Thrombin binds to murine bone marrow-derived macrophages and enhances colony-stimulating factor-1-driven mitogenesis.

The binding and mitogenic properties of thrombin have been established in various transformed cell lines. In such systems, thrombin induces cell division in the absence of exogenous growth factors, and the enzyme is considered to act directly as a mitogen. This study explores thrombin's interaction with nontransformed, growth factor-dependent cells. Binding of 125I-alpha-thrombin to colony-stimulating factor (CSF)-1-dependent bone marrow-derived macrophages is saturable, time-dependent, and displaceable by both unlabeled alpha-thrombin, and esterolytically inactive thrombin. Both dissociation studies of pre-bound radio-labeled thrombin and Scatchard analysis assisted by the program "Ligand" suggest adherence of thrombin-binding data to a multi-site model. There are an estimated 2 x 10(4) high affinity sites (Kd = 7 x 10(-9)M) and 2 x 10(6) low affinity sites (Kd = 9 x 10(-7)M) per cell. Quiescent bone marrow-derived macrophages were cultured with either 10(-8)M thrombin, 1000 units of CSF-1/ml, or both and [3H]thymidine incorporation was determined. Thrombin alone did not induce mitogenesis. CSF-1 induced mitogenesis with peak [3H] thymidine incorporation occurring 24 h after addition of the mitogen. This CSF-1-dependent mitogenic influence was enhanced greater than 2-fold by treatment with thrombin.

Thrombin immobilized to extracellular matrix is a potent mitogen for vascular smooth muscle cells: nonenzymatic mode of action.

Esterolytically inactive diisopropyl fluorophosphate-conjugated thrombin (DIP-alpha-thrombin) stimulated 3H-thymidine incorporation and proliferation of growth-arrested vascular smooth muscle cells (SMCs), similar to native alpha-thrombin. Half-maximal mitogenic response of SMCs was obtained at 1 nM thrombin and was specifically blocked by the leech-derived, high-affinity thrombin inhibitor, hirudin. Native thrombin and a variety of thrombin species that were chemically modified to alter thrombin procoagulant or esterolytic functions were found to induce 3H-thymidine incorporation to a similar extent. Exposure of SMCs to DIP-alpha-thrombin caused a rapid and transient expression of the c-fos protooncogene, determined by Northern blot analysis. These results indicate that thrombin is a potent mitogen for SMCs through a distinct non-enzymatic domain. Binding of 125I-alpha-thrombin to SMC cultures revealed an apparent dissociation constant of 6 nM and an estimated 5.4 x 10(5) binding sites per cell. This binding was inhibited to the same extent by native thrombin and by its nonenzymatic form, DIP-alpha-thrombin. Moreover, the chemotactic fragment of thrombin (CB67-129), which failed to elicit a mitogenic response, competed for 125I-alpha-thrombin binding to SMCs. Cross-linking analysis of 125I-alpha-thrombin to SMCs revealed a specific cell-surface binding site 55 kDa in size. Finally, thrombin immobilized to a naturally produced extracellular matrix retained potent mitogenic activity toward SMCs. These observations lend support to the possibility that in vivo, subendothelial basement membranes sequester thrombin (as well as other bioactive molecules), which may stimulate localized and persistent growth of arterial SMCs. Thrombin may thus be involved directly in progression of atherosclerotic plaque formation.

Evidence for two functionally different fibrinogen receptors on hemopoietic cells: the glycoprotein IIb-IIIa and the mitogenic fibrinogen receptor.

We have previously established that the mitogenic effect of fibrinogen on hemopoietic cell lines Raji and JM is mediated via a specific receptor (Levesque, J.-P. et al.: Proc. Natl. Acad. Sci. USA 83:6494-6498, 1986). In this study, we have further characterized the fibrinogen domain involved in the binding to the mitogenic receptor. This binding was not inhibited either by a monoclonal antibody against the C-terminal sequence of the fibrinogen gamma chains or by synthetic peptides containing the Arg-Gly-Asp sequence. Such inhibition is specific of the platelet fibrinogen receptor, the glycoprotein IIb-IIIa complex. Fragments containing the fibrinogen D domain were the only plasmin degradation products of fibrinogen which were mitogenic. These fragments acted via direct binding on the mitogenic receptor with a Kd of 2.24 X 10(-6) M. This value was similar to the KI value of unlabeled fragments D (2.47 X 10(-6) M). Our results suggest the presence of two different functional types of fibrinogen receptors: the glycoprotein IIb-IIIa receptor responsible both for platelet aggregation and leukocyte adhesion and killing, and the mitogenic receptor involved in proliferation control of hemopoietic cells.

