Endocytosis of Heparin-Binding Protein (CAP37) Is Essential for the ...

2 downloads 182 Views 93KB Size Report
(CAP37) Is Essential for the Enhancement of. Endocytosis of Heparin-Binding Protein. Hiram C. Polk, Jr. and Frederick N. Miller. Michael Heinzelmann, Andreas ...
Endocytosis of Heparin-Binding Protein (CAP37) Is Essential for the Enhancement of Lipopolysaccharide-Induced TNF-a Production in Human Monocytes1,2 Michael Heinzelmann,3*† Andreas Platz,† Hans Flodgaard,‡ Hiram C. Polk, Jr.,† and Frederick N. Miller* Heparin-binding protein (HBP), also known as CAP37, is a proteolytically inactive serine protease homologue that is released from activated granulocytes. However, HBP is not a biologically inactive molecule but rather a multifunctional protein with properties that include the enhancement of LPS-induced TNF-a production from monocytes. We have previously demonstrated that HBP is internalized in monocytes. In the current study, we hypothesize that HBP is internalized in monocytes via endocytosis, and this internalization is an important mechanism by which HBP enhances LPS-induced TNF-a release. Using whole blood from healthy donors and flow cytometry, we found that colchicine (0.1–10 mM), cytochalasin D (1000 mM), NH4Cl (10 –50 mM), and bafilomycin A1 (0.1–3 mM) significantly reduced the affinity of FITC-HBP for CD14-positive monocytes. Using isolated human monocytes and ELISA, we found that colchicine (0.1 mM), cytochalasin D (30 and 300 mM), NH4Cl (30 mM), and bafilomycin A1 (1 mM) significantly reduced the effect of HBP (10 mg/ml) to enhance LPS (10 ng/ml)-induced TNF-a release after 24 h. These findings demonstrate that internalization of HBP in monocytes is essential for the enhancement of LPS-induced TNF-a release. Transport of HBP to an activating compartment depends on intact F-actin polymerization and endosomal acidification, an important mechanism for endosomal protein sorting and trafficking. The Journal of Immunology, 1999, 162: 4240 – 4245.

H

eparin-binding protein (HBP)4, also known as cationic antimicrobial protein of molecular mass kDa (CAP37) or azurocidin, is a proteolytically inactive serine protease homologue that is released from activated granulocytes. HBP is structurally related to serine proteases in circulating white blood cells, but lacks proteolytic activity because of selective mutations of two of the three essential amino acids in the catalytic triad (1). HBP is not a biologically inactive molecule, but rather is a multifunctional protein that targets monocytes, fibroblasts, endothelial cells, and Gram-negative bacteria (2, 3). HBP has many effects on monocytes, including chemotaxis, increased longevity (2), and the enhancement of LPS-induced TNF-a production (4). LPS is an important component of the outer membrane of all Gram-negative bacteria and has a vital function for bacterial viability. In humans and experimental animals, the presence of LPS signals the presence of Gram-negative bacteria and induces a vig-

*Department of Physiology and Biophysics, and †The Price Institute of Surgical Research, Department of Surgery, University of Louisville, School of Medicine, Louisville, KY 40292; and ‡Health Care Discovery, Novo Nordisk, Novo Alle´, Bagsvaerd, Denmark Received for publication July 22, 1998. Accepted for publication January 4, 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 This work was supported in part by the John W. and Caroline Price Trust, Alliant Community Trust, the Mary and Mason Rudd Endowment Fund of Jewish Hospital (Louisville, KY), the American Heart Association Kentucky Affiliate, and the Centre of Applied Microcirculatory Research, University of Louisville, Louisville, KY. 2 Presented in part at the Eighteenth Annual Meeting of the Surgical Infection Society, New York, NY, April 30-May 2, 1998. 3 Address correspondence and reprint requests to: Dr. Michael Heinzelmann, c/o M. Abby, Editorial office, Department of Surgery, University of Louisville, Louisville, KY 40292. E-mail address: [email protected] 4 Abbreviations used in this paper: HBP, heparin-binding protein; CAP37, cationic antimicrobial protein of molecular mass 37 kDa; FL1, fluorescence measured at 530 nm; PE, phycoerythrin.

