Glycosulfopeptides modeled on P-selectin glycoprotein ligand-1 inhibit P-selectin-dependent leukocyte rolling in vivo

Transcription

1 The FASEB Journal express article /fj fje. Published online July 1, Glycosulfopeptides modeled on P-selectin glycoprotein ligand-1 inhibit P-selectin-dependent leukocyte rolling in vivo Anne E.R. Hicks,* Anne Leppänen, Richard D. Cummings, Rodger P. McEver, Paul G. Hellewell,* and Keith E. Norman* A.E.R.H. and A.L. contributed equally to this work *Cardiovascular Research Group, Division of Clinical Sciences (North), University of Sheffield, Sheffield, UK; Department of Biochemistry and Molecular Biology, University of Oklahoma, Health Sciences Centre, Oklahoma City, OK Cardiovascular Biology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK Corresponding author: Keith E. Norman, Cardiovascular Research Group, Clinical Sciences Centre, Northern General Hospital, Sheffield, S5 7AU, UK. ABSTRACT Leukocytic inflammation can be limited by inhibiting selectin-dependent leukocyte rolling. In spite of intensive efforts to develop small molecule selectin inhibitors with defined structureactivity profiles, inhibition of P-selectin-dependent leukocyte rolling in vivo by such a compound has yet to be described. We recently reported that glycosulfopeptides (GSP), modeled on the high affinity selectin ligand PSGL-1, inhibit leukocyte binding to P-selectin in vitro. Here, we have used intravital microscopy to investigate whether GSP can inhibit P-selectin-dependent leukocyte rolling in vivo. Surgical preparation of the mouse cremaster muscle for intravital microscopy induced P-selectin-dependent leukocyte rolling. Baseline rolling was recorded for 1 min followed by i.v. injection of GSP. 2-GSP-6 and 4-GSP-6 substantially reversed P-selectindependent leukocyte rolling, whereas control GSP, which are not fully glycosylated, did not. Inhibition of leukocyte rolling by 2- and 4-GSP-6 lasted 2 4 min. Clearance studies with 125 I- labeled 4-GSP-6 demonstrated rapid reduction in its circulating levels concurrent with accumulation in urine. These data represent the first demonstration that a precisely defined structure based on a natural P-selectin ligand can inhibit P-selectin-dependent leukocyte rolling in vivo. Key words: intravital microscopy inflammation adhesion molecules microcirculation venules L eukocyte recruitment to inflamed tissues is a highly ordered process initiated in many conditions by P-selectin-dependent leukocyte rolling (1 3). P-selectin also supports platelet-leukocyte and platelet-endothelial cell interaction (3) and has procoagulant activity via a mechanism that may involve platelet microparticles or a direct effect of soluble P- selectin (3, 4). Inhibiting selectins therefore holds great promise for the treatment of

2 inflammatory diseases. The selectin family of adhesion molecules has three functionally and structurally related members: E-selectin (expressed by endothelial cells), L-selectin (expressed by leukocytes), and P-selectin (expressed by endothelial cells and platelets) (1, 2). Preclinical investigations have convincingly implicated P-selectin in numerous inflammatory disorders, including ischemia-reperfusion injury (5) and atherosclerosis (6, 7). The realization that selectins recognize sialylated fucosylated glycans, prototypically represented by the tetrasaccharide sialyl Lewis x (sle x ) (8 14), fueled development of carbohydrate-based selectin inhibitors. Data from in vitro binding assays and from models of inflammation support the notion that relatively simple sle x -mimetic drugs inhibit all three selectins (15 20) and, as such, might be efficacious against inflammatory disease. In an intravital microscopy model, however, where leukocyte rolling is observed immediately before and after application of inhibitors, sle x and its close structural mimetics fail to influence P- or L-selectin-dependent rolling, and high doses are required to inhibit E-selectin-dependent rolling (21). This observation is consistent with the notion that sle x and related structures represent only one component of the macromolecular assemblies that represent true selectin ligands (1, 2, 22). The best-characterized selectin ligand is P-selectin glycoprotein ligand-1 (PSGL-1), a dimeric mucin present on all leukocytes (23, 24). Studies with antibodies (25) and with gene-targeted mice lacking PSGL-1 (26) demonstrate that PSGL-1 is the major ligand for P-selectin-dependent leukocyte rolling in the microcirculation. In addition, we recently demonstrated that recombinant PSGL-1 fused to human IgG (rpsgl-ig) could support rolling interactions of microspheres with E- and P-selectins in venules (27) and could competitively inhibit leukocyte rolling on E- and L- as well as P-selectin in vivo (28). With the dual aims of better understanding P-selectin-PSGL-1 interactions and of producing small-molecule P-selectin inhibitors, we generated a series of glycosulfopeptides (GSP) modeled on the mature N-terminus of human PSGL-1, which begins at amino acid residue 42 (29). One of these, glycosulfopeptide-6 (GSP-6), binds tightly to P-selectin in a manner that is Ca 2+ -, sialic acid-, fucose-, and sulfate-dependent. Binding of GSP-6 (K d ~350 nm) to human P-selectin compares favorably with that of PSGL-1 (K d ~300 nm) (29). GSP-6 inhibits neutrophil binding to immobilized P-selectin at < 5 µm, whereas 5 mm free sle x has only slight effects. Positioning of a core-2 based O-glycan containing sle x at Thr 57 (close to locations of potential tyrosine sulfation) is critical for high affinity binding of GSP-6 to human P-selectin. Furthermore, GSP-6 must be sulfated on all three tyrosines to bind optimally to human P-selectin. Equipped with precise knowledge of the structural requirements for optimal binding to human P- selectin, we set out to determine whether derivatives of GSP-6 could compete with cell-bound selectin ligands and modify leukocyte rolling in a physiological setting in mice. Here, we present data that glycosulfopeptides 2-GSP-6 and 4-GSP-6, modeled after the human PSGL-1 structure, competitively inhibit murine leukocyte rolling in vivo. MATERIALS AND METHODS Equilibrium gel filtration chromatograpy

