Inhibitors Possessing Sialyl Lewisx Moieties ...

Endothelial Targeting and Enhanced Antiinflammatory Effects of Complement Inhibitors Possessing Sialyl Lewisx Moieties This information is current as of June 2, 2013.

Michael S. Mulligan, Roscoe L. Warner, Charles W. Rittershaus, Lawrence J. Thomas, Una S. Ryan, Kimberly E. Foreman, Larry D. Crouch, Gerd O. Till and Peter A. Ward J Immunol 1999; 162:4952-4959; ; http://www.jimmunol.org/content/162/8/4952

Subscriptions Permissions Email Alerts

This article cites 40 articles, 20 of which you can access for free at: http://www.jimmunol.org/content/162/8/4952.full#ref-list-1 Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscriptions Submit copyright permission requests at: http://www.aai.org/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/cgi/alerts/etoc

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 9650 Rockville Pike, Bethesda, MD 20814-3994. Copyright © 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.

Downloaded from http://www.jimmunol.org/ by guest on June 2, 2013

References

Endothelial Targeting and Enhanced Antiinflammatory Effects of Complement Inhibitors Possessing Sialyl Lewisx Moieties1 Michael S. Mulligan,* Roscoe L. Warner,† Charles W. Rittershaus,§ Lawrence J. Thomas,§ Una S. Ryan,§ Kimberly E. Foreman,¶ Larry D. Crouch,‡ Gerd O. Till,† and Peter A. Ward2†

M

any acute inflammatory reactions are complement dependent. The complement system serves as a source of peptides (anaphylatoxins, C3a and C5a) that have powerful vasopermeability and/or leukocyte chemotactic activities. Inflammatory reactions that are complement dependent can be suppressed by the inhibition of complement activation products or by blocking activation of the complement system. One effective strategy for suppressing these reactions is the blockade of complement activation products (such as C5a) with specific Ab (1– 4). These Ab-based approaches, however, have the intrinsic problem of the development of immune responses to idiotypic or nonidiotypic domains of the blocking Abs. Blocking activation of the complement system can be achieved by consumptive complement depletion, such as which occurs with repeated i.p. injection of purified cobra venom factor (CVF)3 (5, 6), by Ab-induced inhibition of individual complement components such as C5 (3), or by infu-

Departments of *Surgery and †Pathology, University of Michigan Medical School, Ann Arbor, MI 48109; ‡Department of Physiology, University of Nebraska School of Dentistry, Lincoln, NB 68583; §Avant Immunotherapeutics, Inc., Needham, MA 02494; and ¶Department of Pathology, Loyola University School of Medicine, Maywood, IL 60153 Received for publication November 18, 1998. Accepted for publication January 19, 1999. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1

Supported by National Institutes of Health Grants GM-29507 and HL-31963.

2

Address correspondence and reprint requests to Dr. Peter A. Ward, 1301 Catherine Road, Department of Pathology, Box 0602, University of Michigan Medical School, Ann Arbor, Michigan 48109-0602. E-mail address: [email protected] 3 Abbreviations used in this paper: CVF, cobra venom factor; LHR, long homologous repeats (A-D); MPO, myeloperoxidase; SCR, short consensus repeats; sCR1, soluble complement receptor type 1; sCR1sLex, sCR1 bearing sialyl Lewisx; sCR1[desLHRA], sCR1 constructed by deleting LHR-A; IC50, concentration that inhibits response by 50%; PSGL-1, P-selectin glycoprotein ligand-1.

Copyright © 1999 by The American Association of Immunologists

sion of soluble versions of soluble complement receptor type 1 (sCR1) (7). sCR1 (soluble CD35) is a single chain glycoprotein consisting of 30 homologous protein domains known as short consensus repeats (SCR), followed by transmembrane and cytoplasmic domains (8, 9). Groups of seven SCRs form long homologous repeats (LHRs), which have been designated LHR-A, -B, -C, and -D for the most common human allotype of CR1. sCR1 was prepared by deleting the cytoplasmic and transmembrane domains while retaining LHR-A, -B, -C, and -D (10, 11). This recombinant molecule blocked the assembly of enzymes (convertases) responsible for cleavage of C3 and C5 and subsequent activation of the complement system and served as a cofactor in the proteolysis of C3b and C4b by Factor I. sCR1 has been shown to inhibit both classical and alternative pathways of complement activation (10, 11). An additional soluble version of sCR1 has been constructed by deleting LHR-A (as well as the cytoplasmic and transmembrane domains). The resulting recombinant molecule, consisting of LHR-B, -C, and -D, has been designated sCR1[desLHR-A] (12). Because sCR1[desLHR-A] lost the C4b binding site contained in LHR-A, it should be a relatively selective (but not absolute) inhibitor of the alternative pathway of complement activation. In the first steps of the inflammatory response, recruitment of leukocytes into tissues requires that the vascular endothelium be “activated” to express adhesion molecules, which serve to tether blood leukocytes to the endothelium before their diapedesis (13– 18). Two adhesion molecules expressed during endothelial activation are P- and E-selectin, both containing binding sites for sialated, fucosylated sLex and related motifs present on neutrophils and other leukocytes (13, 17, 18). Developing molecules to interrupt the binding of leukocytes to endothelial selectins has been a goal of many drug development programs. One particular strategy has been the use of soluble oligosaccharides containing the sLex 0022-1767/99/$02.00


The complement inhibitor soluble complement receptor type 1 (sCR1) and a truncated form of sCR1, sCR1[desLHR-A], have been generated with expression of the selectin-reactive oligosaccharide moiety, sialyl Lewisx (sLex), as N-linked oligosaccharide adducts. These modified proteins, sCR1sLex and sCR1[desLHR-A]sLex, were assessed in the L-selectin- and P-selectin-dependent rat model of lung injury following systemic activation of complement by cobra venom factor and in the L-selectin-, P-selectin-, and Eselectin-dependent model of lung injury following intrapulmonary deposition of IgG immune complexes. In the cobra venom factor model, sCR1sLex and sCR1[desLHR-A]sLex caused substantially greater reductions in neutrophil accumulation and in albumin extravasation in lung when compared with the non-sLex-decorated forms. In this model, increased lung vascular binding of sCR1sLex and sCR1[desLHR-A]sLex occurred in a P-selectin-dependent manner, in contrast to the absence of any increased binding of sCR1 or sCR1[desLHR-A]. In the IgG immune complex model, sCR1[desLHR-A]sLex possessed greater protective effects relative to sCR1[desLHR-A], based on albumin extravasation and neutrophil accumulation. Enhanced protective effects correlated with greater lung vascular binding of sCR1[desLHR-A]sLex as compared with the non-sLex-decorated form. In TNFa-activated HUVEC, substantial in vitro binding occurred with sCR1[desLHR-A]sLex (but not with sCR1[desLHR-A]). This endothelial cell binding was blocked by anti-E-selectin but not by anti-P-selectin. These data suggest that sLex-decorated complement inhibitors have enhanced antiinflammatory effects and appear to have enhanced ability to localize to the activated vascular endothelium. The Journal of Immunology, 1999, 162: 4952– 4959.