Thrombin chemotactic stimulation of HL-60 cells: studies on thrombin responsiveness as a function of differentiation.

Thrombin, a major procoagulant enzyme and growth factor, is also selectively chemotactic for monocytes and macrophages but not for neutrophils. This effect stands in contrast to other well-known chemotactic agents such as fMet-Leu-Phe, C5a fragments, and LTB4, which stimulate directed cell movement in both cell types, and have important physiological implications. The human leukemic cell line HL-60, which is capable of differentiating either along granulocytic or monocytic lineages, was therefore used to explore the development of this selective monocyte/macrophage chemotactic response to thrombin. Esterolytically inactive DIP-alpha-thrombin, as well as the thrombin-derived chemotactic peptide CB67-129, elicits a dose-dependent chemotactic response in HL-60 cells differentiated to monocytelike cells by treatment with 1,25(OH)2D3 (HL-60/mono), whereas no such response is evident in either undifferentiated HL-60 cells or in cells differentiated into granulocytes by treatment with DMSO (HL-60/gran). Similarly, early events which characterize stimulation of inflammatory cells by chemotactic agents are also evident, but only in monocyte-differentiated cells. In HL-60/mono, thrombin selectively stimulates rapid cytosolic Ca2+ elevation as well as rapid cytoskeletal association of cytosolic actin. Following thrombin stimulation, maximal actin association in these cells occurs within 30 sec (declining to basal levels at the end of 5 min), and maximal Ca2+ elevations are also evident within 15-20 sec, suggesting a temporal relationship between these two events. Thus, the events accompanying stimulation of HL-60/mono by thrombin are characteristic of those seen following stimulation of inflammatory cells by chemotaxins, with a major difference being the selectivity of thrombin as a chemotaxin for cells of macrophage/monocytic lineage. The selective chemotactic responsiveness of HL-60/mono to thrombin appears to relate to the development of specific receptors on these cells as part of monocytic differentiation: HL-60/mono (but HL-60/gran nor undifferentiated HL-60) are capable of significant specific 125-I-labeled alpha-thrombin-binding (ka approximately 20 nM), and possess an estimated 400,000 thrombin-binding sites per cell. Our findings further suggest that the thrombin response of HL-60 and particularly the expression of thrombin receptors on these cells may serve as a useful model system for exploring the biology of monocyte/macrophage differentiation.

Biologic activities of nonenzymatic thrombin: elucidation of a macrophage interactive domain.

Our studies to date have clearly demonstrated the existence of a unique cell interactive exosite region in thrombin, which mediates specific biologic effects on cells of monocytic-macrophage lineage. These findings are of potential physiologic significance, since even proteolytically degraded forms of thrombin (beta- and gamma-thrombin) or fragments of thrombin arising due to breakdown of thrombin-thrombomodulin complexes are potentially biologically active and are not subject to inhibition by inhibitors such as antithrombin III. Such degraded thrombin forms and proteolytic fragments are likely to be present at sites of inflammation and wound repair where they may be important modulators of macrophage-monocyte responsiveness. Importantly, as emphasized in this communication, the chemotactic and growth-promoting properties of this site, although overlapping, are clearly dissociable. It is apparent that the chemotactic effects require the presence of most, if not all, of the 338-400 sequence, whereas the mitogenic effects are mediated exclusively through the so-called loop B insertion sequence that is a unique feature of thrombin. Although there are ample precedents for differing functional domains within discrete proteins and the presence of hormonelike regions in proteins (such as fibronectin and immunoglobulin G) has been documented, it is fascinating to speculate how functionally unique regions such as this can arise. Data from Craik and associates suggest that unique amino acid sequences in proteins originate from gene insertions at intron-exon junctions and are characteristically surface expressed. These insertion sequences give rise to unique structural and functional features that characterize the particular protein, and also serve to differentiate it from other related members within its family.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification of a thrombin sequence with growth factor activity on macrophages.