3 Hummel-Dreyer equilibrium gel filtration experiments were conducted as previously described (29). A Sephadex G-100 column was equilibrated with 35 SO 3-4-GSP-6 (10,000 cpm/ml, specific activity 1700 cpm/pmol) in subphysiological buffer (20 mm MOPS, ph 7.5, 50 mm NaCl, 2 mm CaCl 2, 2 mm MgCl 2, and 0.02% NaN 3 ). Different amounts of soluble P-selectin were preincubated with buffer plus 35 SO 3-4-GSP-6 and then added to the column. Samples were eluted with buffer plus 35 SO 3-4-GSP-6, and 140 µl fractions were collected at a flow rate of 70 µl/min. Radioactivity was determined by liquid scintillation counting. Bound GSP and total soluble P- selectin concentrations were calculated from equilibrium gel filtration data by dividing the molar amounts of bound 4-GSP-6 and free soluble P-selectin by the peak volume of GSP-soluble P- selectin complex. Animals C57BL/6 mice were purchased from Harlan (Oxon, UK). Male mice weighing between 25 and 35 g were used in these experiments. All procedures were approved by the University of Sheffield ethics committee and by the Home Office Animals (Scientific Procedures) Act 1985 of the UK. Intravital microscopy The cremaster was prepared for intravital microscopy as previously described (27). In brief, mice were anaesthetized with a mixture of ketamine, xylazine, and atropine; cannulations of the trachea, jugular vein and carotid artery were performed; and the cremaster muscle was exposed and spread over a special viewing platform. Temperature was controlled using a thermistorregulated heating pad (PDTronics, Sheffield, UK), and the cremaster was superfused with thermocontrolled (36ºC) bicarbonate-buffered saline. Microscopic observations were made using an upright microscope (Nikon eclipse E600-FN, Nikon UK Ltd, UK) equipped with a water immersion objective (40 /0.80 W). Images were recorded using a CCD camera (Dage MTI DC-330, DAGE MTI, Michigan City, IN) onto svhs videocassettes. Venules (20 40 µm diameter) were selected and typically observed for the entire experimental period. Centerline blood flow velocity (V CL ) was measured in vessels of interest by using a commercially available velocimeter (Circusoft, Hockessin, DE). Vessels with V CL between 1 and 5 mm/s were selected for these studies. Control leukocyte rolling was recorded exactly 30 min after exposure of the cremaster muscle for intravital microscopy because leukocyte rolling at this time is almost exclusively P-selectindependent (30). GSP were injected at 31 min and their effects monitored for 10 min. As a positive control, P-selectin antibody RB40.34 (0.3 mg/kg, Pharmingen, Oxford, UK) was injected at the end of experiments to confirm that rolling was P-selectin-dependent. For comparison, separate mice received a blocking dose (25) of anti-psgl-1 antibody (2PH1, 4 mg/kg, Pharmingen, Oxford, UK). Blood flow velocities and circulating leukocyte concentrations were measured at key times (before and after treatments) during experiments.