The Journal of Immunology

Materials and Methods Recombinant complement inhibitory proteins sCR1 was produced in the Chinese hamster ovary cell line DUKX B11 and purified as previously described (11). sCR1[desLHR-A] was produced in the DUKX B11 cell line that had been transfected with the plasmid pTCR1c6A. The resulting secreted glycoprotein was purified and characterized. sCR1[desLHR-A] was an effective inhibitor of the alternative complement pathway comparable to sCR1 (see below). As expected, sCR1[desLHR-A] was a less effective in vitro inhibitor of the classical complement pathway when compared with sCR1 (12). sCR1[desLHR-A] and sCR1[desLHR-A]sLex were labeled with 125I (New England Nuclear, Boston, MA) using lactoperoxidase techniques. sCR1 and sCR1[desLHRA] produced by DUKX B11 possessed no sLex on their N-linked oligosaccharides (48) (M. D. Picard et al., manuscript in preparation). For sCR1sLex production, the expression plasmid coding for sCR1 was used (11). The plasmid pTCSLDHFR*, coding for a mutant mouse dihydrofolate reductase with an abnormally low affinity for methotrexate (29), was derived from pSV2-DHFR* and cloned into pTCSLneo by direct substitution of the neomycin resistance gene. The Chinese hamster ovary cell line, LEC11 (20), which expresses the a (1–3) fucosyltransferase activity necessary for synthesis of sLex-related oligosaccharides (21), was cotransfected with pTCSLDHFR* together with the plasmid coding for sCR1. Clones grown in medium containing methotrexate were selected for production of high concentrations of sCR1sLex. Carbohydrate analysis of the purified glycoproteins indicated sLex glycosylation of sCR1sLex (48) (M. D. Picard et al., manuscript in preparation). sCR1[desLHR-A] sLex was prepared by transfecting the LEC11 cell line with the pT-CR1c6A plasmid, and the resulting secreted glycoprotein was purified and characterized. sCR1[desLHR-A] and sCR1[desLHR-A]sLex were shown to be equivalent inhibitors of complement in vitro.4 Carbohydrate analysis of the purified glycoproteins indicated N-linked sLex glycosylation of sCR1[desLHR-A]sLex (48).

Animal models and in vivo binding assays CVF was isolated from Naja naja venom by a combination of ion exchange and gel filtration techniques (5). Four units CVF were injected i.v. as a single bolus into young male, specific pathogen-free, Long-Evans rats (300 to 350 g). For binding studies, 0.5 mCi of 125I-sCR1[desLHR-A] or 125IsCR1[desLHR-A]sLex together with unlabeled forms (1.5 mg) of the same compounds was infused i.v. just before i.v. infusion of either sterile saline or CVF. Some animals also received an i.v. infusion of 200 mg of either PB1.3 (IgG1 anti-P-selectin) or MOPC-21 (a subclass-matched Ab) just before infusion of CVF together with 125I-sCR1[desLHR-A] or 125IsCR1[desLHR-A]sLex. PB1.3 is a monoclonal mouse IgG1 with reactivity to human P-selectin and cross-reactivity to rat P-selectin (30). For lung binding studies, animals were killed 20 min after i.v. infusion of CVF. This is the time at which lung vascular P-selectin peaks in vivo (31). For measurement of injury parameters, 0.5 mCi of 125I-BSA was injected i.v. just before infusion of CVF. Unless otherwise indicated, animals were sacrificed 30 min later, and the pulmonary arterial circulation was flushed with 10 ml of sterile saline to remove residual blood (and blood-associated 125 I-BSA). The amount of radioactivity in the lungs was compared with that present in 1.0 ml of blood obtained from the inferior vena cava at the time of sacrifice. Permeability and hemorrhage indices were calculated, as described elsewhere (30). Briefly, the permeability index was calculated by the ratio of radioactivity (125I-BSA) present in saline-perfused lungs 30 min after i.v. infusion of CVF to the amount of radioactivity present in 1.0 ml of blood. Lung injury was also induced with a rabbit polyclonal IgG rich in Ab to BSA (Organon Teknika, West Chester, PA) as previously described (23– 26). As indicated above, lung injury in this model requires engagement of all three selectins. Briefly, 2.5 mg anti-BSA in 300 ml was instilled into rat lungs via a tracheal cannula. In the positive controls, this was followed by an i.v. injection of 10 mg of BSA together with trace amounts of 125I-BSA. In the negative controls, the i.v. infusion of 10 mg BSA was omitted. For binding studies in the IgG immune complex model, similar amounts of nonlabeled and 125I-labeled sCR1[desLHR-A] or sCR1[desLHR-A]sLex together with 4.5 mg of unlabeled sCR1 derivatives were infused i.v. 20 min before sacrifice, at 4 h. This is when up-regulation of P- and E-selectins are maximally in this model of injury (25). Lung injury was quantitated by permeability and MPO measurements, as indicated above.

Inhibition of complement activation in vitro Having established the presence of the sLex tetrasaccharide in the LEC11 glycoproteins, as described above, it was important to examine the effects of such glycosylation on complement inhibitory function. As described elsewhere, the concentrations required to inhibit human complement-mediated lysis of erythrocytes were similar for glycoproteins expressed by LEC11 cells (sLex versions) compared with those expressed by DUKXB11 cells (non-sLex version). Both versions of sCR1 and sCR1[desLHRA] were similar in their capacity to inhibit alternative pathway of complement activation, but the sCR1 form was, as expected, more effective than sCR1[desLHR-A] in the inhibition of the classical complement pathway. Averaging the results of a number of experiments, the concentration of sCR1sLex required to yield half-maximal lysis of sensitized sheep erythrocytes was somewhat higher (IC50 5 0.27 6 0.082 nM, n 5 31) than that required for sCR1 (IC50 5 0.21 6 0.060 nM, n 5 65). This suggests that sCR1sLex is a slightly less effective inhibitor of classical complement activation. Similarly, in the assay of alternative pathway lysis of guinea pig erythrocytes, sCR1sLex appeared somewhat less effective (IC50 5 38 6 16 nM, n 5 37) than sCR1 (IC50 5 19 6 6.6 nM, n 5 10). Analogous results were obtained for the two versions of sCR1[desLHR-A]. In the classical pathway assay, sCR1[desLHR-A]sLex (IC50 5 140 6 32 nM, n 5 3) was somewhat less effective than sCR1[desLHR-A] (IC50 5 58 6 38 nM, n 5 4) but much less effective than either version of sCR1. In the alternative pathway assay, sCR1[desLHR-A]sLex (IC50 5 46 6 9.1 nM, n 5 4) was again somewhat less effective than sCR1 [desLHR-A] (IC50 5 37 6 6.2 nM, n 5 4) but comparable to either version of sCR1.