In contrast to fibroblasts, the exposure of G0/G1-arrested J774 cells, a murine macrophage-like tumor cell line, with either active or esterolytically inactive diisopropyl phosphorofluoridate-conjugated alpha-thrombin (the enzymatically active form of thrombin, EC 3.4.21.5) results in a mitogenic response as measured by increased [3H]thymidine incorporation. This response to thrombin is optimal at 10 nM and is specifically blocked by hirudin, a high-affinity thrombin inhibitor. When prethrombin 1 [a single-chain prothrombin derivative lacking fragment 1, resulting from the action of thrombin on prothrombin] is cleaved with cyanogen bromide, a fragment (peptide CB67-129) is produced that, like the parent thrombin molecule, is mitogenic for J774 cells but not for fibroblasts. Limited tryptic digests of this fragment retain the ability to stimulate macrophages--a function that can be mimicked by a synthetic tetradecapeptide homologue of CB67-129 (representing residues 367-380 of the human thrombin B chain sequence) but not by any of a series of well-known growth promoters, including platelet-derived growth factor, epidermal growth factor, nerve growth factor, and fibroblast epidermal growth factor, nerve growth factor, and fibroblast growth factor. The mitogenic effects of this peptide are not limited to J774 cells but can be expressed in other macrophage-like tumor cell lines, including P388D1, RAW, and PU5. In addition to increased [3H]thymidine incorporation, the synthetic B chain peptide stimulates cell proliferation as evidenced by a dose-dependent increase in total protein per culture well and cell number. We conclude that the thrombin molecule contains a macrophage growth factor domain that is separate and distinct from its active center. Thus, thrombin, in addition to its major role in hemostasis and thrombosis, may also have important functions in such basic processes as the inflammatory response and monocytopoiesis.

Hormone-like activity of human thrombin.

Recently, we have shown that thrombin is a chemotaxin and growth-promoting agent for cells of the mononuclear phagocytic lineage. These activities are independent of thrombin's enzymatic activity. Unlike other chemotactic factors, thrombin is specific for monocytes and does not attract granulocytes. To further explore the cellular specificity we have used a human leukemia cell line HL-60 that is capable of in vitro differentiation toward either monocytes (HL-60/mono) following incubation with 1,25(OH)2D3, or granulocytes (HL-60/gran) following incubation with DMSO. In contrast to undifferentiated HL-60 cells or HL-60/gran, we find that HL-60/mono respond chemotactically to intact human alpha-thrombin, esterolytically inactive iPR2P-alpha-thrombin, and the thrombin-derived peptide CB67-129, previously shown to contain the thrombin chemotactic exosite. In addition, thrombin induces in HL-60/mono association of actin with the cytoskeleton and causes an increase in levels of free cytosolic Ca2+. These phenomena are well characterized as early events occurring concomitant with directed cell movement associated with exposure to chemotactic agents such as FMLP. Furthermore, in contrast to fibroblasts, both iPR2P-alpha-thrombin and the thrombin chemotactic peptide CB67-129 evoke dose-dependent [3H]TdR incorporation, protein synthesis, and cell replication in growth-arrested J-744 cells, a murine macrophage-like cell line. Limited tryptic digests of CB67-129 lose chemotactic activity but retain full mitogenic activity, demonstrating that as with PDGF, the sites on CB67-129 required for chemotaxis and mitogenesis are clearly dissociable. The mitogenic effects of the CB67-129 digest can be mimicked by a synthetic tetradecapeptide analogue of CB67-129 (residues 367-380) that includes the loop B insertion sequence, previously shown to be critical for thrombin's chemotactic effects. From these data, it is apparent that the loop B insertion is critical for thrombin's nonenzymic biological effects on cells, but additional sites are required for stimulation of cell movement.

Growth-promoting effects of esterolytically inactive thrombin on macrophages.