4 Distribution and clearance of 4-GSP-6 4-GSP-6 was radioiodinated ( 125 I) using iodobeads according to the manufacturer s (Pierce, Rockford, IL) instructions, giving a specific activity of 5 mci/µmol. A mixture of 125 I-4-GSP-6 (1 µg) and unlabeled 4-GSP-6 were injected into mice by the jugular vein at a final dose of 4.3 µmol/kg. Blood samples (10 µl) were drawn 1, 2, 4, and 10 min after injection of material. Mice were then exsanguinated and urine drained from the bladder into a syringe. Bladder, kidneys, spleen, liver, heart, lungs, and brain were also collected. Samples were counted on an automatic gamma counter (Wallac 1470, EG&G, Berthold, Milton Keynes, UK), and counts per minute was used to calculate the percentage of injected material located in each of the studied fluids and organs. Total recovered urine and whole organs were counted along with samples of blood. The proportion of 4-GSP-6 remaining in the blood was calculated from sample counts assuming a total blood volume equivalent to 8% of body weight. RESULTS Small modifications to the peptide backbone do not alter affinity of GSP for human P- selectin in vitro Four peptides were synthesized in sufficient scale for comprehensive in vivo investigations (Fig. 1a). 2- and 4-GSP-6 contain all of the structural elements necessary for high affinity P-selectin binding, whereas 2- and 4-GSP-1, while containing 3 tyrosine sulfate residues, carry only N- acetylgalactosamine (GalNAc) at Thr 57 and do not bind P-selectin. The equilibrium binding affinity of 4-GSP-6 to human P-selectin at 50 mm NaCl was 71 ± 16 nm (Fig. 1b, which was comparable to the published (31) value for 2-GSP-6 (76 ± 11 nm)). Neither 2-GSP-1 nor 4-GSP- 1 bound detectably to P-selectin in equilibrium binding experiments; therefore, these GSP served as negative controls. 2-GSP-6 and 4-GSP-6 competitively inhibit P-selectin-dependent leukocyte rolling in vivo By mimicking the extreme N-terminus of mature PSGL-1, and carrying all of the structural elements required for high affinity binding to human P-selectin, 2-GSP-6 and 4-GSP-6 can potentially compete with cell-bound P-selectin ligands and inhibit P-selectin-dependent leukocyte rolling in vivo. We used intravital microscopy of the mouse cremaster muscle to investigate this potential. Surgical preparation of mice for intravital microscopy stimulated P- selectin-dependent rolling as previously described (30). Baseline rolling was observed 30 min after surgery, and GSP were injected at 31 min. Effects of GSP on the number and velocity of rolling cells were determined from recordings taken between 32 and 42 min after surgery. Both 2-GSP-6 and 4-GSP-6 reduced preexisting P-selectin-dependent leukocyte rolling, whereas 2- GSP-1 and 4-GSP-1 did not (Fig. 2a (A QuickTime movie demonstrating the effect of 2-GSP-6 on P-selectin-dependent rolling can be accessed with this article's Supplemental Data link.)). Effects of 2-GSP-6 and 4-GSP-6 could not be attributed to changes in blood flow or circulating leukocyte counts, because blood flow velocity remained stable throughout observation and systemic leukocyte counts increased slightly following treatment with either 2-GSP-6 or 4-GSP- 6 (Table 1). Interestingly, 2-GSP-6 inhibited P-selectin-dependent rolling to a greater extent than 4-GSP-6, although neither compound matched the complete inhibition given by P-selectin

5 blocking antibody. The magnitude of inhibition given by 4-GSP-6 was similar to that given by anti-psgl-1 antibody 2PH1. Others (32) have reported slightly superior effects of an alternative PSGL-1 antibody (4RA10) although some P-selectin-dependent rolling still remains. The effects of 4-GSP-6 were dose-dependent and were maximal at 4.3 µmol/kg (Fig. 2a). 2-GSP-6 was not synthesized in sufficient quantities for dose response studies. The effects of 2-GSP-6 and 4-GSP-6 were short-lived. Both these GSP caused significant inhibition of leukocyte rolling 1 2 min after injection at 4.3 µmol/kg, but this effect reversed within 2 3 min (Fig. 2b. A higher dose (12.9 µmol/kg) of 4-GSP-6 did not further inhibit rolling, but it did prolong the effect. In addition to reducing the number of cells rolling through a given vessel, selectin inhibitors can also increase the velocity of cells that continue to roll. Although 4- GSP-6 failed to reduce the proportion of leukocytes rolling when given at 1.43 µmol/kg, this dose of peptide did cause a significant increase in leukocyte rolling velocity 1 min after application (indicated by a shift to the right of the distribution; Fig. 3 left panel). This effect was reversed after 2 min. The increase in velocity caused by 4-GSP-6 at 4.3 µmol/kg was greater in magnitude but was similarly reversed after 2 min (Fig. 3, center panel). In contrast, application of 4-GSP-6 at 12.9 µmol/kg caused a more sustained increase in leukocyte rolling velocity that was still apparent after 10 min (Fig. 3, right panel). Because surgically induced rolling evolves from a purely P-selectin-dependent response initially to one that is dependent on both P- and L- selectins between 30 and 60 min (30), we did not study the duration of action of 4-GSP-6 beyond 10 min. Rapid clearance of 4-GSP-6 To inhibit rolling, selectin antagonists must remain intact at sufficient concentrations in the blood. We used 125 I-radiolabeled 4-GSP-6 to investigate the kinetics of its clearance from the circulation. A mixture of radiolabeled and unlabeled 4-GSP-6 was injected into the jugular vein giving a final 4-GSP-6 dose of 4.3 µmol/kg. Blood samples (10 µl) were drawn from the carotid artery at serial time points after application of material and counted in a gamma counter. More than 60% of injected 4-GSP-6 was removed from the blood within 1 min of injection (Fig 4a). Following an initial rapid fall in blood concentration, we observed a more gradual clearance between 2 and 10 min. After collection of the final blood sample, mice were rapidly killed and various organs and fluids harvested. Approximately 30% of injected 4-GSP-6 was detected in the urine within 10 min of its application at 4.3 µmol/kg (Fig 4b). Subsequent analysis by highperformance liquid chromatography demonstrated that the 4-GSP-6 recoverable from urine was intact (data not shown). Counting of various organs showed little evidence of preferential accumulation at sites other than the urine (Fig 4b). DISCUSSION The therapeutic potential of inhibiting leukocyte adhesion to inflamed endothelium has been rigorously investigated in animal models, with a number of strategies (anti-icam-1, anti-cd18, anti-selectin) progressing to clinical investigation. Clinical trials with antibodies targeting leukocyte adhesion have proven disappointing however. Inhibition of ICAM-1 (33, 34) or CD18 (35) fails to improve prognosis in conditions in which preclinical data promise convincing potential. There is also scant evidence for any of the large number of selectin inhibitors