In vitro binding assays using endothelial cells For in vitro binding of sCR1[desLHR-A] or sCR1[desLHR-A]sLex to endothelial cells, freshly isolated HUVEC (passage 1–3) in 12-well tissue culture plates were incubated for 4 h at 37°C in the presence or absence of 50 ng/ml TNF-a. The cells were washed twice with HBSS containing 1% BSA and then incubated with sCR1[desLHR-A] or sCR1[desLHR-A]sLex (10 mg/ml) and containing trace amounts of 125Ilabeled proteins in the presence or absence of Abs (5 mg/ml) directed against human E-selectin (CL-3), P-selectin (PB1.3), or an isotypematched irrelevant Ab (MOPC-21). Following a 10-min incubation, the


moiety (19). In the current report, we assessed sLex-decorated versions of sCR1 (and the truncated forms). Such altered molecules might possess two novel features. First, they could physically block the initial, selectin-dependent in vivo leukocyte binding (associated with the “rolling” phenomenon) and consequently act as antiinflammatory agents. Second, the presence of the sLex motif within natural oligosaccharides of sCR1 could also serve to localize this molecule to areas of inflammation by binding to endothelial selectins that are reactive with sLex. One method to obtain proteins with the sLex motif attached to naturally expressed oligosaccharides would be in vitro chemical or enzymatic modification of proteins after their synthesis. This method would likely result in limited amounts of the final desired product. The alternative, chosen for the current studies, would be production of the recombinant protein in a cell line containing the specific fucosyl transferase activity that would allow addition of a1–3 linked fucose during the course of normal oligosaccharide synthesis (20, 21). This would allow production of the sLex motif within natural N-linked oligosaccharides. This strategy was applied for production of modified proteins, sCR1sLex and sCR1[desLHR-A]sLex. These proteins have been shown to possess sLex as a portion of their natural N-linked oligosaccharides (48) (M. D. Picard et al., manuscript in preparation). This report describes the in vivo efficacy of sCR1sLex and sCR1[desLHRA]sLex in lung inflammatory models of neutrophil-mediated lung injury: systemic activation of complement, which induces injury that is P- and L-selectin dependent (22–25), and intrapulmonary deposition of IgG immune complexes, which induces injury that is P-selectin, L-selectin, and E-selectin dependent (23–26). Previously, in both models of lung injury, sCR1 and the soluble sLex tetrasaccharide have been shown to be protective (7, 27, 28) The current studies indicate that sLex-decorated versions of sCR1 are more protective in vivo and have the ability to localize to the activated vascular endothelium.

4953

ANTIINFLAMMATORY EFFECTS OF sCR1sLex

4954 Table I. Protective effects of sCR1 and sCR1sLex after CVF infusion

Parameter of Injurya Material Infused

sCR1 sCR1sLex

Dose (mg)

Permeability

MPO

0.30 1.5 4.5 0.30 1.5 4.5

1.05 6 0.05 (NS)* 0.90 6 0.02 (0.004)* 0.81 6 0.05 (0.02)* 0.60 6 0.06 (,0.001)** 0.462 6 0.03 (,0.001)** 0.38 6 0.03 (,0.001)**

1.11 6 0.05 (0.04)* 0.95 6 0.06 (0.002)* 0.88 6 0.05 (0.03)* 0.95 6 0.03 (0.003)** 0.62 6 0.03 (0.003)** 0.43 6 0.02 (,0.001)**

a Permeability values for positive and negative control groups were 1.28 6 0.11 and 0.15 6 0.01, respectively. The MPO values for positive and negative control groups were 1.29 6 0.05 and 0.27 6 0.02, respectively. For each group n $ 5. pp values when compared with positive control group that was otherwise untreated. ppp values when the companion groups in the same columns were compared at similar doses of sCR1 or sCR1sLex.

cells were washed to remove any unbound radiolabel and then lysed with 2% Triton X-100 for isotope counting. Nonspecific binding to the cells was measured in the presence of 50- to 100-fold excess of unlabeled sCR1[desLHR-A] or sCR1[desLHR-A]sLex.

Whole lungs were homogenized with a Polytron homogenizer (Tekmar, Cincinnati, OH) (4 to 10 s at setting of 4) in a volume of 6 ml, using a homogenizer buffer (50 mM phosphate, pH 6.0). Samples were then subjected to centrifugation (3000 3 g, 30 min) at 4°C. Myeloperoxidase (MPO) activity in supernatant fluid was assayed by measuring the change (per min) in absorbance at 460 nm resulting from the oxidation of o-dianisidine in the presence of H2O2 (7).

Immunohistochemistry For immunostaining of frozen sections, rat lungs were frozen in optimal cutting temperature (O.C.T.) compound (Miles, Elkhart, IN) and stained with rabbit polyclonal IgG Ab to human sCR1, which was diluted 1:1000 in PBS (pH 7.4) containing 0.1% BSA for 1 h in a humidified chamber. Slides were then washed two times in PBS and incubated for 1 h with horseradish peroxidase-conjugated goat anti-rabbit IgG-specific Ab (Rockland, Gilbertsville, IL) diluted 1:10,000 in PBS. Slides were washed two times in PBS, dried, and incubated with horseradish peroxidase-specific substrate True Blue (Kirkegaard & Perry, Gaithersburg, MD) for 5 min, washed, then dipped in 100% ethanol, dried, and mounted under coverslips.

Statistical analysis The data were subjected to ANOVA. Paired or unpaired Student’s t test multigroup comparisons were also determined using the Schaffer t test, as well as the Fisher protected least significant difference test. All statistical comparisons were made between treatment groups and positive controls after mean background negative control values had been subtracted from each data point. All values were expressed as mean 6 SEM. Statistical significance was defined as p , 0.05.

Results CVF-induced lung injury: comparison of the protective effects of sCR1 and sCR1sLex The relative efficacy of sCR1 and sCR1sLex was determined in the CVF model of lung injury employing i.v. doses of 0.30, 1.5, and 4.5 mg per animal. Inhibitors (200 ml) were infused i.v. immediately before i.v. infusion of 4 U of CVF together with 5 mCi 125IBSA (100 ml), which was used to measure lung vascular permeability. Animals were sacrificed 30 min later. Results are shown in Table I. After infusion of CVF, there were 5- to 10-fold increases in albumin leak and in MPO content when compared with negative controls (no CVF) (Table I, footnotes). At doses of 0.30, 1.5, and 4.5 mg, sCR1 reduced permeability indices by 20%, 34%, and 42%, respectively. The latter two values showed statistically significantly greater protection ( p , 0.05) when compared with the values of the otherwise untreated positive controls (CVF alone). In the companion group treated with sCR1sLex, the doses of 0.30,

Immunostaining of sCR1 and sCR1sLex in rat lung after CVF infusion Using immunostaining to detect binding of sCR1 and sCR1sLex to the lung vasculature, 0.30 mg of each inhibitor was injected i.v. 10 min after i.v. infusion of 4 U of CVF. Animals were then sacrificed 10 min after the infusion of either sCR1 or sCR1sLex, and the lungs were prepared for immunostaining. At this point (20 min after infusion of CVF), lung vascular P-selectin is maximally upregulated (31). A polyclonal affinity-purified Ab to sCR1 (which also detects sCR1sLex) was employed. The results of these studies are shown in Fig. 1. Rats treated with sCR1 and CVF revealed no detectable binding of sCR1 to the lung vasculature (Fig. 1, A and C), whereas treatment with sCR1sLex and CVF resulted in obvious evidence of binding of sCR1sLex to lung interstitial capillaries and venules (Fig. 1, B and D). These data provide direct evidence for


