Annexin 1 and neutrophil apoptosis - Semantic Scholar

1 downloads 0 Views 68KB Size Report
*The William Harvey Research Institute, Queen Mary University of London, ..... 32 Alldridge, L.C., Harris, H.J., Plevin, R., Hannon, R. and Bryant, C.E. (1999).
Apoptosis in Myeloid Cells

Annexin 1 and neutrophil apoptosis M. Perretti*1 and E. Solito† *The William Harvey Research Institute, Queen Mary University of London, Bart’s and the London, Queen Mary School of Medicine and Dentistry, Charterhouse Square, London EC1M 6BQ, U.K., and †Molecular and Cellular Neuroscience, Imperial College, Hammersmith Campus, Du Cane Road, London W12 ONN, U.K.

Abstract ANXA1 (annexin 1), a member of the ‘annexin’ family of calcium- and phospholipid-binding proteins, was originally identified as an endogenous mediator of the anti-inflammatory actions of glucocorticoids. However, this protein exerts multiple inhibitory effects on the host inflammatory response, including a preferential regulation of the adhesion step of blood-borne neutrophil within the microenvironment of an inflamed vasculature. It is now emerging that ANXA1 is endowed with other roles, since the protein is abundant in inflammatory exudates as it is produced and released by the extravasated neutrophil. In the present paper, we review the novel proapoptotic effect of ANXA1 and discuss its potential with respect to the pathophysiology of inflammation and leucocyte recruitment.

Glucocorticoids and annexin 1 Since the seminal work of Hench et al. [1], GCs (glucocorticoids) have been recognized for their pivotal role in controlling the symptoms of several chronic inflammatory pathologies. In the early 1980s, Allan Munck et al. [2] reinterpreted the experimental results obtained with the use of natural and synthetic GCs, explaining how these molecules were important in preventing the host inflammatory reaction from over-shooting; thus these hormones are endogenously released to ensure that the inflammatory response decreases, i.e. to ensure that it does not damage the host. In a way, systemic administration of synthetic derivatives hijacks this hormonal function and uses it for therapeutic purposes. It is now clear that effects of GCs are observed in several distinct ways such that multiple intracellular and extracellular pathways/mechanisms can be influenced by their application. More than two decades ago, a novel mechanism was unravelled discovering a GC-inducible 37 kDa protein, termed lipocortin [3], capable of inhibiting PLA2 (phospholipase A2 ) and hence prostaglandin generation from perfused lungs and activated macrophages [4]. The protein was cloned in 1986 [5] and then shown to affect the acute inflammatory response [6]. In the latter study, lipocortin 1 or annexin 1 (ANXA1; current acronym for this mediator) was particularly effective in the carrageenin paw oedema model, in relation to the phase characterized by the accumulation of intense neutrophils. When recombinant preparations of the protein or peptido-mimetic drawn from the N-terminal region were tested, acute models of neutrophil extravasation insensitive to inhibitors of lipid metabolism, displayed a marked inhibitory effect [7,8]. These pharmacological studies opened the way to several others, such that today we refer to the ‘ANXA1 system’ as an endogenous biochemical process,

Key words: calcium flux, glucocorticoids, leucocyte, lipocortin 1, phagocytosis. Abbreviations used: ANXA1, annexin 1; GC, glucocorticoid; PLA2 , phospholipase A2 ; TNFα, tumour necrosis factor α. 1 To whom correspondence should be addressed (email [email protected]).

which operates in the context of the adherent extravasating neutrophil [9,10], briefly summarized below.

