC Immunol.

    (2002)
  • M. Owhashi et al.

    Parasitol. Res.

    (1998)
  • I.N. Holcomb et al.

    EMBO J.

    (2000)
  • Cited by (0)

    View full text