6 developed to date providing clear clinical benefits. However, agents that effectively inhibit P- selectin may have an advantage over agents targeting ICAM-1 or CD18, because the former may also inhibit leukocyte-platelet and platelet-endothelial cell interactions. In addition, extensive intravital microscopy investigations lead us to believe that failure to actually inhibit leukocyte rolling in vivo may underlie the disappointing performance of earlier selectin inhibitors. (21, 28, 36). A P-selectin inhibitor with clearly defined biological activity may provide a strong therapeutic effect by inhibiting leukocyte-endothelial cell, leukocyte-platelet, and platelet-endothelial cell interaction. P-selectin inhibition may exert effects via inhibition of soluble or platelet microparticle expressed P-selectin. Most reported selectin inhibitors are relatively simple oligosaccharide derivatives of sle x or sialyl Lewis a (sle a ) (5, 37). One likely reason for the disappointing performance of such simple derivatives in clinical trials is that their effects in vitro are often extrapolated to predict a likely effect in vivo. Such extrapolations assume that molecules capable of inhibiting soluble selectin binding to surface-bound selectin ligands or binding of leukocytes to surfaces coated with selectins or selectin ligands under static conditions will also inhibit leukocyte rolling on those selectins in vivo. For many oligosaccharide derivatives, however, in vitro inhibition of selectin binding has not translated into in vivo inhibition of selectin-dependent leukocyte rolling (21, 36). Leukocyte rolling is supported by rapid formation of selectin-selectin ligand bonds at the front of a cell, coupled with detachment at the rear. With a constant requirement for new bond formation, leukocyte rolling is therefore sensitive to treatments that block the molecules involved in this response. In keeping with this model, application of antibodies that block the selectins or PSGL- 1 reverse existing leukocyte rolling in vivo (25, 30, 38). Charged polysaccharides such as fucoidin and dextran sulfate can also inhibit preexisting leukocyte rolling (39 42), presumably by binding to and blocking the selectins. Difficulties of large-scale synthesis and fears of immune reactions limit the use of antibodies for therapy, whereas a high possibility of nonspecific side effects limit the use of fucoidin and similar agents. Small molecules of defined structure that selectively bind to selectins with high affinity and prevent binding to ligands therefore represent attractive drug candidates. We (29, 31, 43) and others (44 46) have studied the relationship between structure and function of the selectin ligand PSGL-1 to enhance our understanding of its contribution to leukocyte recruitment in vivo and provide clues as to how this response can be modified for clinical therapy. These investigations revealed that P-selectin bound with high affinity to a short N- terminal region of PSGL-1 that required tyrosine sulfation and a specifically positioned core-2 O-glycan capped with sle x (31). Glycosulfopeptides modeled after this region of PSGL-1 were synthesized and demonstrated to bind to P-selectin with high affinity and to inhibit binding of neutrophils to immobilized P-selectin in a static adhesion assay. Using enzymatic methods as described (29), we synthesized sufficient quantities of two glycosulfopeptides (2-GSP-6 and 4-GSP-6) for in vivo investigation and found that these agents reversed preexisting, surgically induced leukocyte rolling. This observation is consistent with a model wherein soluble selectin-binding molecules compete with cell-bound ligands preventing