Copyright © 1999 by The American Association of Immunologists

orous cellular response, and LPS is believed to be responsible for many of the toxic manifestations of severe Gram-negative sepsis (5). LPS is known to interact with a variety of cell types, including endothelial cells, smooth muscle cells, granulocytes, thrombocytes, and macrophages/monocytes. In monocytes/macrophages, the binding of LPS to CD14 induces the activation of many inducible genes and bioactive substances, such as cytokines (e.g., TNF-a), adhesive proteins, and enzymes that produce low m.w. proinflammatory mediators (6, 7). Collectively, these products upregulate host defense systems with the apparent goal of eliminating the bacterial infection (8, 9). Unfortunately, these same mediators also contribute to the development of septic shock (10, 11). After murine peritonitis is induced by cecal ligation and puncture (12), we have shown that i.p. administration of HBP increases monocyte recruitment into the peritoneum and increases survival. This indicates that HBP may play a critical role during sepsis. Previous studies have established that HBP has a high affinity for human monocytes, but not for the LPS receptor CD14, and that HBP enhances the LPS-induced release of prostaglandin E2 from isolated monocytes (13), suggesting a generalized monocyte activation. In confocal microscopy studies, we have found that HBP is internalized in monocytes within 30 min at 37°C (13). In other studies, we have found that the affinity of HBP for monocytes is reduced by the addition of fucoidan (14). Fucoidan is a substance known to compete for binding sites on the scavenger receptor that mediates endocytosis via the clathrin-coated pathway (15). These results suggest that endocytosis-mediated internalization of HBP in monocytes is a potentially important step for the enhancement of LPS-induced TNF-a release. In the current study, we inhibited endocytosis at different levels and assessed the effect of HBP on LPS-induced TNF-a production from isolated human monocytes. Considerable knowledge about the organization and function of endocytic processes has been accumulated over the past decade, 0022-1767/99/$02.00

The Journal of Immunology but current knowledge of the entire endocytic pathway is still fragmentary (16, 17). Macromolecular entry to the cell is mediated via different pathways: caveolae, nonclathrin-coated vesicles, clathrincoated pits and vesicles, macropinosomes, and phagosomes. The clathrin-coated vesicle is the best-characterized mechanism for entry into the cell. Endocytosis vesicles with membrane proteins, lipids, and solutes deliver their cargo to the early endosome for further processing. From the early endosome, the internalized substrates are directed to late endosomes and lysosomes along the endocytic pathway. Acidification of the early endosomes (pH 6.0 – 6.5) plays a crucial role in the sorting of internalized ligands, receptors, and solutes in intracellular trafficking and proteolytic cleavage (18). Ammonium chloride (NH4Cl) and bafilomycin A1, a specific inactivator of the proton pump vacuolar ATPase (19), inhibit acidification of the early endosome and impair the transport between early and late endosomes. Lysosomes are the site of degradation of obsolete extracellular macromolecules following endocytosis and phagocytosis (20). Some endocytosed receptors are recycled back to the cell surface from the early endosome via a highly tubulated, colchicine-sensitive, recycling compartment. There is now a large body of evidence showing that microtubules play an important role in membrane traffic. In contrast, the role of the actin cytoskeleton in membrane traffic is not completely understood (16). Classic studies of endocytosis have characterized two fates for endocytosed protein: degradation following transport to lysosomes or return to the cell surface, such as occurs with class II MHCrestricted Ag presentation. However, it is now clear that the endocytic pathway is more complex, and that many cells have the capacity to sequester some endocytosed proteins into specialized compartments (16). It is currently unclear whether the internalization of HBP (13) is a process that is needed to enhance LPSinduced TNF-a production in monocytes or merely a pathway that leads to the lysosome and degradation. We hypothesize that HBP is internalized in monocytes via endocytosis, and this internalization is an important mechanism by which HBP enhances LPS-induced TNF-a release from isolated human monocytes. Herein, we show that internalization of HBP in monocytes is essential for the enhancement of LPS-induced TNF-a release, and trafficking of HBP to an activating compartment depends on intact F-actin polymerization and endosomal acidification.

