Macrophage Stimulating Protein (MSP)

First printed in R&D Systems' 1998 Catalog.


The macrophage plays a variety of roles in the immune defense system, including a central role in innate or natural immunity. First described as a phagocyte by Metchnikoff in the 1800's1, 2, this cell is now known to exhibit a wide variety of functions, including phagocytosis, tumor cytotoxicity, cytokine secretion and (possibly) antigen presentation.3, 4 A number of factors are known that "activate" or engage macrophages in these activities. These include antibodies, chemotactic agents, membrane-bound and soluble receptors, and cytokines (particularly IFN-gamma).1-3, 5 In addition to IFN-gamma, one of the more intriguing molecules associated with macrophage phagocytic activity is MSP (macrophage stimulating protein).6 This hepatocyte growth factor-like protein may be particularly important to macrophage phagocytosis in its very earliest stages and, as such, may be positioned to significantly impact macrophage-influenced downstream adoptive and continuing innate immune responses.

Structural Information

Human MSP is an 85 kDa, disulfide-linked, heterodimeric glycoprotein that shares considerable homology with hepatocyte growth factor (HGF).7-10 MSP is synthesized initially as a prepropeptide. Following cleavage of the signal peptide, the 693 amino acid (aa) residue propeptide is proteolytically cleaved at an Arg-Val bond to generate two polypeptide chains that remain associated to form a heterodimeric protein. This protein consists of a 55 kDa alpha-chain of 465 aa and a 28 kDa beta-chain of 228 aa linked together by one disulfide bridge.7, 8 Structurally, the N-terminal alpha-chain contains four 80 aa kringle (or triple disulfide) loops, while the C-terminal beta-chain shows an (inactive) serine protease domain (SPD).7, 8, 11 As a result, MSP is now considered to be a member of the kringle domain-containing family of serine proteases.12 In addition to possessing kringle domains and SPDs, all family members circulate freely, are secreted as inactive precursors, and undergo cleavage (and thus activation) between the kringle domain(s) and the SPD.9, 12 All members except HGF and MSP also show serine protease activity, with the lack of enzyme activity in MSP attributed to three non-catalytic aa substitutions in the SPD.8 Mouse and rat MSP have also been isolated, with mouse MSP having 82% aa sequence identity to human MSP,13 and rat MSP having 80% aa sequence identity to human MSP. In a comparasin of rat to mouse proteins, MSP shows 93% aa sequence identity.9 Human MSP is known to be active in mice.14 MSP has been isolated from serum at nM concentrations,15, 16 but it appears to exist only in the uncleaved or proform.17, 18 Although a number of enzymes are known that can potentially cleave MSP,18, 19 it has recently been suggested that physiologically important MSP is only generated at the surface of macrophages (or other target cells), resulting in locally restricted levels of the cytokine.20 Cells known to produce MSP are limited and include hepatocytes,21 developing sperm,9 plus embryonic floor plate and myotome cells.12


The human receptor for MSP has been cloned and found to a 185 kDa, 1400 aa transmembrane protein that exhibits intrinsic tyrosine kinase activity.22, 23 Named RON for Receptur d'Origine Nantaise,22 the receptor shows significant homology to MET, the HGF receptor, and is classified as a type IV protein tyrosine kinase.24, 25 Members of this subfamily show a heterodimeric structure, where a small, strictly extracellular alpha-chain is covalently cross-linked to a large transmembrane signal transducing beta-chain.24 For RON, this structure begins with the synthesis of a 1400 aa prepropeptide followed by intracellular cleavage into an approximately 40 kDa alpha-chain of 280 aa and an approximately 150 kDa beta-chain of 1090 aa.22, 26 MSP has been reported to bind to RON with a Kd of 500 pM.27 Under normal physiological circumstances, for the standard receptor isoform, cleavage is believed to be necessary for expression, subsequent ligand binding, and receptor activation.26 However, an alternatively spliced form of human RON has also been identified that shows a 49 aa deletion in the beta-chain region. Although the proform of this shorter version is neither proteolytically cleaved nor expressed on the surface, this variant does appear to undergo spontaneous intracellular oligomerization and activation, suggesting a possible role for this spliced form in development of some tumors.23 Mouse RON (or STK for stem cell-derived tyrosine kinase) has also been cloned and found to be 74% identical to human RON at the aa sequence level.28, 29 As in the human, the mouse receptor has an alternatively spliced form. This variant is severely truncated, however and lacks most of the extracellular region of the receptor. Nevertheless, this shorter version is expressed at higher levels than the longer form.29 The significance of this form is unclear.