Tissue myeloperoxidase content

1.5, and 4.5 mg caused reductions of 57%, 69%, and 80%, respectively, all of which were significantly greater than values obtained with sCR1 (Table I). Thus, at similar doses, sCR1sLex was substantially more protective than sCR1. Lung neutrophil accumulation as defined by lung MPO content was also compared for each of the three doses of the two sCR1 compounds (Table I). At doses of 0.30, 1.5, or 4.5 mg, sCR1 caused reductions in MPO content in lung of 18%, 33%, and 40%, respectively, all being statistically significant ( p , 0.05) when compared with the otherwise untreated positive controls. In animals treated with sCR1sLex at doses of 0.30, 1.5, or 4.5 mg, MPO levels fell by 33%, 56%, and 75%, respectively, each being statistically significantly more protective when compared with the effects of sCR1 (Table I). Thus, sCR1sLex is a more effective inhibitor than is sCR1. The observed differences in the efficacy of sCR1 and sCR1sLex could not be explained by differences in the pharmacokinetic or complement-inhibiting properties of the compounds. The pharmacokinetics of sCR1 and sCR1sLex were determined in normal adult male rats (n $ 5 for each group) given a bolus i.v. infusion (10 mg/kg) of either sCR1 or sCR1sLex. Plasma samples (obtained at 0, 3, 10, 30, and 60 min during the first hour) were quantitated for sCR1 and sCR1sLex by enzyme immunoassay techniques. During the first hour, the blood levels of the two compounds were not statistically different (data not shown). It would appear in the CVF model of lung injury (where sacrifice occurs at 30 min) that the greater protective effects of sCR1sLex cannot be explained by differences in pharmacokinetics. With respect to complement-inhibiting properties of sCR1 and sCR1sLex, similar concentrations of sCR1 and sCR1sLex showed similar inhibition in human serum complement-mediated lysis of Ab-sensitized sheep erythrocytes (see above).


4955

the binding of sCR1sLex, but not sCR1, to the lung vasculature of animals following i.v. infusion of CVF. Protective effects of sCR1[desLHR-A] and sCR1[desLHR-A]sLex after CVF infusion We evaluated the protective effects of sCR1[desLHR-A] and sCR1[desLHR-A]sLex in the CVF model of lung injury. The results are shown in Table II. At sCR1[desLHR-A] doses of 0.30, 1.5, and 4.5 mg, the permeability indices fell by 6%, 41%, and 53%, respectively, the latter two values being statistically significantly different from those of the reference positive control groups not otherwise treated. In animals treated with sCR1 [desLHRA]sLex, at the same doses, the permeability indices were reduced 20%, 53%, and 69%, respectively. Only the highest dose gave statistically greater protection ( p 5 0.03) when compared with the undecorated form of the inhibitor. When MPO measurements were determined, the sCR1 [desLHR-A] doses of 0.30, 1.5, and 4.5 mg caused reductions of 8%, 34%, and 46%, respectively (Table II). Statistically, when compared with the otherwise untreated positive controls, the two higher doses of sCR1 [desLHR-A] achieved statistical significance. When animals were treated with sCR1 [desLHR-A] sLex at doses of 0.30, 1.5, and 4.5 mg, the reductions in MPO content were 22%, 44%, and 66%, respectively. Again, only the highest dose of

sCR1 [desLHR-A] sLex achieved statistical significance ( p 5 0.01) when compared with sCR1 [desLHR-A]. Thus, although the sLex-decorated compound was a more effective inhibitor, these effects were only seen at the highest dose, unlike the full-length sCR1s described in Table I. Binding of sCR1[desLHR-A] and sCR1[desLHR-A]sLex to lung vasculature after CVF infusion With availability of 125I-versions of the decorated and undecorated sCR1[desLHR-A] compounds, we evaluated the lung vascular binding of 125I-sCR1[desLHR-A] and 125I-sCR1[desLHR-A]sLex after i.v. infusion of PBS or 4 U of CVF in the presence or absence of 200 mg i.v. infused PB1.3 IgG1 (anti-P-selectin) or MOPC-21 IgG1 (irrelevant control Ab). Twenty minutes later, lung binding was determined in lungs after the pulmonary artery had been infused with sterile saline (10 ml) to clear residual blood. The composite results of these studies are shown in Table III. The binding of sCR1[desLHR-A] and sCR1 [desLHR-A]sLex to the lung vasculature in the presence of circulating MOPC-21 after infusion of either saline was comparable (5.78 6 0.53 vs 6.34 6 0.715 mg, respectively; p, NS). The binding to lung of sCR1[desLHR-A]sLex in the presence of MOPC-21 in saline-infused animals was similar to sCR1[desLHR-A] to the binding after saline infusion (6.34 6 0.31) and in CVF-infused animals (6.63 6 0.71). However, sCR1[desLHR-A]sLex binding to lungs of

Table II. Protective effects of sCR1 (desLHR-A) and sCR1 (desLHR-A) sLex in CVF model Parameter of Injurya Material Infused

sCR1 [desLHR-A] sCR1 [desLHR-A]sLex

Dose (mg)

Permeability

MPO

0.30 1.5 4.5 0.30 1.5 4.5

0.61 6 0.03 (NS)* 0.44 6 0.02 (,0.001)* 0.38 6 0.02 (,0.001)* 0.54 6 0.02 (NS)** 0.38 6 0.02 (NS)** 0.30 6 0.02 (0.03)**

0.48 6 0.03 (NS)* 0.38 6 0.02 (0.002)* 0.33 6 0.02 (,0.001)* 0.43 6 0.02 (NS)** 0.34 6 0.02 (NS)** 0.25 6 0.01 (0.01)**

a Permeability values for positive and negative control groups were 0.64 6 0.02 and 0.15 6 0.01, respectively. The MPO values for positive and negative control groups were 0.85 6 0.02 and 0.11 6 0.02, respectively. For each group n 5 5. pp values when compared to positive control group that was otherwise untreated. ppp values when similar doses for the two groups were compared at similar doses. NS, not significant (p . 0.05).


FIGURE 1. Immunostaining for sCR1 in frozen lung sections of rats infused i.v. with CVF followed by sCR1 or sCR1sLex. Animals received 4 U of CVF followed 10 min later by sCR1 or sCR1sLex, with sacrifice 10 min thereafter. A, Lung section from a CVF-infused animal receiving sCR1 showed no staining using affinity-purified rabbit anti-sCR1 (3100). B, A lung venule of an animal receiving CVF followed by sCR1sLex showed staining with anti-sCR1 (3100). C, Lung capillaries (arrows) of an animal receiving CVF followed by sCR1 showed no staining with anti-sCR1 (3200). D, In contrast, lung interstitial capillaries (arrow) in a CVF-infused animal receiving sCR1sLex demonstrated staining with anti-sCR1 (3200). All sections were counterstained with eosin.