Materials and Methods Preparation of human recombinant HBP HBP was expressed in Spodopterea frugiperda (SF9) cells (Invitrogen, San Diego, CA) and purified as previously described (4). Briefly, we constructed a 770-bp BamHI-HindIII fragment from a human bone marrow cDNA library (Clontech Laboratories, Heidelberg, Germany) by using PCR technology. We then inserted the fragment into the baculovirus transfer vector pBlueBacIII (Invitrogen), which resulted in the transfer plasmid pSX556. SF9 cells were transfected using linear Autographa californica nuclear polyhedrosis virus DNA and transfection plasmid (Invitrogen). We collected the insect cell culture medium 3– 4 days after transfection and purified HBP by glass microfiber filtration (Whatman GF/A), CM-Sepharose cation exchange columns (Pharmacia, Piscataway, NJ), and Sephadex G-25 gel-filtration columns (Pharmacia).

Reagents and mAbs Colchicine, cytochalasin D, NH4Cl, bafilomycin A1, potassium bicarbonate, DMSO, EDTA, trypan blue, and Escherichia coli 0111:B4 LPS were purchased from Sigma (St. Louis, MO). Phycoerythrin (PE)-coupled monoclonal anti-CD14 Ab Mo2 was purchased from Coulter (Hialeah, FL). Enzyme-linked immunosorbent assay was used to measure TNF-a (BioSource, Camarillo, CA).

4241 HBP affinity studies Whole blood from healthy volunteers was collected in acid citrate dextrose Vacutainers (Becton Dickinson, Cockeysville, MD) at room temperature. The affinity studies were conducted in a final volume of 100 ml. Whole blood was preincubated for 60 min with increasing concentrations of colchicine, cytochalasin D, NH4Cl, and bafilomycin A1. FITC-HBP (10 mg/ ml, final concentration) was added, and the samples were incubated for 60 min at 37°C with 5% CO2. At the end of the incubation with FITC-HBP, Mo2-PE (500 ng in 5 ml) was added, and the samples were incubated for 25 min at 4°C. Subsequently, erythrocytes were removed by hypotonic lysis (150 mM NH4Cl, 12 mM potassium bicarbonate, 0.1 mM EDTA), samples were washed twice with FTA hemagglutination buffer with 0.1% sodium azide (Becton Dickinson), fixed in 1% paraformaldehyde, and analyzed by flow cytometry.

Flow cytometry A FACScan from Becton Dickinson with an argon laser (488 nm) was used to assess FITC-HBP fluorescence on CD14-positive cells. Fluorescence values derived from FITC-HBP were measured at 530 nm (FL1). CD14positive monocytes were gated based on the combination of fluorescence derived from the anti-CD14 Ab Mo2-PE (measured at 580 nm) and the sideways light scatter. A total of 4000 –5000 CD14-positive monocytes were analyzed per sample, and acquired data were processed with Cellquest version 1.2 software (Becton Dickinson). The FL1 fluorescence distribution was displayed as a single histogram. The percentage of FL1 fluorescent cells and the mean fluorescence intensity were determined in each case.

Monocyte isolation and culture Human monocytes were isolated by dextran sedimentation and density gradient centrifugation (21). Briefly, whole blood was collected in EDTA Vacutainers, and 1 part of 6% dextran-500 in 0.9% saline (w/v) (Sigma) was added to 10 parts of EDTA blood. Leukocyte-rich plasma was harvested after 45 min of sedimentation and layered on top of 3 ml of 1-StepMonocyte (1068 gradient; Accurate Scientific, Westbury, NY). The gradient was centrifuged at 600 3 g for 15 min at room temperature. The upper layer consisted of plasma and was discarded. The middle layer contained the monocytes and was harvested and washed twice with a washing solution that contained 0.9% saline, 0.13% EDTA (Sigma), and 1% FCS (BioWhittaker, Walkerville, MD). The cell suspension was centrifuged for 7 min at 600 3 g and eventually resuspended in culture medium. Culture media (RPMI 1640 with glutamine; Sigma) was supplemented with 1% antibiotics (100 mg/ml streptomycin, 100 U/ml penicillin; BioWhittaker) and 1% antimycotics (0.25 mg/ml, amphotericin B; BioWhittaker). The cells were counted with a hemocytometer, and the percentage of CD14positive monocytes was assessed by flow cytometry. A total of 2 3 105 cells in 250 ml of supplemented culture medium were added to each well (96-well plate; Costar, Cambridge, MA) and incubated at 37°C with 5% CO2. Cells were pretreated for 60 min with colchicine (0.1 mM, final concentration), cytochalasin D (30 and 300 mM, final concentration), NH4Cl (100 mg/ml, final concentration), bafilomycin A1 (1 mM, final concentration) and stimulated for 24 h with LPS (10 ng/ml, final concentration), HBP (10 mg/ml, final concentration), or a combination of LPS 1 HBP.