It has been recognized for many years that alpha-thrombin, like other better known mitogens (eg, PDGF, EGF, etc) is capable of initiating proliferation in quiescent cells belonging to the fibroblast family. However, unlike these other peptides, thrombin is a serine protease whose function as a growth stimulator for fibroblasts is intimately linked to its esterolytic activity. Thus, while native alpha-thrombin is capable of evoking DNA synthesis in G0/G1-arrested cells, neither enzymatically inactive thrombin (eg, iPR2P-alpha-thrombin) nor partially degraded thrombin (eg, gamma-thrombin) shares in this capability. Data from our laboratory have shown that thrombin is chemotactic for peripheral blood monocytes and for cells belonging to the monocyte/macrophage family and that this activity is not dependent upon thrombin's enzymatic properties. Our recent findings demonstrate that thrombin also serves as a growth factor for these cells, and this mitogenic capability is independent of esterolytic function and resides in the same region of the molecule as that responsible for chemotaxis. Additionally, by means of techniques such as computer modeling and peptide synthesis, we have now been able to delineate a distinct mitogenic subsite within this chemotactic thrombin sequence. Thus, the sequence in the thrombin B chain that mediates chemotaxis represents a true cell interactive exosite additionally capable of stimulating growth and possibly other biological functions in cells of macrophage/monocyte lineage.

Localization of a chemotactic domain in human thrombin.

The cyanogen bromide fragment CB67-129 of human prethrombin 1, corresponding to residues 54-116 of the thrombin B chain, is a potent chemotaxin for human peripheral blood monocytes and the murine macrophage like cell line, J774. Both of these cell types have been shown to respond chemotactically to alpha-thrombin and iPr2P-alpha-thrombin. Effective concentrations for stimulating directed cell movement with the fragment vary from 10(-11) to 10(-7) M. Moreover, CB67-129 and its parent protein compete for the same chemotactic receptor site. Fragment CB67-129, representing residues 54-116 of the human thrombin B chain sequence, contains a nine-residue insertion ("loop B") that is absent in homologous sequences derived from the closely related proteases chymotrypsin and trypsin. Unlike iPr2P-alpha-thrombin, iPr2P derivatives of these latter enzymes possess little or no chemotactic activity, suggesting a relationship between the insertion sequence and thrombin chemotactic activity. The loop B sequence is unique insofar as it contains all of the carbohydrate moieties known to reside in alpha-thrombin. However, chemotactic activity is only minimally reduced subsequent to hydrolysis by both neuraminidase and beta-galactosidase, indicating that receptor recognition and stimulated cell movement are mainly a function of structure of the cyanogen bromide derived fragment rather than of asparagine-linked carbohydrates.

Receptor-mediated chemotactic response of a macrophage cell line (J774) to thrombin.

Human alpha-thrombin, the procoagulant activation product of prothrombin, elicits chemotaxis in several macrophage-like continuous cell lines, most notably J774. Effective alpha-thrombin concentrations eliciting cell movement range from 10(-10) to 10(-6) M, with the optimal response occurring at about 10(-8) M. At the latter concentration, the response is equivalent, on a molar basis, to that observed with f-Met-Leu-Phe-OH, the positive control used in these experiments. Blockage of alpha-thrombin's active center with diisopropylfluorophosphate or by tryptic proteolysis of the procoagulant exosite (i.e., gamma-thrombin) does not decrease chemotactic activity. However, formation of enzymatically inactive complexes with AT3 or hirudin eliminates chemotactic potency. Competition chemotaxis assays, carried out by comparing the abilities of varying concentrations of test substances placed in the upper compartment of Boyden chambers to inhibit gradient-oriented movement of cells toward a fixed concentration of thrombin in the lower compartment, demonstrate that, although the modified forms of thrombin cross-compete, formylated peptide will not inhibit thrombin's chemotactic effects. Binding studies carried out using 125I-alpha-thrombin on paraformaldehyde-fixed J774 cells show a Kd of approximately 7.5 nM with an estimated 14,100-binding sites/cell. This Kd is in agreement with the optimal chemotactic thrombin dose (approximately 10 nM). On the basis of the competition chemotaxis assays and the radiolabeled thrombin-binding data, it is proposed that unique thrombin-specific chemotactic receptors exist on J774 cell membranes, and that these receptors are distinct from those mediating chemotaxis by agents such as formylated peptides.

Chemotactic response of monocytes to thrombin.