7 the formation of new bonds required for continued maintenance of leukocyte rolling. Because surgically induced rolling is P-selectin-dependent, these data demonstrate that 2- and 4-GSP-6 are active P-selectin antagonists in vivo. Studies with 125 I-labeled 4-GSP-6 demonstrated that glycosulfopeptides are rapidly cleared from the circulation. Only 20% of injected material remained in the blood 2 min after intravenous injection. Our first measurement of rolling flux was made 1-2 min after injection. Assuming that the blood volume of a mouse is 8% of body weight, the blood concentration of 4-GSP-6 2 min after injection of 4.3 µmol/kg would be 10 µm. Because some of this material may be bound to blood elements, the active circulating concentration of 4-GSP-6 is comparable to the concentration of GSP-6 required to inhibit neutrophil binding to P-selectin in vitro (4.7 µm) (29). Our clearance studies (blood sampling for 10 min followed by sacrifice and organ harvest) were designed to track GSP removal from the blood rather than accumulation in specific tissues. 4- GSP-6 rapidly disappeared from the blood and a significant portion of material was cleared to the urine within 10 min. Because we saw no preferential accumulation in other organs, we assume that the remaining material is distributed fairly evenly throughout the body. The fact that high-dose (12.9 µmol/kg) 4-GSP-6 caused a sustained increase of leukocyte rolling velocity suggests that material cleared from the blood and into tissues slowly returns to the blood and is cleared to the urine more gradually. Modifications to prolong the circulating half-life of GSPbased agents may be necessary to enhance their therapeutic effects. We recently demonstrated that a less defined structure (recombinant PSGL-1 fused to IgG, rpsgl-ig) could also competitively reverse existing P-selectin-dependent leukocyte rolling (28). The molar inhibitory activity of GSP compares favorably with that of rpsgl-ig in that 4.3 µmol/kg of 4-GSP-6 (equating to 15 mg/kg) inhibited leukocyte rolling by 50 70%, whereas 30 mg/kg rpsgl-ig was required for a similar effect. The activity of the GSP is all the more remarkable when clearance kinetics are considered (rpsgl-ig has a half-life of hundreds of hours [47]). Activity of the GSP also compares favorably with less selective inhibitors such as fucoidin (41). In recent investigations, we have discovered a large discrepancy between the antiinflammatory effects of selectin inhibitors and their ability to directly influence leukocyte rolling. This discrepancy is typified by the activity of rpsgl-ig, which reduces inflammation and leukocyte rolling if given as a pretreatment at 1 mg/kg but can only reverse existing leukocyte rolling if doses are increased 30-fold (28). As discussed previously, selectins mediate events other than leukocyte rolling, including formation of mixed aggregates of leukocytes and platelets and transduction of signals into leukocytes (48). Thus, the antiinflammatory activity of selectin antagonists (perhaps including GSP) might depend in part on these other events, which may have distinct dose-response characteristics. Future investigations will examine the ability of glycosulfopeptides to modulate responses other than leukocyte rolling and to provide antiinflammatory protection. We will also investigate effects of GSP on E- and L-selectindependent rolling and consider strategies to increase the activity of GSP and prolong their biological availability.

8 In summary, we have demonstrated that GSP modeled on the N-terminus of mature PSGL-1 can competitively inhibit surgically induced leukocyte rolling in vivo. This represents the first demonstration that a precisely defined structure based on a natural P-selectin ligand can inhibit P-selectin-dependent leukocyte rolling in living blood vessels. The GSP described here will form the prototypes for development of more active and more stable analogues. ACKNOWLEDGMENTS This work was supported by grant FS98051 from the British Heart Foundation, grants and from the Wellcome Trust, and National Institute of Health grants HL 65631, AI 44902, and AI REFERENCES 1. Kansas, G. S. (1996) Selectins and their ligands: Current concepts and controversies. Blood 88, McEver, R. P., Moore, K. L., and Cummings, R. D. (1995) Leukocyte trafficking mediated by selectin-carbohydrate interactions. Journal of Biological Chemistry 270, McEver, R. P. (2001) Adhesive interactions of leukocytes, platelets and the vessel wall during hemostasis and inflammation. Thrombosis & Haemostasis 86, Hartwell, D. W., Mayadas, T. N., Berger, G., Frenette, P. S., Rayburn, H., Hynes, R. O., and Wagner, D. D. (1998) Role of P-selectin cytoplasmic domain in granular targeting in vivo and in early inflammatory responses. J Cell Biol 143, Lefer, D. J. (2000) Pharmacology of selectin inhibitors in ischemia/reperfusion states. Annu Rev Pharmacol Toxicol 40, Eriksson, E. E., Xie, X., Werr, J., Thoren, P., and Lindbom, L. (2001) Direct viewing of atherosclerosis in vivo: plaque invasion by leukocytes is initiated by the endothelial selectins. FASEB Journal 15, Johnson, R. C., Chapman, S. M., Dong, Z. M., Ordovas, J. M., Mayadas, T. N., Herz, J., Hynes, R. O., Schaefer, E. J., and Wagner, D. D. (1997) Absence of P-selectin delays fatty streak formation in mice. Journal of Clinical Investigation 99, Berg, E. L., Robinson, M. K., Mansson, O., Butcher, E. C., and Magnani, J. L. (1991) A carbohydrate domain common to both sialyl Le(a) and sialyl Le(X) is recognized by the endothelial cell leukocyte adhesion molecule ELAM-1. Journal of Biological Chemistry 266,