Cells known to express RON include macrophages,28 keratinocytes,22, 30 columnar epithelium,31 osteoclasts,32, 33 neutrophils,34 megakaryocytes,35 chromaffin cells of the adrenal medulla,31 respiratory ciliated columnar epithelium,29 and sperm.9

Biological Activity

Much of the activity attributed to MSP involves effects on macrophage phagocytosis. It was recognized early on that MSP increased phagocytosis of immunoglobulin-coated RBCs and was required for macrophages to respond to complement in a chemotactic fashion.6, 36 Subsequent studies have demonstrated that MSP does indeed influence the activity of both the C5a receptor and the C3b receptor (CR1)15, and that MSP (in vitro) promotes complement-driven phagocytosis.12 Once phagocytosis has occurred, MSP seems to block the generation of nitric oxide and related intermediates, molecules that are often considered central to microbicidal activity.14, 37 While this effect may be somewhat counterintuitive, NO is known to 1) induce macrophage apoptosis, 2) downregulate superoxide activity, and 3) inhibit cellular respiration, a function that is known to increase in response to phagocytosis. Thus, on balance, the inhibition of NO may actually facilitate the overall process of phagocytosis.15

Although MSP seems to impact some aspects of chemotaxis, it apparently does not provide a stimulus for monocyte migration out of the blood. Indeed, monocytes are not even known to express the MSP receptor RON.15, 28, 38 What MSP is suggested to do, however, is to induce the terminal differentiation of recently migrated and resident phagocytes into "mature" macrophages.28, 38 In order to respond to MSP though, local cells that were previously RON negative must now become RON positive, and the stimulus for the upregulation of RON is unknown.28 This local conversion from an unresponsive to a responsive phenotype is paralleled by the local conversion of MSP from an inactive to an active growth factor state. MSP is not considered to circulate in a bioactive form (i.e., as a disulfide-linked heterodimer). Rather, as mentioned above, bio-inactive proMSP is believed to be constitutively produced by hepatocytes, to circulate freely throughout the body, and to be activated by proteolytic cleavage only at the surface of target cells.9, 17, 18, 20, 21 Thus, the activity of target cells (macrophages) would appear to be the limiting factor in availability of bioactive MSP.

MSP also is suggested to play a role in embryogenesis, particularly in development of the myotome and induction of the floor plate.12 Finally, MSP is proposed to enhance megakaryocyte maturation, perhaps by stimulating IL-6 production.35


  1. Rabinovitch, M. (1995) Trends Cell Biol. 5:85.
  2. Langermans, J.A.M. et al. (1994) J. Immunol. Methods 174:185.
  3. Young, H.A. and K.J. Hardy (1995) J. Leukoc. Biol. 58:373.
  4. Peters, J.H. et al. (1996) Immunol. Today 17:273.
  5. Fearon, D.T. and R.M. Locksley (1996) Science 272:50.
  6. Leonard, E.J. and A.H. Skeel (1978) Exp. Cell Res. 114:117.
  7. Yoshimura, T. et al. (1993) J. Biol. Chem. 268:15461.
  8. Han, S. et al. (1991) Biochemistry 30:9768.
  9. Ohshiro, K. et al. (1996) Biochem. Biophys. Res. Commun. 227:273.
  10. Shimamoto, A. et al. (1993) FEBS Lett. 333:61.
  11. Wahl, R.C. et al. (1997) J. Biol. Chem. 272:15053.
  12. Thery, C. and C.D. Stern (1996) Acta Anat. 156:162.
  13. Degen, S.J.F. et al. (1991) Biochemistry 30:9781.
  14. Wang, M-H. et al. (1994) J. Biol. Chem. 269:14027.
  15. Skeel, A. et al. (1991) J. Exp. Med. 173:1227.
  16. Wang, M.H. et al. (1993) J. Leukoc. Biol. 54:289.
  17. Leonard, E.J. and A. Skeel (1996) J. Leukoc. Biol. 60:453.
  18. Wang, M-H. et al. (1994) J. Biol. Chem. 269:3436.
  19. Wang, M-H. et al. (1994) J. Biol. Chem. 269:13806.
  20. Wang, M-H. et al. (1996) J. Clin. Invest. 97:720.
  21. Bezerra, J.A. et al. (1993) Hepatology 18:394.
  22. Ronsin, C. et al. (1993) Oncogene 8:1195.
  23. Collesi, C. et al. (1996) Mol. Cell. Biol. 16:5518.
  24. Bardelli, A. et al. (1994) J. Biotechnol. 37:109.
  25. Park, M. et al. (1987) Proc. Natl. Acad. Sci. USA 84:6379.
  26. Wang, M-H. et al. (1994) Science 266:117.
  27. Sakamoto, O. et al. (1997) J. Clin. Invest. 99:701.
  28. Iwama, A. et al. (1995) Blood 86:3394.
  29. Iwama, A. et al. (1994) Blood 83:1360.
  30. Wang, M-H. and E.J. Leonard (1995) J. NIH Res. 7:47.
  31. Gaudino, G. et al. (1995) Oncogene 11:2627.
  32. Kurihara, N. et al. (1996) Blood 87:3704.
  33. Quantin, B. et al. (1995) Dev. Dyn. 204:383.
  34. Gaudino, G. et al. (1994) EMBO J. 13:3524.
  35. Banu, N. et al. (1996) J. Immunol. 156:2933.
  36. Leonard, E.J. and A.H. Skeel (1979) Adv. Exp. Med. Biol. 121B:181.
  37. MacMicking, J. et al. (1997) Annu. Rev. Immunol. 15:323.
  38. Skeel, A. and E.J. Leonard (1994) J. Immunol. 152:4618.