Human alpha-thrombin, the procoagulant activation product of prothrombin, elicits chemotaxis in human peripheral blood monocytes and several macrophagelike continuous cell lines, most notably J-774.2, but not in human peripheral blood granulocytes. alpha-Thrombin is effective in stimulating cell movement at concentrations ranging from 10(-10) to 10(-6) M but is optimally active at 10(-8) M. At the latter concentration, the degree of response is equivalent, on a molar basis, to that observed with the peptide formylmethionylleucylphenylalanine, (FMP). In contrast to thrombin, prothrombin produces a minimal chemotactic response in monocytes and J-774.2. Blockade of alpha-thrombin's active center with diisopropylfluorophosphate (DIP-F) or tryptic proteolysis of the procoagulant exosite (i.e., gamma-thrombin) fails to alter chemotactic activity. On the other hand, addition of equimolar amounts of antithrombin III (AT3) to alpha-thrombin reduces thrombin-mediated chemotaxis by 60%, and increased ratios of AT3 to enzyme completely suppress chemotaxis. We conclude that thrombin is a potent monocyte chemotaxin and that the domains in thrombin involved in stimulating cell movement are distinct from the catalytic site and the fibrin recognition exosite. These chemotactic domains appear to be sequestered in prothrombin and in the thrombin-AT3 complex and, as such, are unavailable to the chemotactic receptor on the monocyte cell membrane.

T lymphocyte recognition of peptide antigens: evidence favoring the formation of neoantigenic determinants.

To more precisely define the nature of exogenous antigenic determinants recognized by T cells, the response to fibrinopeptide fragment B beta 7-14 and a peptide of the inverted amino acid sequence of B beta 7-14 was examined. Strain 2 guinea pig T cells immunized with B beta 7-14 showed in vitro proliferative responses with B beta 7-14, but failed to respond to the inverted B beta 7-14 peptide. Moreover, the inverted B beta 7-14 peptide was nonimmunogenic and failed to prime strain 2 T cells for responses to native or inverted B beta 7-14. These results suggest that T cell recognition of peptide antigens involves more than simple interactions with amino acid side chains and that the ordering of the amino acids within the peptide is critical. One interpretation of these results is that T cells exhibit polarity in reading of antigenic determinants and peptides become associated with self in some fashion to form a neoantigenic determinant. To test this possibility, a Gly residue was added to the carboxyl end of B beta 7-14 (B beta 7-15), which is the likely site of attachment to self. It was found that strain 13 guinea pigs, which are totally unresponsive to B beta 7-14, produced T cell responses to B beta 7-15. This observation is consistent with the interpretation that Gly spaces the B beta 7-14 away from self to form an antigenic determinant complementary to strain 13 T cell antigen recognition structures. These results are discussed with respect to several mechanisms for immune response gene control of T cell responses.

Fine specificity of genetic regulation of guinea pig T lymphocyte responses to angiotensin II and related peptides.

Guinea pig T lymphocyte responses to the octapeptide antigen angiotensin II (NH(2)-Asp(1)-Arg(2)-Val(3)-Tyr(4)-Ile(5)-His(6)-Pro(7)-Phe(8)-OH; AII) were examined using various synthetic peptide analogues and homologues. Each peptide antigen was assessed for immunogenicity and antigenicity in strain 2 and strain 13 guinea pigs as determined by in vitro T cell proliferative responses. The genetic control of T cell responses to these peptides was found to be highly specific and capable of distinguishing subtle differences in the antigens. For example, strain 2 guinea pigs responded to AII and were low responders to [Val(5)]-AII, whereas strain 13 animals responded to [Val(5)]-AII but not to AII. The genetic control in this case involved the difference of one methyl group between Val(5) and Ile(5). Differences in T cell responsiveness by strain 2 and strain 13 guinea pigs were also observed with analogues involving para substitutions on the phenyl ring of Tyr(4) and of Phe(8). However, the genetic regulation of T cell responses did not seem to be based on a single peptide residue. For example, removal of Asp(1) allowed strain 13 animals to respond to the Ile(5)-containing analogue, but eliminated responsiveness to the Val(5)-containing analogue. Thus, the first and fifth AII residues are both involved in the regulation of strain 13 T cell responses. Substitutions for Tyr(4) and Phe(8) suggested that the same residue may serve to alter the specificity of T cell responses in one strain, and determine responsiveness or unresponsiveness in the other strain. One of the most striking observations is that T cell responsiveness to the various AII analogues and homologues randomly fluctuates between strain 2 and strain 13 guinea pigs, and in general neither strain responds to the same peptide antigens. This suggests that strain 2 and strain 13 T cell responses are rarely directed against the same antigenic determinants, and that the T cell antigen-combining diversity is usually exclusive between these two strains. These results are discussed with respect to the specificity of Ir gene control and the relationship between Ir gene function and antigen recognition by T cells. Note added in proof: More recent experiments using a new lot of [Val(5)]- AII have indicated that [Val(5)]-AII-immune strain 2 T cells show significant stimulation with AII but remain relatively low responders with [Val(5)]-AII, as shown in Table I. The difference in priming for cross-reactivity for AII with the different lots of [Val(5)]-AII is at present unknown.