ANXA1 and the neutrophil In resting neutrophils, ANXA1 is predominantly localized in the cytosol, and a good proportion of this is localized in cytoplasmic granules [11], and possibly other organelles, such that it can be rapidly mobilized to the cell surface when the neutrophil adheres on to endothelium [12]. On the plasma membrane, ANXA1 interacts in an autocrine/paracrine fashion with receptors that belong to a specific family of ‘chemotactic’ G-protein-coupled receptors, most probably the lipoxin A4 receptor [10,13]. From other post-receptor mechanisms, the end-point is controlled neutrophil activation and detachment from the post-capillary venule endothelium [14,15]. The counter-regulatory action of ANXA1 is probably terminated by the activity of specific proteolytic enzymes [9,16]. The model illustrated above explains, at least in part, how endogenous ANXA1 can down-regulate neutrophil recruitment to the site of inflammation, and indicates the existence of a complex biological system which acts ‘selectively’ within the microenvironment of the adherent neutrophil. However, there are scant indications that extravasated neutrophils re-synthesize ANXA1 [17]. In addition, the causal role for extravasating neutrophils for marked exudate ANXA1 protein content was chiefly demonstrated in models of colitis [18] and then confirmed in the injured heart [19]. Why does the neutrophil make more ANXA1 once extravasated if the step sensitive to the protein (neutrophil adhesion/diapedesis) has already passed?

Annexin and apoptosis: the first studies When the word ‘annexin’ is mentioned in relation to apoptosis, most of the scientists would immediately think of annexin V only. This is because of the large use of this other member of the annexin superfamily [20] in apoptosis research,  C 2004

Biochemical Society

507

508

Biochemical Society Transactions (2004) Volume 32, part 3

where its ability to bind acid phospholipids in the presence of calcium is used to detect phosphatidylserine externalization at the surface of the cell entering in apoptosis [21]. However, all annexins could be used for this property, although annexin V is certainly more stable (in view of a very short N-terminal domain [20]). Annexin V is a useful tool and tells us little about a potential biological role for this annexin in the context of cell apoptosis. In any case, more studies are emerging to appreciate the role of endogenous ANXA1 in the process of apoptosis, not necessarily of the neutrophil, as well as the effect of exogenously applied ANXA1. These will be discussed in the remaining part of this mini-review. Immunohistochemical analysis of breast cancer tissue samples during the regression phase indicated a high degree of ANXA1 expression [22]. Tumour regression is due to marked apoptosis of the previously proliferating cells. It was only after a few years that this potentially causal role for ANXA1 in the context of cell apoptosis was revisited and began to be appreciated. A few studies have now described specific roles for this protein in the context of apoptosis, with the result of augmenting the complexity of the biochemical system centred on ANXA1. In 1996, Hirata and co-workers (who originally detected ANXA1 in circulating neutrophil [23]) reported the proapoptotic effect of this protein. Using rat thymocytes in culture, it was observed that exogenous ANXA1 addition to cells ‘reduced’ cell necrosis promoted by hydrogen peroxide, but ‘augmented’ cell apoptosis, probably via modulation of the activity of cytosolic PLA2 , and perhaps by the inhibition of the generation of anti-apoptotic prostanoids (e.g. PGE2 ) [24]. Congruently, an anti-ANXA1 monoclonal antibody enhanced H2 O2 -induced thymocyte necrosis and reduced H2 O2 -induced cell apoptosis. However, potential changes of cellular ANXA1 after cell exposure to the oxidant were not monitored. When we transfected U937 cells with fulllength ANXA1, cells entered spontaneously into apoptosis [25,26] and the clone died within a week [25]. Therefore to make the system more flexible and perform mechanistic studies, we stably transfected U937 cells with a plasmid encoding the N-terminal plus the first repeat of ANXA1. In this case, the cells survived transfection. Stable clones could be selected and they demonstrated a marked sensitivity to apoptosis, as observed after the addition of TNFα (tumour necrosis factor α) or the intercalating agent etoposide [26]. It is noteworthy that stable transfection of the antisense plasmid effectively reduced cell-associated protein expression, but this did not result in any detectable modification of cell survival. Thus the control operated by the intracellular ANXA1 system on apoptosis follows a one-way pattern, such that only higher ANXA1 expression promotes cell apoptosis (and reduces necrosis), whereas lower amounts of the protein are essentially ineffective. It is worth noticing that U937 cell incubation with TNFα augmented de novo ANXA1 synthesis and protein expression as early as 3–6 h of incubation [25]. Two further comments are now due. The first is to recall the protective effect of programmed cell death or apoptosis,  C 2004

Biochemical Society

such that the apoptotic cell, at variance from the necrotic one, implodes and does not release toxic mediators or proteolytic enzymes; thus apoptosis is safe for the host. The second is a reminder of the fact, partially mentioned above, that GC and also proinflammatory cytokines augment ANXA1 cell contents [27]; this suggests a potential relevance of this phenomenon at later phases of the inflammatory response (see below).