9 9. Phillips, M. L., Nudelman, E., Gaeta, F. C., Perez, M., Singhal, A. K., Hakomori, S., and Paulson, J. C. (1990) ELAM-1 mediates cell adhesion by recognition of a carbohydrate ligand, sialyl-lex. Science 250, Polley, M. J., Phillips, M. L., Wayner, E., Nudelman, E., Singhal, A. K., Hakomori, S., and Paulson, J. C. (1991) CD62 and endothelial cell-leukocyte adhesion molecule 1 (ELAM-1) recognize the same carbohydrate ligand, sialyl-lewis x. Proceedings of the National Academy of Sciences (USA) 88, Foxall, C., Watson, S. R., Dowbenko, D., Fennie, C., Lasky, L. A., Kiso, M., Hasegawa, A., Asa, D., and Brandley, B. K. (1992) The three members of the selectin receptor family recognize a common carbohydrate epitope, the sialyl lewisx oligosaccharide. Journal of Cell Biology 117, Imai, Y., Lasky, L. A., and Rosen, S. D. (1992) Further characterization of the interaction between L-selectin and its endothelial ligands. Glycobiology 2, Lowe, J. B., Stoolman, L. M., Nair, R. P., Larsen, R. D., Berhend, T. L., and Marks, R. M. (1990) ELAM-1-dependent cell adhesion to vascular endothelium determined by a transfected human fucosyltransferase cdna. Cell 63, Zhou, Q., Moore, K. L., Smith, D. F., Varki, A., McEver, R. P., and Cummings, R. D. (1991) The selectin GMP-140 binds to sialylated, fucosylated lactosaminoglycans on both myeloid and nonmyeloid cells. Journal of Cell Biology 115, Lefer, D. J., Flynn, D. M., Phillips, M. L., Ratcliffe, M., and Buda, A. J. (1994) A novel sialyl lewis x analog attenuates neutrophil accumulation and myocardial necrosis after ischemia and reperfusion. Circulation 90, Murohara, T., Margiotta, J., Phillips, L. M., Paulson, J. C., Defrees, S., Zalipsky, S., Guo, L. S., and Lefer, A. M. (1995) Cardioprotection by liposome-conjugated sialyl Lewisxoliogsaccharide in myocardial ischaemia and reperfusion injury. Cardiovascular Research 30, Kim, M. K., Brandley, B. K., Anderson, M. B., and Bochner, B. S. (1998) Antagonism of selectin-dependent adhesion of human eosinophils and neutrophils by glycomimetics and oligosaccharide compounds. Am J Respir Cell Mol Biol 19, Eguchi, H., Ikeda, H., Murohara, T., Yasukawa, H., Haramaki, N., Sakisaka, S., and Imaizumi, T. (1999) Endothelial injuries of coronary arteries distal to thrombotic sites: role of adhesive interaction between endothelial P-selectin and leukocyte sialyl LewisX. Circ Res 84, Kanwar, S., Johnston, B., and Kubes, P. (1995) Leukotriene C 4 /D 4 induces P-selectin and sialyl Lewis(x)- dependent alterations in leukocyte kinetics in vivo. Circulation Research 77,

10 20. Mulligan, M. S., Paulson, J. C., De Frees, S., Li, Z., Lowe, J. B., and Ward, P. A. (1993) Protective effects of oligosaccharides in P-selectin-dependent lung injury. Nature 364, Norman, K. E., Anderson, G. P., Kolb, H., Ley, K., and Ernst, B. E. (1998) Sialyl Lewis x (sle x ) and a sle x mimetic, CGP69669A, disrupt E-selectin-dependent rolling in vivo. Blood 91, Cummings, R. D., and Lowe, J. B. (1999) Selectins. In Essentials of Glycobiology (Varki, A., Cummings, R. D., Esko, J., Freeze, H., Hart, G., and Marth, J., eds) pp , Coldspring Harbor Laboratory Press, Boston 23. McEver, R. P., and Cummings, R. D. (1997) Role of PSGL-1 binding to selectins in leukocyte recruitment. Journal of Clinical Investigation 100, Yang, J., Furie, B. C., and Furie, B. (1999) The biology of P-selectin glycoprotein ligand- 1: its role as a selectin counterreceptor in leukocyte-endothelial and leukocyte-platelet interaction. Thromb Haemost 81, Borges, E., Eytner, R., Moll, T., Steegmaier, M., Campbell, M. A., Ley, K., Mossmann, H., and Vestweber, D. (1997) The P-selectin glycoprotein ligand-1 is important for recruitment of neutrophils into inflamed mouse peritoneum. Blood 90, Yang, J., Hirata, T., Croce, K., Merrill-Skoloff, G., Tchernychev, B., Williams, E., Flaumenhaft, R., Furie, B. C., and Furie, B. (1999) Targeted gene disruption demonstrates that P-selectin glycoprotein ligand 1 (PSGL-1) is required for P-selectinmediated but not E- selectin-mediated neutrophil rolling and migration. J Exp Med 190, Norman, K. E., Katopodis, A. G., Thoma, G., Kolbinger, F., Hicks, A. E., Cotter, M. J., Pockley, A. G., and Hellewell, P. G. (2000) P-selectin glycoprotein ligand-1 supports rolling on E- and P-selectin in vivo. Blood 96, Hicks, A. E. R., Hellewell, P. G., and Norman, K. E. (2001) The anti-inflammatory effects of rspsgl-1 are not related to competitive inhibition of selectin dependent leukocyte rolling in vivo (Abstract). FASEB Journal 15, A Leppanen, A., Mehta, P., Ouyang, Y. B., Ju, T., Helin, J., Moore, K. L., van Die, I., Canfield, W. M., McEver, R. P., and Cummings, R. D. (1999) A novel glycosulfopeptide binds to P-selectin and inhibits leukocyte adhesion to P-selectin. J Biol Chem 274, Ley, K., Bullard, D. C., Arbones, M. L., Bosse, R., Vestweber, D., Tedder, T. F., and Beaudet, A. L. (1995) Sequential contribution of L- and P-selectin to leukocyte rolling in vivo. Journal of Experimental Medicine 181,