Phagocytosis assay Whole blood was obtained from healthy volunteers and collected in acid citrate dextrose Vacutainers. A modification of a previously described method (22, 23) was used to assess phagocytosis in granulocytes and monocytes. Whole blood (90 ml) was incubated with different doses of cytochalasin D (10 ml; final concentration, 30 mM-3 mM) for 60 min before the addition of Bodipy-E. coli (Molecular Probes, Eugene, OR). The number of neutrophils from each donor was assessed with an automated cell counter (Coulter) and confirmed with a hemocytometer. The number of E. coli was adjusted to a ratio of five bacteria per one neutrophil. After 30 min, the phagocytosis process was interrupted by the addition of 4°C-cold lysing reagent (150 mM NH4Cl, 12 mM potassium bicarbonate, 0.1 mM EDTA), and samples were kept at 4°C for 6 min to lyse erythrocytes. Samples were washed twice with FTA-azide (Becton Dickinson), fixed in 1% paraformaldehyde, and analyzed by flow cytometry for fluorescence of Bodipy (530 nm, FL1). To quench adherent Bodipy E. coli, trypan blue (Sigma) was added at a final concentration of 5 mM to each tube before the samples were acquired (24, 25).

4242

FIGURE 1. Effect of colchicine on FITC-HBP affinity for CD14-positive monocytes. Whole blood from healthy donors was pretreated with increasing concentrations of colchicine (0.1, 0.3, 1, 3, and 10 mM) for 60 min and then incubated for 60 min with FITC-labeled HBP (10 mg/ml) or control FITC-IgG (10 mg/ml). CD14-positive monocytes were tagged with the monoclonal anti-CD14 Ab Mo2-PE. Samples were analyzed by flow cytometry. A, The mean fluorescence intensity in CD14-positive monocytes with increasing concentrations of colchicine. B, An overlay of representative histograms demonstrating that colchicine induces a shift in FITC-HBP fluorescence. The peaks (from left to right) are: negative FITCIgG control, FITC-HBP in 10 mM colchicine, in 0.1 mM colchicine, and in saline. Values are mean 6 SEM (n 5 5 donors). p, p , 0.05 compared with saline by ANOVA and Fisher’s probable least-squares difference posthoc test.

Statistical analysis Statistical significance was determined with ANOVA and Fisher’s probable least-squares difference analysis (Statview 4.5; Abacus Concepts, Berkeley, CA) to compare data between multiple groups. A p value ,0.05 was considered significant.

Results Effect of colchicine In the first experiments, we blocked microtubule assembly with colchicine (26) and measured the effect of colchicine on HBP internalization and monocyte activation. Pretreatment with colchicine significantly reduced FITC-HBP fluorescence in CD14-positive monocytes (Fig. 1, A and B). Fluorescence values decreased from 220 6 7 in the saline-treated group to 42 6 2 at the highest concentration of colchicine (10 mM).