4956

Table III. Binding of sCR1[desLHR-A] and sCR1[desLHR-A]sLex to rat lungs after infusion of saline or CVFa Amount (mg) Boundb Material Infused

Ab (200 mg)

Saline

MOPC-21

NS

5.78 6 0.53 NS

CVF

sCR1[desLHR-A]sLex

sCR1[desLHR-A]

0.005

6.63 6 0.71

MOPC-21

NS CVF

7.91 6 0.29 ,0.05

0.005

7.93 6 0.74

PB1.3

0.04

NS 5.41 6 0.29

PB1.3

0.02

13.7 6 0.76

NS Saline

6.34 6 0.31

10.1 6 0.54

Statistical analysis (p value) is indicated by the numbers between lines connecting the groups compared. NS, p . 0.05. b Determined using 125I-labeled compounds, as described in the text. a

Protective effects of sCR1[desLHR-A] and sCR1[desLHR-A]sLex in injury caused by IgG immune complexes The complement inhibitors sCR1[desLHR-A] and sCR1[desLHRA]sLex were evaluated in the L-selectin-, P-selectin-, and E-selectin-dependent model of acute vascular injury caused by the intrapulmonary deposition of IgG immune complexes. These inhibitors were infused at a dose of 4.5 mg. Permeability indices and MPO activity were assessed (as described above). The results are summarized in Table IV. Treatment with sCR1[desLHR-A]

reduced the permeability index and MPO values by 45% ( p 5 0.04) and 50%, ( p , 0.001), respectively, while treatment with sCR1[desLHR-A]sLex reduced these values by 63% ( p , 0.002) and 71%, ( p 5 0.005), respectively. For both permeability indices and MPO values, the differences in effects of sCR1[desLHR-A] as compared with those of sCR1[desLHR-A]sLex were statistically significantly different, with the latter being more protective. Thus, in both models of acute lung injury, sCR1[desLHR-A]sLex at the 4.5-mg dose demonstrated greater protective effects than sCR1[desLHR-A]. Lung vascular binding of sCR1[desLHR-A] and sCR1[desLHR-A]sLex after intrapulmonary deposition of IgG immune complexes We evaluated the binding of 125I-sCR1[desLHR-A] and 125IsCR1[desLHR-A]sLex 3 h and 45 min after initiation of IgG immune complex-induced lung injury. The animals were infused with 4.5 mg of sCR1[desLHR-A] or sCR1[desLHR-A]sLex, together with 125I-labeled compounds, to provide 800,000 cpm per animal. Fifteen minutes later, the animals were sacrificed. Binding data are shown in Table V. Binding values for sCR1[desLHR-A] in lungs of negative (no IgG immune complexes) and positive (IgG immune complexes) control groups were 5.60 6 0.50 and 5.40 6 0.30 mg, respectively. In the negative control groups, the binding of sCR1[desLHR-A]sLex, as compared with binding of sCR1[desLHR-A], was doubled in the negative control group, to Table V. Binding of sCR1[desLHR-A] and sCR1[desLHR-A]sLex to rat lungs after IgG immune complex deposition

Table IV. Comparison of protective effects of sCR1[desLHR-A] and sCR1[desLHR-A]sLex in rat lungs after IgG immune complex deposition Group

Parameter of Lung Injurya Treatment (4.5 mg)

Permeability

0.54 6 0.1

None

MPO

0.36 6 0.1 x

sCR1[desLHR-A]sLe

0.29 6 .02

Negative controls

0.95 6 .03

,0.002

0.61 6 .02 0.47 6 .02

IgG immune complex-treated 0.005

The mean permeability value for the negative control group was 0.14 6 0.02. The MPO value for the negative control group was 0.27 6 0.04. Numbers connecting the lines between groups represent p values. For each group, n 5 5. a

sCR1[desLHR-A]

5.60 6 0.50

sCR1[desLHR-A]sLex 0.03

11.1 6 1.0 0.002

NS

,0.001

0.004

sCR1[desLHR-A]

Amount (mg) Bounda

5.40 6 0.30

25.4 6 1.0 ,0.001

a Details are described in text. Binding was assessed 4 h after initiation of the inflammatory response. Statistical analysis (positive) is indicated by the numbers between lines connecting the groups compared. For each group, n 5 5. Binding was determined using 125I-labeled compounds.


animals infused with CVF and MOPC-21 was much higher (13.7 6 0.76 mg). In animals infused with saline in the presence of anti-Pselectin (PB1.3), the binding of both sCR1[desLHR-A] and sCR1[desLHR-A]sLex was relatively low and similar (5.41 6 0.29 and 7.93 6 0.29, respectively; p, NS). Binding of sCR1[desLHR-A] in the presence of PB1.3 was similar in lungs of animals infused with saline (5.41 6 0.29) or CVF (7.93 6 0.74) ( p, NS). With sCR1[desLHR-A]sLex, binding in saline-infused animals also treated with PB1.3 was not statistically different (7.91 6 0.29) ( p, NS). However, in CVF-treated rats in the presence of PB1.3, the binding of sCR1[desLHR-A]sLex fell to 10.1 6 0.54 mg in CVF-infused animals, a significant drop ( p 5 0.04) from the 13.7 6 0.76 value in the CVF-infused animals also receiving MOPC-21. These data indicate that, in CVF-treated rats, there is increased specific binding of sCR1[desLHR-A]sLex to the lung vasculature when compared with the undecorated form of this molecule and that this binding can be significantly diminished by the presence of anti-P-selectin (PB1.3).


4957

11.1 6 1.0 mg, perhaps due to vascular perturbations following airway instillation of anti-BSA (in the absence of i.v. infused BSA). In the positive control group, binding of sCR1[desLHRA]sLex rose to 25.4 6 1.0 mg, more than 2-fold above the level in the negative control group, and nearly 5-fold when compared with amounts of sCR1[desLHR-A] bound in negative or positive control lungs. Accordingly, the binding of sCR1[desLHR-A] and sCR1[desLHR-A]sLex to the lung vasculature correlated with the protective effects of these compounds (Table IV). In vitro binding of 125I-sCR1[desLHR-A] and 125 I-sCR1[desLHR-A]sLex to nonactivated and activated endothelial cells The binding of 125I-sCR1[desLHR-A] and 125I-sCR1[desLHRA]sLex to unstimulated and TNF-a-stimulated (4 h) HUVEC was assessed. When unstimulated endothelial cells were incubated with 125 I-sCR1[desLHR-A], very little binding occurred (Fig. 2A). This low level of binding was unaffected by the presence of anti-Eselectin (CL-3), anti-P-selectin (PB1.3), or a class-matched irrelevant IgG1 (MOPC-21). In TNF-a-stimulated HUVEC, there was, likewise, no significant increase in binding of sCR1[desLHR-A] (Fig. 2B). In contrast, the binding patterns of sCR1[desLHRA]sLex were quite different. Although binding of sCR1[desLHRA]sLex to unstimulated HUVECS (Fig. 2C) was about 2.5-fold higher than the binding of sCR1[desLHR-A] to unstimulated cells (Fig. 2A), there was a significantly greater increase in the binding of sCR1[desLHR-A]sLex to the TNF-a-stimulated HUVECS in the absence of Ab (Fig. 2D). This binding was greatly reduced in the presence of anti-E-selectin (CL-3) but not in the presence of either anti-P-selectin or the subclass-matched irrelevant mouse IgG1 (MOPC-21). Thus, in vitro binding of sCR1[desLHR-A]sLex was increased in TNF-a-stimulated HUVECS in a manner that proved to be E-selectin dependent.