11 31. Leppanen, A., White, S. P., Helin, J., McEver, R. P., and Cummings, R. D. (2000) Binding of glycosulfopeptides to P-selectin requires stereospecific contributions of individual tyrosine sulfate and sugar residues. J Biol Chem 275, Sperandio, M., Thatte, A., Foy, D., Ellies, L. G., Marth, J. D., and Ley, K. (2001) Severe impairment of leukocyte rolling in venules of core 2 glucosaminyltransferase-deficient mice. Blood 97, Enlimomab-Investigators (2001) Use of anti-icam-1 therapy in ischemic stroke: Results of the Enlimomab Acute Stroke Trial. Neurology 57, Salmela, K., Wramner, L., Ekberg, H., Hauser, I., Bentdal, O., Lins, L. E., Isoniemi, H., Backman, L., Persson, N., Neumayer, H. H., Jorgensen, P. F., Spieker, C., Hendry, B., Nicholls, A., Kirste, G., and Hasche, G. (1999) A randomized multicenter trial of the anti-icam-1 monoclonal antibody (enlimomab) for the prevention of acute rejection and delayed onset of graft function in cadaveric renal transplantation: a report of the European Anti-ICAM-1 Renal Transplant Study Group. Transplantation 67, Baran, K. W., Nguyen, M., McKendall, G. R., Lambrew, C. T., Dykstra, G., Palmeri, S. T., Gibbons, R. J., Borzak, S., Sobel, B. E., Gourlay, S. G., Rundle, A. C., Gibson, C. M., and Barron, H. V. (2001) Double-Blind, Randomized Trial of an Anti-CD18 Antibody in Conjunction With Recombinant Tissue Plasminogen Activator for Acute Myocardial Infarction: Limitation of Myocardial Infarction Following Thrombolysis in Acute Myocardial Infarction (LIMIT AMI) Study. Circulation 104, Hicks, A. E., Ridger, V. C., Dixon, R., Dupre, B., Hellewell, P. G., and Norman, K. E. (2000) Anti-inflammatory effects of TBC1269 are not due to competitive inhibition of leukocyte rolling (Abstract). FASEB Journal 15, A Nelson, R. M., Dolich, S., Aruffo, A., Cecconi, O., and Bevilacqua, M. P. (1993) Higheraffinity oligosaccharide ligands for E-selectin. Journal of Clinical Investigation 91, Kunkel, E. J., Jung, U., Bullard, D. C., Norman, K. E., Wolitzky, B. A., Vestweber, D., Beaudet, A. L., and Ley, K. (1996) Absence of trauma-induced leukocyte rolling in mice deficient in both P-selectin and intercellular adhesion molecule-1 (ICAM-1). Journal of Experimental Medicine 183, Ley, K., Linnemann, G., Meinen, M., Stoolman, L. M., and Gaehtgens, P. (1993) Fucoidin, but not yeast polyphosphomannan PPME inhibits leukocyte rolling in venules of the rat mesentery. Blood 81, Ley, K., Cerrito, M., and Arfors, K. E. (1991) Sulfated poysaccharides inhibit leukocyte rolling in rabbit mesentery venules. Journal of Physiology 260, H1667