ANXA1 and neutrophil apoptosis What about the neutrophil? GC prolongs neutrophil survival; however, when these cells were incubated with ANXA1, an acceleration of cell apoptosis was quantified (using FITC–annexin 5 binding, cell-cycle analysis and staining of condensed nuclei) [28]. This effect was mediated by the ANXA1 receptor and was not shared by annexin 5. As said above, the latter annexin possesses a short (five amino acids) N-terminus, whereas the remainder of the protein (>300 amino acids) displays high homology with ANXA1 [20]. All this indicates a role for the N-terminal region of ANXA1 and indeed, a chimaeric annexin 5 protein with the 44-amino-acid-long ANXA1 N-terminus attached, fully retained the neutrophil proapoptotic property. ANXA1induced neutrophil apoptosis required entry of extracellular calcium and was unevenly accompanied by other classical neutrophil responses. In fact, ANXA1 and the chimaeric protein, but not annexin 5, caused transient entry of calcium fluxes, L-selectin shedding and a modest chemotactic effect, whereas no changes were detected in inositol phosphate metabolism and reactive oxygen species generation [28].

ANXA1-mediated apoptosis: clues for mechanisms? Although at a first glance ANXA1-induced neutrophil apoptosis is surprising in view of the anti-apoptotic effect of GC on this cell type, it is worth recalling that extravasated neutrophils markedly contribute to ANXA1 expression within inflammatory exudates and inflamed tissues, and this occurs in a GC-independent manner [29]. So, how does ANXA1 promote apoptosis? Very little information is available regarding the molecular mechanisms that might be activated by the protein. Also, the issue of extracellular versus intracellular ANXA1 should be carefully considered, as it is highly probable that different molecular determinants mediate ANXA1 roles in relation to its localization, hence probably in relation to the cell type and the phase of the inflammatory response. Having mentioned ANXA1-mediated inhibition of PLA2 activity in rat thymocytes, it was strange that the overexpression of U937 cells in ANXA1 fragment was associated with higher PLA2 activity [26]. In the latter cell type, increasing intracellular ANXA1 did not modify basal levels of Bax and Bcl-2 (B-cell lymphocytic-leukaemia proto-oncogene 2) [25], but increased the susceptibility of the cell clones to TNFα and etoposide-induced apoptosis. What changed was the activity

Apoptosis in Myeloid Cells

of one of the central enzymes in the apoptotic process, caspase 3. U937 clones transfected with full-length ANXA1 entered into apoptosis spontaneously, and this was paralleled by augmented caspase 3 activity [25]. The higher susceptibility to caspase 3 activation was also measured in the clones stably transfected with the ANXA1 fragment and stimulated with either TNFα or etoposide. A subsequent study has reinforced the link between caspase 3 and ANXA1. Monitoring of ANXA1 levels in a broncho-alveolar epithelial cell line (BZR cells) during ceramide-induced apoptosis demonstrated a parallelism between apoptosis and ANXA1 degradation [30], as detected by monitoring the appearance of the 33 kDa fragment cleaved at the N-terminus. Ceramide-induced BZR cell apoptosis was preceded by changes in Bcl-2 (decrease) and Bax (increase) gene expression, and this phenomenon was temporally followed by caspase 3 activation. Consistently, blockade of caspase 3 activity by means of a specific inhibitor prevented cell apoptosis and also ANXA1 proteolysis. This interesting study [30] poses a yet unanswered question: there are no caspase recognition motifs in the ANXA1 amino acid sequence, suggesting involvement of other proteolytic enzymes (e.g. calpain or elastase [31]), which in turn are activated by caspase 3. It is not clear whether changes in intracellular signal events reported in cells overexpressing ANXA1 would have an impact on apoptosis [32]. Furthermore, the proapototic effect of exogenously added ANXA1 to human neutrophils indicated a requirement for Ca2+ , and not L-selectin shedding, as well as dephosphorylation, and hence deactivation of the proapoptotic signal Bad. The putative role for caspase 3 activation in ANXA1-treated neutrophil remains untested. Also, no hypothesis is available as to which, if any, phosphatase might be activated by ANXA1 to deactivate Bad. In analogy with the discussion above for the intracellular protein, much more work is required for dissecting the molecular pathways activated by exogenous and endogenous ANXA1 to dictate cell entrance into apoptosis. These pathways are probably distinct, at least in their initial step, but they may well converge on common final executioners.