Discussion Adhesiveness of neutrophils to the endothelium is a key event in the early inflammatory response. Rapid endothelial expression of P-selectin appears to facilitate early adhesive interactions involving counterreceptors on the neutrophil, such as P-selectin glycoprotein ligand-1 (PSGL-1) and, to a lesser extent, E-selectin ligand-1 (ESL-1). The current report describes a novel strategy that takes advantage of the known interactions between certain selectins (chiefly L- and P-selectins) and sialated, fucosylated oligosaccharides, especially sLex (13, 14). A process that permits incorporation of the sLex moiety into natural N-linked oligosaccharides has been developed. This has the advantage of sLex tetrasaccharide being attached to large, naturally occurring oligosaccharides. In addition to the possibility that sLex-bearing glycoproteins may inhibit the binding of selectins to their counterreceptors, these modified proteins may also become localized in vivo to sites on the activated endothelium because of their content of sLex, hereby allowing complement inhibition to be focused at the site of the activated endothelium. In addition or alternatively, this binding might compete with the ability of neutrophils to bind to their counterreceptors (selectins) on endothelial cells. Ischemic injury of the myocardium (32), of the hind limbs (33), and of lung (34) is in every case associated with participation of selectins. Blocking Ab to P-selectin or the use of sLex tetra- or pentasaccharide has been shown to be protective in several of these models of injury (31, 34). Ischemia-reperfusion injury appears to be associated with up-regulation of endothelial P-selectin, perhaps in part due to complement activation (C5a and/or C5b-9). Accordingly, ischemic models are logical applications for complement inhibitors decorated with sLex. Complement inhibitors would interfere with these pathways by reducing generation of C5a and/or


FIGURE 2. The binding of 125I-sCR1[desLHR-A] to unstimulated HUVECS (A) and to TNF-a-stimulated HUVEC (B). When used, blocking Abs to human E-selectin (CL-3) and to P-selectin (PB1.3) were added at concentrations of 5 mg/ml. MOPC-21 was a subclass-matched control. Stimulation of HUVEC was accomplished by addition of 50 ng/ml human TNF-a at 37°C for 4 h before washing and further additions. The binding of 125I-sCR1[desLHRA]sLex to unstimulated HUVEC (C) and TNF-a-stimulated HUVEC (D) is shown.

4958

sCR1[desLHR-A] was less effective than the sLex-decorated forms of sCR1 and sCR1[desLHR-A]. This correlated with the enhanced binding activities of these compounds to the lung vasculature (Fig. 1 and Table III). Using the IgG immune complex model of lung injury, which is L-selectin-, P-selectin-, and E-selectin-dependent, the ability of sLexdecorated and undecorated forms of sCR1[desLHR-A] was evaluated both for protective effects as well as binding to the lung vasculature. Undeniably, both inhibitors showed protective effects, with the decorated versions being more effective (Table IV). This also correlated with the higher binding of the sLex-decorated form to the lung vasculature (Table V). What is curious is why sCR1[desLHR-A] would have any protective effects in the IgG immune complex model (which would be assumed to be predominately engaging the classical complement pathway). SCR1[desLHR-A], when compared with sCR1, had somewhat reduced blocking activity for the alternative pathway (IC50 values of 37 6 6.2 nM vs 19 6 6.6 nM, respectively). If the developing IgG immune complex response at some point were to engage the alternative pathway due to generation of C3b, this could explain why sCR1[desLHR-A] had protective effects (albeit diminished). Such a possibility is supported by published data (38 – 41). When sCR1 and sCR1[desLHR-A] were evaluated in vitro for their complement inhibiting activities, the IC50 values for inhibition of the classical and alternative pathways were 0.21 nM and 19 nM for sCR1 and 58 nM and 37 nM for sCR1[desLHR-A] (above and Footnote 4). Therefore, if present in sufficient concentrations, sCR1[desLHR-A] would contain the ability to block activation of the classical pathway. Intravenous infusion of decorated or undecorated sCR1[desLHR-A] (at 15 mg/kg body weight) would yield a plasma concentration in the range of 2 mM, well above the IC50 values for inhibition of the classical pathway. As to why the sLex-decorated form of sCR1[desLHR-A] was a more effective inhibitor of injury in the IgG immune complex model of lung injury than was the undecorated form, recent observations that this model is both P-selectin and L-selectin dependent (and also E-selectin dependent (see above)) would be consistent with the ability of sLex containing oligosaccharides to interact with all three selectins. As to why sCR1[des LHRA]sLex but not the sLex-undecorated form binds to TNF-a-stimulated HUVECS in an E-selectin manner, these data are consistent with published evidence that E-selectin recognizes the monomeric sLex moiety and that the strength of binding between E-selectin and monomeric sLex is comparable to the binding interactions between P- or L-selectin and monomeric sLex (42–47). Interactions between P-selectin and its primary counterreceptor, PSGL-1, depend on sLex-decoration of PSGL-1. It is believed that this interaction is higher affinity than that between P-selectin and monomeric sLex because of 1) tyrosine sulfation on PSGL-1 that contributes to binding affinity, 2) the multivalent nature of the sLex moieties on O-linked glycans present on PSGL-1, and 3) the overall tertiary conformation assumed by these glycans. Similar considerations also apply to interactions between Lselectin on rolling neutrophils and PSGL-1 displayed by adherent neutrophils, which is the major counterreceptor recognized by L-selectin that is operative in neutrophil recruitment. The nature of bona fide neutrophil E-selectin ligands is less clear. There is evidence that mono- and polyfucosylated glycolipids are physiological ligands, and there is evidence that PSGL-1 is also a ligand. The strategy to develop complement inhibitors that can be “targeted” to the selectin-expressing activated endothelium is attractive, since this should provide a way to achieve localization of a complement inhibitor along surfaces of the activated endothelium. Collectively, our data suggest that decoration of sCR1 or sCR1[desLHR-A] with sLex enhances their binding to the selectinexpressing vascular endothelium and, in turn, enhances protection