Fibrin Polymerization. 1. Alkylating peptide inhibitors of fibrin polymerization.

A series of analogues relating to the NH2-terminal region of the fibrin alpha chain, i.e., Gly-Pro-Arg-Pro, were prepared by stepwise solid-phase synthesis, and their abilities to inhibit fibrin polymerization and to prolong thrombin-initiated clotting time were evaluated. Among the analogues systematically modified at different positions, replacement of the NH2-terminal three residues of Gly-Pro-Arg-Pro by either chlorambucil, p-nitrophenyl-L-alanine, or p-aminophenyl-L-alanine gave inactive compounds in the thrombin time assay, whereas similar substitution or extension of the COOH terminus produced the highly active analogues Gly-Pro-Arg-Phe(4-NH2), 22%; Gly-Pro-Arg-Pro-Phe(4-NO2), and Gly-Pro-Arg-Pro-Phe(4-NH2), 105%; relative to Gly-Pro-Arg-Pro = 100% in the fibrin polymerization inhibitory assay. As potential photoaffinity labeling probes, analogues containing a nitrophenylalanine residue in position 4 or 5 underwent photolysis under the experimental photoactivation conditions. As a potential alkylating probe, Chl-Pro-Arg-Pro was selectively effective in inhibiting thrombin amidolysis and fibrin polymerization. In the latter assay, Chl-Pro-Arg-Pro was approximately 20 times more potent than Gly-Pro-Arg-Pro in inhibiting fibrin aggregation.

Nature of T lymphocyte recognition of macrophage-associated antigens. V. Contribution of individual peptide residues of human fibrinopeptide B to T lymphocyte responses.

Guinea pig T lymphocyte responses to a decapeptide antigen (NH1-Asp5-Ans6-Glu7-Glu8-Gly9-Phe10-Phe11-Ser12-Ala13-Arg14-OH) of human fibrinopeptide B (hFPB) were examined using various synthetic peptide analogues containing single residue substitutions. Each analogue was examined for antigenicity as determined by an in vitro proliferative responses of hFPN-immune strain 2 guinae pig T cells. In addition, both strain 2 and strain 13 animals were immunized with each analogue and immunogenicity assessed by in vitro T cell-proliferative responses with the homologous immunizing analogue and the parent peptide. Replacement of arginine14 with lysine formed an immunogenic analogue which showed no antigenic cross-reactivity with the native peptide in strain 2 T cell responses. In addition, substitution of arginine14 with blocked lysine again produced a unique immunogenic analogue that showed little or no antigenic identity with the intact lysine analogue or the native peptide. In similar fashion, substitution of resideu phenylalanie10 with tyrosine or Phe(4-NO2) created unique immunogenic analogues with little or no antigenic identity to the native peptide with strain 2 T cells. By contrast, replacement of phenylalanine11 with either tyrosine or Phe(4-NO2) resulted in analogues with a total loss of immunogenicity and antigenicity in strain 2 T cell responses. An analogue in which glutamic acid7,8 were replaced with glutamine retained a small degree of antigenicity with hFPB-immune T cells, but T cells from strain 2 animals immunized with the Gln analogue responded only marginally to the Gln analogue while producing good proliferative responses with the native peptide. On the other hand, an analogue in which asparatic acid5 was replaced with asparagine retained most of the antigenic identity with hFPB for strain 2 T cell responses. None of thee analogues were immunogenic for strain 13 guinea pigs. These observations are discussed with respect to the contribution of each substituted residue to T cell respones, mechanism of Ir gene function, and a model for T cell recognition of small peptide antigens.