Does ANXA1 have more than one effect/role in apoptosis? Finally, a recent exciting study has highlighted another facet of the ANXA1/apoptosis link. Jurkat T lymphocyte apoptosis is associated with a Ca2+ - and caspase-dependent externalization of ANXA1 on to the plasma membrane [33]. The protein appeared to be intact (37 kDa) here, indicating that the cleavage observed by Debret et al. [30] could be cell-specific. The pattern of membrane ANXA1 expression in apoptotic Jurkat cells is reminiscent of that reported for adherent neutrophils [12] and thus may rely on fusion of cytosolic organelles with the plasma membrane. But why does ANXA1 move on to the cell surface of apoptotic T cells? Arur et al. [33] demonstrated a role for membrane ANXA1 as an ‘eat me’ signal. Membrane ANXA1 in Jurkat cells promoted clustering of phosphatidylserine receptors,

ensuring in this manner efficient engulfment of the apoptotic cells. It is noteworthy that GCs promote phagocytosis of apoptotic neutrophils in a non-phlogistic manner [34]: we do not know if apoptotic neutrophils externalize ANXA1 or if the large amount of soluble ANXA1 measured in inflammatory exudates could promote this phenomenon. However, preliminary results indicate ANXA1 externalization in apoptotic neutrophils (E. Solito, unpublished work).

Conclusion In the last 5 years, few papers have appeared linking ANXA1 and cell apoptotis. Whether determining apoptosis itself, engulfment of apoptotic cells or activation of proapoptotic pathways, these few studies have indicated a consensus for a positive effect of the protein on the process as a whole. This may have implications for the development of annexinomimetics as novel anti-inflammatory agents.

Research activities on ANXA1 were funded by the Arthritis Research Campaign UK (fellowship P0587) and the Wellcome Trust UK (programme grant no. 069234/Z/02/Z).

