View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Immunity 366

Mellroth, P., Karlsson, J., and Steiner, H. (2003). J. Biol. Chem. 278, Xu, M., Wang, Z., and Locksley, R.M. (2004). Mol. Cell. Biol. 24, 7059–7064. 7949–7957. Stenbak, C.R., Ryu, J.H., Leulier, F., Pili-Floury, S., Parquet, C., Zaidman-Re´ my, A., Herve´ , M., Poidevin, M., Pili-Floury, S., Herve, M., Chaput, C., Boneca, I.G., Lee, W.J., Lemaitre, B., and Kim, M.-S., Blanot, D., Oh, B.-H., Ueda, R., Mengin-Lecreulx, Mengin-Lecreulx, D. (2004). J. Immunol. 173, 7339–7348. D., and Lemaitre, B. (2006). Immunity 24, this issue, 463– Tanji, T., and Ip, Y.T. (2005). Trends Immunol. 26, 193–198. 473.

Immunity 24, April 2006 ª2006 Elsevier Inc. DOI 10.1016/j.immuni.2006.03.008

Alveolar Macrophage injected bacteria before there is ‘‘spillover’’ of bacteria to DCs and before adaptive immunity is induced (Ma- in the Driver’s Seat cLean et al., 1996). Macrophages are specialized hematopoietic cells dis- tributed throughout different tissues of the body where Although alveolar macrophages are normally quies- they play a central role in homeostasis, tissue remodel- cent to prevent damaging the alveoli, in this issue ing, host defense, and the response to foreign materials, of Immunity, Takabayshi et al. (2006) demonstrate including particulates. One of the key functional charac- that alveolar macrophages can self-regulate their teristics of macrophages is that, depending on their function on demand to mount an appropriate immune state of differentiation and microenvironmental factors response. they encounter in a particular tissue, they can be specif- ically modified to have whatever functions needed to The function of the lung is to allow the uptake of oxygen deal most effectively with a particular inciting stimulus. and the excretion of carbon dioxide. Gas exchange oc- In the lung, resident alveolar macrophages are continu- curs in lung alveoli, which are made up of thin type I al- ously encountering inhaled substances due to their ex- veolar epithelial cells (AECs) and more cuboidal type II posed position in the alveolar lumen (Figure 1). To avoid AECs that produce surfactant and have self-renewal collateral damage to type I and type II cells in response and differentiation potential. Lung capillaries are situ- to harmless antigens, they are kept in a quiescent state, ated in close approximation to the type I cells, sepa- producing little inflammatory cytokines and displaying rated only by a 0.2 mm thick fused basement membrane, poor phagocytic activity, as evidenced by downregu- allowing the easy diffusion of gas. With its large surface lated expression of the phagocytic receptor CD11b area, the lung is exposed to many environmental chal- (Holt, 1978). In addition, alveolar macrophages actively lenges and is a portal of entry for many pathogens be- suppress the induction of adaptive immunity through cause the air we breathe is contaminated with infectious their effects on alveolar and interstitial DCs and T cells. agents, toxic gases, and (fine) particulate matter. The in- Elegant studies have demonstrated that in vivo elimina- haled microbes and inhaled toxic substances can then tion of alveolar macrophages using clodronate-filled li- gain easy access to the bloodstream across the delicate posomes leads to overt inflammatory reactions to other- alveolar-capillary membrane. Defense of this barrier is wise harmless particulate and soluble antigens (Thepen not easy and needs to be tightly controlled because et al., 1989). Alveolar macrophages adhere closely to too much edema, inflammation, and cellular recruitment AECs at the alveolar wall and are separated by a distance will lead to thickening of the alveolar wall and will jeop- of only 0.2–0.5 mm from interstitial DCs. In macrophage- ardize the diffusion of gases. Considering the large sur- depleted mice, the DCs have a clearly enhanced anti- face area of the respiratory epithelium and the volume of gen-presenting function (Holt et al., 1993). When mixed air inspired on a daily basis, it is remarkable that there is with DCs in vitro, alveolar macrophages suppress T cell so little inflammation under normal conditions. activation through release of nitric oxide (mainly in ro- Nonspecific and specific defense mechanisms pro- dents), prostaglandins, interleukin-10 (IL-10), and trans- tect the lung from environmental pathogens. Coughing forming growth factor-b (TGFb). and sneezing as well as the mucociliary blanket remove Until recently, it was largely unknown how the lung most of the larger particulates from the upper airways. environment instructs alveolar macrophages to sup- The innate immune system is very well developed in press innate and adaptive immunity. A few years ago, the deeper parts of the lung and is made up of a humoral an important step forward was made by the finding that arm (lactoferrins, lyzozyme, surfactant proteins, man- avb6 integrin-deficient mice have activated alveolar nose binding lectin, and defensins) and a cellular arm, macrophages due to a lack of TGFb activation in the consisting mainly of alveolar macrophages that express lung (Morris et al., 2003). This integrin has the potential numerous pattern recognition receptors for foreign anti- to activate latent TGFb by binding to the latency gen. If these nonspecific mechanisms fail, a highly de- associated peptide (LAP), an N-terminal inactivating veloped network of epithelial and alveolar dendritic cells fragment of TGFb. Takabayshi et al. in this issue of (DCs) is responsible for mounting the adaptive immune Immunity further build on these early findings and dem- response. It has been estimated that the pool of alveolar onstrate how the avb6 integrin-TGFb axis influences im- macrophages can handle up to 109 intratracheally mune homeostasis in the lung (Takabayashi et al., 2006). Previews 367