ENDOCYTOSIS OF HEPARIN-BINDING PROTEIN (CAP37)

FIGURE 3. Effect of cytochalasin D on FITC-HBP affinity for CD14positive monocytes (A) and on phagocytosis of E. coli (B). Whole blood was pretreated for 60 min with increasing concentrations of cytochalasin D (30, 100, 300, and 1000 M (black bars)), carrier control, i.e., NaCl 0.9% (white bar), or DMSO in equimolar concentration. A, The effect of cytochalasin D on FITC-HBP affinity (10 mg/ml, 60 min incubation) for CD14-positive monocytes. Values are mean 6 SEM (n 5 5 donors). p, p , 0.05 by ANOVA and Fisher’s PLSD posthoc test: Cytochalasin D (1000 mM, black bars) compared with the corresponding DMSO concentration (1%, gray bars). B, The effect of cytochalasin D (black bars), corresponding DMSO concentrations (gray bars), or saline (white bar) on phagocytosis of Bodipy-labeled E. coli by monocytes after 30 min. The mean fluorescence intensity of phagocytosed E. coli was measured after quenching of extracellular bacteria with trypan blue (24, 25). Values are mean 6 SEM (n 5 5 donors). p, p , 0.05 between the cytochalasin D and the DMSO control (unpaired t test). These results demonstrate that cytochalasin D concentration $30 mM were sufficient to block F-actin polymerization.

In the saline control group, LPS (10 ng/ml) increased the release of TNF-a from isolated monocytes, and HBP significantly enhanced this LPS-induced production of TNF-a (Fig. 2). This confirmed previous reports by our laboratory (13) and others (4). In comparison with the control group pretreated with saline, the release of TNF-a from monocytes pretreated with colchicine was markedly reduced after stimulation with LPS (10 ng/ml) or a combination of HBP 1 LPS. LPS still induced a significant ( p 5 0.0003) increase in TNF-a release in the colchicine-treated group (190 6 19 pg/ml) when compared with saline (30 6 9 pg/ml), but HBP did not enhance the LPS-induced release of TNF-a from monocytes. Intact microtubule formation may also be necessary to release TNF-a from monoctyes. To rule out that TNF-a had not been produced and “trapped” within the colchicine-treated monocytes, we conducted a separate experiment. We lysed the adherent monocytes/macrophages used in our current study by adding 150 ml of ultrapure water and then freeze-thawing the samples. The lysate was analyzed for TNF-a by ELISA and showed low TNF-a values (83 6 22 to 114 6 29 pg/ml), with no difference between the stimulated and nonstimulated samples in saline control or the colchicine-treated groups. Effect of cytochalasin D

FIGURE 2. Effect of colchicine on TNF-a production from isolated monocytes. Human monocytes were isolated by density gradient centrifugation (21). After pretreatment with colchicine (0.1 mM) or saline for 60 min, cells were incubated for 24 h with saline, HBP (10 mg/ml), LPS (10 ng/ml), or a combination of HBP 1 LPS. Values are mean 6 SEM (n 5 5 donors). p , 0.05 by ANOVA and Fisher’s PLSD posthoc test. Significant differences within the saline control or colchicine groups: p, vs saline; †, vs saline, HBP, and LPS; ‡, vs saline and HBP. Significant differences between the control and the colchicine groups: §, vs LPS; ¶, vs HBP 1 LPS.

In the second set of experiments, we assessed the effect of cytochalasin D on HBP internalization and monocyte activation. Cytochalasin D blocks actin filament elongation by binding to high affinity sites that are associated with F-actin (27). Cytochalasin D concentrations of 30, 100, 300, and 1000 mM were used to test whether FITC-HBP fluorescence would decrease in CD14-positive monocytes (Fig. 3A). Fluorescence values in the saline group (186 6 24) were not different compared with cytochalasin D at 30 mM (220 6 12), 100 mM (194 6 11), or 300 mM (190 6 6). However, cytochalasin D at the highest concentration of 1000 mM significantly ( p , 0.05) reduced FITC-HBP fluorescence (164 6

The Journal of Immunology

FIGURE 4. Effect of cytochalasin D on TNF-a production from isolated monocytes. Human monocytes were isolated by density gradient centrifugation (21). After pretreatment with cytochalasin D (30 mM) or 0.03% DMSO control for 60 min, cells were incubated for 24 h with saline, HBP (10 mg/ml), LPS (10 ng/ml), or a combination of HBP 1 LPS. Values are mean 6 SEM (n 5 5 donors). p , 0.05 by ANOVA and Fisher’s PLSD posthoc test. Significant differences within the DMSO control or colchicine groups: p, vs saline and HBP; †, vs saline, HBP, and LPS. Significant differences between the control and cytochalasin-D groups: ¶, vs HBP 1 LPS.