C5b-9, both of which have been shown to induce P-selectin expression on endothelial cells (35–37). In these various models of ischemia-reperfusion injury, neutrophil recruitment seems to be an important event related to full development of injury. The method by which sCR1sLex and sCR1[desLHR-A]sLex are produced is described elsewhere.4 The analysis of oligosaccharides on these molecules confirms that they do indeed possess the sLex moiety. The goal of the current report has been to characterize the biological activity and the efficacy of the sLex-decorated glycoproteins in two selectin-dependent models of acute lung injury. We first assessed the protective functions of the sLex-decorated and undecorated forms of sCR1 and sCR1[desLHR-A] in the P-selectin- and L-selectin-dependent model of acute lung injury, which occurs following systemic activation of complement following i.v. infusion of CVF. In this model, blocking of P-selectin with antiP-selectin Ab (PB1.3) has been shown to reduce neutrophil accumulation and lung vascular injury, as measured by albumin leak, hemorrhage, and MPO content (see above). In the same model, infusion of L-selectin- or P-selectin-Ig chimeric proteins (but not E-selectin-Ig chimeric protein) protected against full development of lung injury (24). Furthermore, i.v. infusion of penta- or tetrasaccharide sLex was protective in this model, these effects being associated with reduced accumulation of neutrophils and development of injury (19). The increased efficacy of sCR1sLex as compared with sCR1 in protecting against CVF-induced lung injury seems to be related to the ability of sCR1sLex to bind to the activated endothelium. By immunohistochemical methods, we were able to demonstrate that sCR1sLex, but not sCR1, was bound to the lung vascular endothelium following systemic activation of complement (Fig. 1). This may imply that endothelial-bound sCR1sLex is more effective than sCR1 (which does not bind) in preventing local complement activation, and, therefore, in inhibiting complement activation, which has been linked to up-regulation of endothelial P-selectin (31). In addition, the binding of sCR1sLex to either endothelial cells or to neutrophils might interfere with the binding interactions between these cells (via endothelial P-selectin and/or neutrophil L-selectin, or both), resulting in diminished accumulation of neutrophils within the lung. The presence of sLex on sCR1, which results in a more protective molecule, appears not to be associated with altered pharmacokinetics, or due to sCR1sLex having superior complement inhibitory activities (see above). Further, although platelets in addition to endothelial cells can be stimulated by complement activation products to express P-selectin (35–37), platelets appear unlikely to be the source of P-selectin in the CVF model of lung injury, since in this model platelet depletion did not reduce the intensity of lung injury or lung vascular expression of P-selectin (31). In the CVF model, we also observed increased efficacy of sCR1[desLHR-A]sLex relative to sCR1[desLHR-A] based on reduced lung permeability and diminished content of MPO. That correlated with greater binding of sCR1[desLHR-A]sLex to the lung vasculature when compared with sCR1[desLHR-A] (Table III). This binding was P-selectin-dependent and was determined by the use of anti-P-selectin. Therefore, the enhanced protective effects of sCR1[desLHR-A]sLex are consistent with the interpretation that sCR1[desLHR-A]sLex has greater binding to the activated endothelium. As expected, deletion of domain A in sCR1 (sCR1[desLHR-A]) as compared with intact sCR1 (19 vs 37; 38 vs 46 nM) had little effect on its inhibitory activity in the alternative pathway, while, at the same time, causing nearly 50-fold reduction in inhibition of the classical pathway (see above). There seems little question that, even in the CVF model of lung injury,


The Journal of Immunology against neutrophil-mediated injury. Whether the enhanced protective effects of sCR1sLex and sCR1[desLHR-A]sLex are due to their increased concentration at sites of the vascular endothelium (thus more effectively inhibiting local complement activation) or the sLex-decorated compounds complete with selectin-dependent binding interactions of neutrophils to the activated endothelium remains to be determined. sCR1sLex and sCR1[desLHR-A]sLex are clearly more effective antiinflammatory agents when compared with the forms lacking sLex. This may suggest a novel strategy for development of antiinflammatory compounds.

Acknowledgments We thank Henry C. Marsh, Jr. for insightful input; Robin Kunkel, Susanne Scesney, and Christopher Honan for technical support; and Beverly Schumann for secretarial assistance.

References

22. Mulligan, M. S., M. J. Polley, R. J. Bayer, M. F. Nunn, J. C. Paulson, and P. A. Ward. 1992. Neutrophil-dependent acute lung injury: requirement for Pselectin (GMP-140). J. Clin. Invest. 90:1600. 23. Mulligan, M. S., M. Miyasaka, T. Tamatani, M. L. Jones, and P. A. Ward. 1994. Requirements for L-selectin in neutrophil-mediated lung injury in rats. J. Immunol. 152:832. 24. Mulligan, M. S., S. R. Watson, C. Fennie, and P. A. Ward. 1993. Protective effects of selectin chimeras in neutrophil-mediated lung injury. J. Immunol. 151: 6410. 25. Mulligan, M. S., J. Varani, M. K. Dame, C. L. Lane, C. W. Smith, D. C. Anderson, and P. A. Ward. 1991. Role of endothelial-leukocyte adhesion molecule 1 (ELAM-1) in neutrophil-mediated lung injury in rats. J. Clin. Invest. 88:13966. 26. Bless, N. M., S. J. Tojo, H. Kawarai, Y. Natsume, A. B. Lentsch, V. Padgaonkar, B. J. Czermak, H. Schmal, H. P. Friedl, and P. A. Ward. 1998. Differing patterns of P-selectin expression in lung injury. Am. J. Pathol. 153:1113. 27. Mulligan, M. S., J. C. Paulson, S. De Frees, Z.-L. Zheng, J. B. Lowe, and P. A. Ward. 1993. Protective effects of oligosaccharides in P-selectin-dependent lung injury. Nature 364:149. 28. Mulligan, M. S., J. B. Lowe, R. D. Larsen, J. Paulson, Z.-L. Zheng, S. DeFrees, K. Maemura, M. Fukuda, and P. A. Ward. 1993. Protective effects of sialylated oligosaccharides in immune complex-induced acute lung injury. J. Exp. Med. 178:623. 29. Simonsen, C. C., and A. D. Levinson. 1983. Isolation and expression of an altered mouse dihydrofolate reductase cDNA. Proc. Natl. Acad. Sci. USA 80:2495. 30. Mulligan, M. S., M. J. Polley, R. J. Bayer, M. F. Nunn, J. C. Paulson, and P. A. Ward. 1992. Neutrophil-dependent acute lung injury: requirement for Pselectin (GMP-140). J. Clin. Invest. 90:1600. 31. Mulligan, M. S., E. Schmid, G. O. Till, T. E. Hugli, H. P. Friedl, R. A. Roth, and P. A. Ward. 1997. C5a-dependent up-regulation in vivo of lung vascular P-selectin. J. Immunol. 158:1857. 32. Weyrich, A. S., X.-L. Ma, D. J. Lefer, K. H. Albertine, and A. M. Lefer. 1993. In vivo neutralization of P-selectin protects feline heart and endothelium in myocardial ischemia and reperfusion injury. J. Clin. Invest. 91:2620. 33. Seekamp, A., M. S. Mulligan, G. O. Till, and P. A. Ward. 1993. Requirements for neutrophil products and L-arginine in ischemia-reperfusion injury. Am. J. Pathol. 142:1217. 34. Eppinger, M. J., M. L. Jones, M. Deeb, S. F. Bolling, and P. A. Ward. 1995. Pattern of injury and the role of neutrophils in reperfusion injury in rat lung. J. Surg. Res. 58:713. 35. Hattori, R., K. K. Hamilton, R. P. McEver, and P. J. Sims. 1989. Complement proteins C5b-9 induce secretion of high molecular weight multimers of endothelial von Willebrand factor and translocation of granule membrane protein GMP140 to the cell surface. J. Biol. Chem. 264:905. 36. Foreman, K. E., A. A. Vaporciyan, B. K. Bonish, M. L. Jones, K. J. Johnson, M. M. Glovsky, S. M. Eddy, and P. A. Ward. 1994. C5a-induced expression of P-selectin in endothelial cells. J. Clin. Invest. 94:1147. 37. Kilgore, K. S., P. A. Ward, and J. S. Warren. 1998. Neutrophil adhesion to human endothelial cells is induced by the membrane attack complex: the roles of Pselectin and platelet activating factor. Inflammation 22:583. 38. Farries, T.C., K. L. Knutzen Steuer, and J. P. Atkinson. 1990. The mechanism of activation of the alternative pathway of complement by cell-bound C4b. Mol. Immunol. 27:1155. 39. Lachmann, P. J., and N. C. Hughes-Jones. 1984. Initiation of complement activation. Springer Semin. Immunopathol. 7:143. 40. Meri, S., V. Koistinen, A. Miettinen, T. Tornroth, and J. T. Seppala. 1992. Activation of the alternative pathway of complement by monoclonal l light chains in membranoproliferative glomerulonephritis. J. Exp. Med. 175:939. 41. Meri, S., and M. Pangburn. 1990. A mechanism of activation of the alternative complement pathway by the classical pathway: protection of C3b from inactivation by covalent attachment to C4b. Eur. J. Immunol. 20:2555. 42. Norman, K. E., G. P. Anderson, H. C. Kolb, K. Ley, and B. Ernst. 1998. Sialyl Lewis(x) (sLe(x)) and an sLe(x) mimetic, CGP69669A, disrupt E-selectin-dependent leukocyte rolling in vivo. Blood 91:475. 43. Stahn, R., H. Schafer, F. Kernchen, and J. Schreiber. 1998. Multivalent sialyl Lewis x ligands of definite structures as inhibitors of E-selectin mediated cell adhesion. Glycobiology 8:311. 44. Welply, J. K., S. Z. Abbas, P. Scudder, J. L. Keene, K. Broschat, S. Casnocha, C. Gorka, C. Steininger, S. C. Howard, and J. J. Schmuke. 1994. Multivalent sialyl-Lex: potent inhibitors of E-selectin-mediated cell adhesion; reagent for staining activated endothelial cells. Glycobiology 4:259. 45. Jacob, G. S., C. Kirmaier, S. Z. Abbas, S. C. Howard, C. N. Steininger, J. K. Welphy, and P. Scudder. 1995. Binding of sialyl Lewis x to E-selectin as measured by fluorescence polarization. Biochemistry 34:1210. 46. Brunk, D. K., D. J. Goetz, and D. A. Hammer. 1996. Sialyl Lewis(x)/E-selectinmediated rolling in a cell-free system. Biophys. J. 71:2902. 47. Lowe, J. B. 1997. Selectin ligands, leukocyte trafficking, and fucosyltransferase genes. Kidney Int. 51:1418. 48. Rittershaus, C. W., L. J. Thomas, D. P. Miller, M. D. Picard, K. M. GeoghaganBarek, S. M. Scesney, L. D. Henry, A. C. Sen, A. M. Bertino, G. Hannig, et al. 1999. Recombinant glycoproteins that inhibit complement activation and also bind the selectin adhesion molecules. J. Biol. Chem. In press.