Figure 1. Alveolar Macrophages Push and Release the Brakes on Lung Immunity In baseline conditions (1), alveolar macro- phages closely adhere to alveolar epithelial cells, in this way inducing the expression of the avb6 integrin, a TGFb-activating integrin. Upon recognition of dangerous antigens by TLR triggering, macrophages detach from al- veolar epithelial cells, and avb6 integrin ex- pression is rapidly lost. In this way, macro- phages escape TGFb inhibition and display innate immune functions such as phagocyto- sis and secretion of proinflammatory cyto- kines (2). After a few days, activated T cells stimulate macrophage production of MMP- 9, which activates latent TGFb and puts the brake back on the alveolar macrophage, and eventually, homeostasis is restored (3).

They show that under homeostatic conditions alveolar To avoid collateral damage and to restore gas ex- macrophages closely adhere to AECs, and this in turn change as quickly as possible, there needs to be a rapid induces the expression of the integrin avb6 on AECs, mechanism to put the brakes back on macrophage acti- in a TGFb-dependent manner, thus leading to localized vation, and again, the macrophage is in the driver seat. activation of TGFb in the vicinity of the macrophage Takabayshi et al. (2006) show that a few days after deliv- (Figure 1). Binding of activated TGFb to its receptors ex- ery of the infectious threat activated lymphocytes se- pressed on macrophages induces phosphorylation of creting IFN-g stimulate the production of matrix metallo- SMAD-2 and -3 and suppresses macrophage phagocy- proteinase MMP-9 by macrophages (Figure 1). MMP-9 tosis and cytokine production. The inhibition of macro- has the potential to activate latent TGFb, and in this phage function by avb6-TGFb complex is unique to the way, tonic inhibition of macrophage function is restored lung, illustrating the microenvironmental specialization (as evidenced by the return of SMAD-2 and SMAD-3 of macrophages to meet the needs of the tissue. phosphorylation), macrophages again adhere to AECs, The tonic inhibition of macrophage function in the and avb6-integrin expression is restored. It has been lung is so robust that it has long been enigmatic how in- described previously by Holt et al. that freshly recruited fection might lead to macrophage activation. Takabay- monocytes gradually acquire the phenotype of resident shi et al. (2006) show that infectious agents known to suppressive alveolar macrophages over a period of days trigger Toll-like receptors (TLRs) and nonpattern recog- (Bilyk and Holt, 1993). Whether this would be a prede- nition receptors on macrophages open a window of op- fined process or an instruction by the lung TGFb-rich en- portunity for macrophage activation. TLR stimulation of vironment remains to be shown. Certainly, the prolonged macrophages leads to a rapid loss of contact with presence of activating cytokines such as GM-CSF keeps AECs, which in turn induces a rapid loss of expression inflammatory monocytes from acquiring suppressive of avb6 integrin expression on AECs (Figure 1). Under activities (Bilyk and Holt, 1993). An additional advantage these conditions, TGFb is no longer activated, releasing of enhanced TGFb production would be the stimulation the brakes on macrophage activation and innate im- of collagen synthesis in interstitial fibroblasts, necessary mune function, and macrophages become primed to for restoring alveolar wall architecture. secrete proinflammatory cytokines (tumor necrosis fac- The findings of Takabayshi et al. have implications for tor and IL-6) and to phagocyose particulate matter. understanding immunologically mediated lung dis- Once activated, alveolar macrophages can clear the in- eases. Lung fibrosis can not only occur after severe lung fectious threat on their own (Takabayashi et al., 2006). infections but also as an idiopathic disease (Gross and Others had already shown that many infectious agents Hunninghake, 2001). One consistent finding of these induce the recruitment of CCR2+ inflammatory mono- diseases is a loss of type I AEC and a concomitant cytes to the alveolar space (Warmington et al., 1999). hyperplasia of type II AECs. In fibrotic lung disease, These freshly recruited monocytes are clearly proin- the amounts of activated TGFb are strongly enhanced, flammatory and display phagocytosis and killing and yet alveolar macrophages are persistently activated, promote rather than suppress T cell and DC activation. producing too much TNF, IL-1, and IL-6. If type I AECs It takes a few days before these monocytes acquire the are lost, this might lead to a loss of anchoring points suppressive phenotype of alveolar macrophages, al- for alveolar macrophages and for avb6. In this way, lowing for another ‘‘window of opportunity’’ for initiation macrophages miss an important feedback mechanism of innate and adaptive responses in the lung (Bilyk and and produce their toxic products that inflict collateral Holt, 1995). damage. At the same time, persistent MMP-9 activation Immunity 368