12 41. Lindbom, L., Xie, X., Raud, J., and Hedqvist, P. (1992) Chemoattractant-induced firm adhesion of leukocytes to vascular endothelium in vivo is critically dependent on initial leukocyte rolling. Acta Physiol. Scand. 146, Granert, C., Raud, J., Xie, X., Lindquist, L., and Lindbom, L. (1994) Inhibition of leukocyte rolling with polysaccharide fucoidin prevents pleocytosis in experimental meningitis in the rabbit. Journal of Clinical Investigation 93, Liu, W., Ramachandran, V., Kang, J., Kishimoto, T. K., Cummings, R. D., and McEver, R. P. (1998) Identification of N-terminal residues on P-selectin glycoprotein ligand- 1 required for binding to P-selectin. J Biol Chem 273, Sako, D., Comess, K. M., Barone, K. M., Camphausen, R. T., Cumming, D. A., and Shaw, G. D. (1995) A sulfated peptide segment at the amino terminus of PSGL-1 is critical for P-selectin binding. Cell 83, Goetz, D. J., Greif, D. M., Ding, H., Camphausen, R. T., Howes, S., Comess, K. M., Snapp, K. R., Kansas, G. S., and Luscinskas, F. W. (1997) Isolated P-selectin glycoprotein ligand-1 dynamic adhesion to P- and E- selectin. J Cell Biol 137, Somers, W. S., Tang, J., Shaw, G. D., and Camphausen, R. T. (2000) Insights into the molecular basis of leukocyte tethering and rolling revealed by structures of P- and E- selectin bound to slex and PSGL-1. Cell 103, Khor, S. P., McCarthy, K., DuPont, M., Murray, K., and Timony, G. (2000) Pharmacokinetics, pharmacodynamics, allometry, and dose selection of rpsgl-ig for phase I trial. J Pharmacol Exp Ther 293, Zimmerman, G. A. (2001) Two by two: The pairings of P-selectin and P-selectin glycoprotein ligand 1. Proc Natl Acad Sci USA 98, Video file (shown in Supplemental Data link) Video 1 shows leukocyte rolling 31 min after surgical preparation of the mouse cremaster for intravital microscopy. Leukocyte rolling is almost exclusively P-selectin-dependent at this time. Baseline rolling was observed for 1 min between 30 and 31 min (not shown). 2-GSP-6 (4.3 µmol/kg) was injected via the jugular vein at precisely 31 min. The video runs at four times normal speed. Effects of 2-GSP-6 begin to develop at 31 min 20 s and are maximal by 31 min 35 s. Received March 7, 2002; revised May 16, 2002.

13 Table 1 Effects of 2-GSP-6 and 4-GSP-6 on circulating leukocyte count and blood flow in mice with P-selectindependent leukocyte rolling. Treatment Systemic leukocytes Centerline blood (cells/µl blood) flow velocity (mm/s) 30 min 32 min 30 min 32 min 2-GSP ± ± ± ± µmol/kg 4-GSP ± ± ± ± µmol/kg

14 Fig. 1 Figure 1. A) Structures of glycosulfopeptides (GSP). With the exception of the C-terminal amino acid, 2-GSP-6 is identical to 4-GSP-6 and 2-GSP-1 to 4-GSP-1. B) Equilibrium binding affinity of 4-GSP-6 for soluble P-selectin at subphysiological salt concentration. Bound GSP and free soluble P-selectin concentrations were calculated from equilibrium gel filtration data (data not shown), and the dissociation constant was derived using a rectangular hyperbola equation to calculate a best-fit curve.

15 Fig. 2 Figure 2. A) Effects of different GSP on proportion of leukocytes rolling through observed venules. Baseline leukocyte rolling was determined from a 1-min observation recorded 30 min after surgical preparation of the cremaster for intravital microscopy. Glycosulfopeptides were injected at 31 min, and effects were determined between 32 and 33 min. P-selectin (RB40.34) or PSGL-1 (2PH1) blocking antibodies were injected at 45 min as a positive control. Numbers in parentheses indicate the dose of GSP in µmol/kg. Asterisks indicate significant difference from baseline rolling (*P<0.05, **P<0.01, ***P<0.001). B) Time course of effect of GSP on leukocyte rolling. Individual vessels were tracked for up to 10 min after injection of GSP. Open diamonds indicate rolling after 4.3 µmol/kg 2-GSP6. Closed circles, squares, and triangles indicate rolling after 1.43, 4.3, and 12.9 µmol kg 4-GSP-6, respectively. Data in A and B are presented as mean ±SE of n=9 12 venules from 3 4 mice per treatment.

16 Fig. 3 Figure 3. Effect of different doses of 4-GSP-6 on leukocyte rolling velocity. Average rolling velocities were determined every min for 10 min after injection of 4-GSP-6. Data are from 9 12 venules from 3 4 mice per treatment. Asterisks indicate significant difference from baseline rolling (*P<0.05, **P<0.01).

17 Fig. 4 Figure 4. A) Clearance of 125 I-4-GSP-6 from blood. B) Accumulation of 125 I-4-GSP-6 in urine and different organs. Data are presented as mean ±SE of measurements taken from three experiments. Asterisks indicate significant difference from baseline rolling (*P<0.05, **P<0.01).