References 1 Hench, P.S., Kendall, E.C., Slocumb, C.H. and Polley, H.E. (1949) Proc. Staff Meet. Mayo Clin. 24, 181–197 2 Munck, A., Guyre, P.M. and Holbrook, N.J. (1984) Endocr. Rev. 5, 25–44 3 Di Rosa, M., Flower, R.J., Hirata, F., Parente, L. and Russo-Marie, F. (1984) Prostaglandins 28, 441–442 4 Flower, R.J. (1988) Br. J. Pharmacol. 94, 987–1015 5 Wallner, B.P., Mattaliano, R.J., Hession, C., Cate, R.L., Tizard, R., Sinclair, L.K., Foeller, C., Chow, E.P., Browning, J.L., Ramachandran, K.L. et al. (1986) Nature (London) 320, 77–81 6 Cirino, G., Peers, S.H., Flower, R.J., Browning, J.L. and Pepinsky, R.B. (1989) Proc. Natl. Acad. Sci. U.S.A. 86, 3428–3432 7 Perretti, M. and Flower, R.J. (1993) J. Immunol. 150, 992–999 8 Perretti, M., Ahluwalia, A., Harris, J.G., Goulding, N.J. and Flower, R.J. (1993) J. Immunol. 151, 4306–4314 9 Perretti, M. (1997) Trends Pharmacol. Sci. 18, 418–425 10 Perretti, M. (2003) Trends Pharmacol. Sci. 24, 574–579 11 Perretti, M., Christian, H., Wheller, S.K., Aiello, I., Mugridge, K.G., Morris, J.F., Flower, R.J. and Goulding, N.J. (2000) Cell Biol. Int. 24, 163–174 12 Perretti, M., Croxtall, J.D., Wheller, S.K., Goulding, N.J., Hannon, R. and Flower, R.J. (1996) Nat. Med. (N.Y.) 22, 1259–1262 13 Perretti, M., Chiang, N., La, M., Fierro, I.M., Marullo, S., Getting, S.J., Solito, E. and Serhan, C.N. (2002) Nat. Med. (N.Y.) 8, 1296–1302 14 Lim, L.H., Solito, E., Russo-Marie, F., Flower, R.J. and Perretti, M. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 14535–14539 15 Gavins, F.N., Yona, S., Kamal, A.M., Flower, R.J. and Perretti, M. (2003) Blood 101, 4140–4147 16 Smith, S.F., Tetley, T.D., Guz, A. and Flower, R.J. (1990) Environ. Health Perspect. 85, 135–144 17 Oliani, S.M., Paul-Clark, M.J., Christian, H.C., Flower, R.J. and Perretti, M. (2001) Am. J. Pathol. 158, 603–615 18 Vergnolle, N., Comera, ´ C. and Bueno, ´ L. (1995) Eur. J. Biochem. 232, 603–610 19 La, M., D’Amico, M., Bandiera, S., Di Filippo, C., Oliani, S.M., Gavins, F.N., Flower, R.J. and Perretti, M. (2001) FASEB J. 15, 2247–2256 20 Gerke, V. and Moss, S.E. (2002) Physiol. Rev. 82, 331–371 21 Koopman, G., Reutelingsperger, C.P.M., Kuijten, G.A.M., Keehnen, R.M.J., Pals, S.T. and van Oers, M.H.J. (1994) Blood 84, 1415–1420 22 McKanna, J.A. (1995) Anat. Rec. 242, 1–10 23 Hirata, F., Schiffman, E., Venkatasubramanian, K., Salomon, D. and Axelrod, J. (1980) Proc. Natl. Acad. Sci. U.S.A. 77, 2533–2536 24 Sakamoto, T., Repasky, W.T., Uchida, K., Hirata, A. and Hirata, F. (1996) Biochem. Biophys. Res. Commun. 220, 643–647  C 2004

Biochemical Society

509

510

Biochemical Society Transactions (2004) Volume 32, part 3

25 Solito, E., de Coupade, C., Canaider, S., Goulding, N.J. and Perretti, M. (2001) Br. J. Pharmacol. 133, 217–228 26 Canaider, S., Solito, E., de Coupade, C., Flower, R.J., Russo-Marie, F., Goulding, N.J. and Perretti, M. (2000) Life Sci. 66, L265–L270 27 de Coupade, C., Ajuebor, M.N., Russo-Marie, F., Perretti, M. and Solito, E. (2001) Am. J. Pathol. 159, 1435–1443 28 Solito, E., Kamal, A.M., Russo-Marie, F., Buckingham, J.C., Marullo, S. and Perretti, M. (2003) FASEB J. 17, 1544–1546 29 Vergnolle, N., Comera, ´ C., More, ´ J., Alvinerie, M. and Bueno, ´ L. (1997) Am. J. Physiol. 273, R623–R629 30 Debret, R., El Btaouri, H., Duca, L., Rahman, I., Radke, S., Haye, B., Sallenave, J.M. and Antonicelli, F. (2003) FEBS Lett. 546, 195–202

 C 2004

Biochemical Society

31 Ando, Y., Imamura, S., Hong, Y.-M., Owada, M.K., Kakunaga, T. and Kannagi, R. (1989) J. Biol. Chem. 264, 6948–6955 32 Alldridge, L.C., Harris, H.J., Plevin, R., Hannon, R. and Bryant, C.E. (1999) J. Biol. Chem. 274, 37620–37628 33 Arur, S., Uche, U.E., Rezaul, K., Fong, M., Scranton, V., Cowan, A.E., Mohler, W. and Han, D.K. (2003) Dev. Cell 4, 587–598 34 Liu, Y., Cousin, J.M., Hughes, J., Van Damme, J., Seckl, J.R., Haslett, C., Dransfield, I., Savill, J. and Rossi, A.G. (1999) J. Immunol. 162, 3639–3646

Received 20 November 2003