and active TGFb might lead to prolonged activation of Selected Reading fibroblasts and fibrosis. Takabayshi et al. (2006) have provided important in- Bilyk, N., and Holt, P.G. (1993). J. Exp. Med. 177, 1773–1777. sights into how macrophage innate immune function is Bilyk, N., and Holt, P.G. (1995). Immunology 86, 231–237. closely regulated in the demanding environment of the Fainaru, O., Woolf, E., Lotem, J., Yarmus, M., Brenner, O., Golden- lung. In the future, it will be important to study if the berg, D., Negreanu, V., Bernstein, Y., Levanon, D., Jung, S., and Groner, Y. (2004). EMBO J. 23, 969–979. avb6 integrin-TGFb–macrophage axis also slows down Gross, T.J., and Hunninghake, G.W. (2001). N. Engl. J. Med. 345, adaptive immunity and its DC control in homeostatic 517–525. conditions. This scenario is certainly a possibility, as Holt, P.G. (1978). Am. Rev. Respir. Dis. 118, 791–793. mice deficient in Runx3, another downstream signaling Holt, P.G., Oliver, J., Bilyk, N., McMenamin, C., McMenamin, P.G., molecule involved in TGFb signaling, have constitutively Kraal, G., and Thepen, T. (1993). J. Exp. Med. 177, 397–407. activated alveolar DCs that mount immune responses to MacLean, J.A., Xia, W.J., Pinto, C.E., Zhao, L.H., Liu, H.W., and otherwise harmless environmental antigens (Fainaru Kradin, R.L. (1996). Am. J. Pathol. 148, 657–666. et al., 2004). Clearly, more research will be needed be- Morris, D.G., Huang, X., Kaminski, N., Wang, Y., Shapiro, S.D., fore we understand the full impact of this important Dolganov, G., Glick, A., and Sheppard, D. (2003). Nature 422, new pathway in lung immune homeostasis. 169–173. Takabayashi, K., Corr, M., Hayashi, T., Redecke, V., Beck, L., 1 Guiney, D., Sheppard, D., and Raz, E. (2006). Immunity 24, this issue, Bart N. Lambrecht 475–487. 1 Department of Pulmonary Medicine Thepen, T., Van Rooijen, N., and Kraal, G. (1989). J. Exp. Med. 170, Erasmus University Medical Center 494–509. Dr Molewaterplein 50 Warmington, K.S., Boring, L., Ruth, J.H., Sonstein, J., Hogaboam, 3015 GE Rotterdam C.M., Curtis, J.L., Kunkel, S.L., Charo, I.R., and Chensue, S.W. The Netherlands (1999). Am. J. Pathol. 154, 1407–1416.