1. Stevens, J. H., P. O’Hanley, J. M. Shapiro, F. G. Mihm, P. S. Satoh, J. A. Collins, and T. A. Raffin. 1986. Effects of anti-C5a antibodies on the adult respiratory distress syndrome in septic primates. J. Clin. Invest. 77:1812. 2. Smedegard, G., L. Cui, and T. E. Hugli. 1989. Endotoxin-induced shock in the rat. Am. J. Pathol. 135:489. 3. Rinder, C. S., H. M. Rinder, B. R. Smith, J. C. K. Fitch, M. J. Smith, J. B. Tracey, L. A. Matis, S. P. Squinto, and S. A. Rollins. 1995. Blockade of the C5a and C5b-9 generation inhibits leukocyte and platelet activation during extracorporeal circulation. J. Clin. Invest. 96:1564. 4. Mulligan, M. S., E. Schmid, B. Beck-Schimmer, G. O. Till, H. P. Friedl, R. B. Brauer, T. E. Hugli, M. Miyasaka, R. L. Warner, K. J. Johnson, and P. A. Ward. 1996. Requirement and role of C5a in acute lung inflammatory injury in rats. J. Clin. Invest. 98:503. 5. Cochane, C. G., H. J. Muller-Eberhard, and B. S. Aikin. 1970. Depletion of plasma complement in vivo by a protein of cobra venom: its effect on various immunologic reactions. J. Immunol. 105:55. 6. Hill, J. H., and P. A. Ward. 1970. The phlogistic role of C3 leukotactic fragments in myocardial infarcts of rats. J. Exp. Med. 133:885. 7. Mulligan, M. S., C. G. Yeh, A. R. Rudolph, and P. A. Ward. 1992. Protective effects of soluble CR1 in complement- and neutrophil-mediated tissue injury. J. Immunol. 148:1479. 8. Klickstein, L. B., W. W. Wong, J. A. Smith, J. H. Weis, J. G. Wilson, and D. T. Fearon. 1987. Human C3b/C4b receptor (CR1): demonstration of long homologous repeating domains that are composed of the short consensus repeats characteristic of C3/C4 binding proteins. J. Exp. Med. 165:1095. 9. Klickstein, L. B., T. J. Bartow, V. Miletic, L. D. Rabson, J. A. Smith, and D. T. Fearon. 1988. Identification of distinct C3b and C4b recognition sites in the human C3b/C4b receptor (CR1, CD35) by deletion mutagenesis. J. Exp. Med. 168:1699. 10. Yeh, C. G., H. C. Marsh, G. R. Carson, L. Berman, M. F. Concino, S. M. Scesney, R. E. Kuestner, R. Skibbens, K. A. Donahue, and S. H. Ip. 1990. Recombinant soluble human complement receptor type 1 inhibits inflammation in the reversed passive arthus reaction in rats. J. Immunol. 146:250. 11. Weisman, H. F., T. Bartow, M. K. Leppo, H. C. Marsh, G. R. Carson, M. F. Concino, M. P. Boyle, K. H. Roux, M. L. Weisfeldt, and D. T. Fearon. 1990. Soluble human complement receptor type 1: in vivo inhibitor of complement suppressing post-ischemic myocardial inflammation and necrosis. Science 249:146. 12. Scesney, S. M., S. C. Makrides, M. L. Gosselin, P. L. Ford, B. M. Andrews, E. G. Hayman, and H. C. Marsh. 1996. A soluble deletion mutant of the human complement receptor type 1, which lacks the C4b binding site, is a selective inhibitor of the alternative complement pathway. Eur. J. Immunol. 26:1729. 13. Lasky, L. A. 1992. Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science 258:964. 14. Varki, A. 1994. Selectin ligands. Proc. Natl. Acad. Sci. USA 91:7390. 15. Springer, T. A. 1990. Adhesion receptors of the immune system. Nature 346:425. 16. Zimmerman, G. A., S. M. Prescott, and T. M. McIntyre. 1992. Endothelial cell interactions with granulocytes: tethering and signaling molecules. Immunol. Today 13:93. 17. Imhof, B. A., and D. Dunon. 1995. Leukocyte migration and adhesion. Adv. Immunol. 58:345. 18. Rosen, S. D., and C. R. Bertozzi. 1994. The selectins and their ligands. Curr. Opin. Cell. Biol. 6:663. 19. Mulligan, M. S., J. C. Paulson, S. De Frees, Z. L. Zheng, J. B. Lowe, and P. A. Ward. 1993. Protective effects of oligosaccharides in P-selectin-dependent lung injury. Nature 364:149. 20. Campbell, C., and P. Stanley. 1984. The Chinese hamster ovary glycosylation mutants LEC11 and LEC12 express two novel GDP-fucose: N-acetylglucosaminide 3-a-L-fucosyltransferase enzymes. J. Biol. Chem. 259:11208. 21. Stanley, P., and P. H. Atkinson. 1988. The LEC11 Chinese hamster ovary mutant synthesizes N-linked carbohydrates containing sialated, fucosylated lactosamine units. J. Biol. Chem. 263:11374.

4959