4) when compared with the equimolar DMSO control group (219 6 5). These results raised the possibility that the lower concentrations of cytochalasin D might not have affected F-actin polymerization in our experimental groups. Therefore, we tested the efficiency of cytochalasin D in a phagocytosis assay, because phagocytosis depends more on F-actin polymerization than does endocytosis (16). Whole blood was incubated with cytochalasin D (30 –3000 mM) and Bodipy-labeled E. coli, and phagocytosis was assessed by flow cytometry. It was important to measure ingested bacteria only; therefore, we quenched extracellular bacteria with trypan blue (24, 25). Our results (Fig. 3B) showed a dose-dependent inhibition of phagocytosis by cytochalasin D and further demonstrated that even the lowest dose (30 mM) significantly reduced phagocytosis in monocytes when compared with the equimolar DMSO concentrations (Fig. 3B). Cytochalasin D (30 mM) also significantly reduced phagocytosis in granulocytes (data not shown) to confirm inhibition of F-actin polymerization at the lowest cytochalasin D concentration tested. Pretreatment with 30 mM of cytochalasin D (Fig. 4) significantly ( p , 0.001) prevented the effect of HBP to enhance the LPSinduced release of TNF-a (1775 6 554 pg/ml for HBP 1 LPS in cytochalasin D compared with 5511 6 1029 pg/ml for HBP 1 LPS in DMSO control). However, this 30 mM concentration of cytochalasin D did not reduce ( p 5 0.513) the TNF-a release produced by LPS alone (2192 6 754 pg/ml for LPS in cytochalasin D compared with 2702 6 625 pg/ml for LPS in DMSO control). Even though the internalization of HBP is not inhibited by cytochalasin D, as these results indicate (Fig. 3A), intracellular trafficking after internalization is necessary for HBP to enhance the LPS-induced release of TNF-a because F-actin acts only inside the cell (Fig. 4). Effect of NH4Cl In the third set of experiments, we used NH4Cl to prevent endosomal acidification (28, 29). This selective alkalization of endosomes by NH4Cl has been reported to inhibit protein sorting and trafficking and to inhibit intracellular dissociation of receptor-ligand complexes (18). Pretreatment with NH4Cl significantly ( p , 0.05) reduced FITC-HBP fluorescence in CD14-positive cells (Fig.

4243

FIGURE 5. Effect of NH4Cl on FITC-HBP affinity for CD14-positive monocytes (A) and on TNF-a production from isolated monocytes (B). A, Whole blood from healthy donors was pretreated for 60 min with increasing concentrations of NH4Cl (10, 30, and 50 mM) and then incubated for 60 min with FITC-labeled HBP (10 mg/ml). CD14-positive monocytes were tagged with the monoclonal anti-CD14 Ab Mo2-PE. Mean FITCHBP fluorescence intensity in CD14-positive monocytes was measured by flow cytometry. Values are mean 6 SEM (n 5 5 donors). p, p , 0.05 compared with saline by ANOVA and Fisher’s PLSD posthoc test. B, Human monocytes were isolated by density gradient centrifugation (21). After pretreatment with NH4Cl (30 mM) or saline for 60 min, cells were incubated for 24 h with saline (S, white bars); HBP (10 mg/ml) (H, gray bars); LPS (10 ng/ml) (L, hatched bars); or a combination of HBP with LPS (H1L, black bars). Values are mean 6 SEM (n 5 5 donors). p , 0.05 by ANOVA and Fisher’s PLSD posthoc test. Significant differences within the control or NH4Cl groups: p, vs saline; †, vs saline, HBP, and LPS. Significant differences between the control and the NH4Cl groups: §, vs LPS; ¶, vs HBP 1 LPS

5A). Fluorescence values were 255 6 10 in the saline control group compared with 149 6 20, 97 6 17, and 64 6 6 in the NH4Cl-treated group (10 mM, 30 mM, and 50 mM, respectively). Pretreatment with 30 mM of NH4Cl (Fig. 5B) reduced ( p 5 0.04) the stimulatory effect of LPS to release TNF-a from isolated monocytes (1254 6 154 pg/ml for LPS in NH4Cl compared with 1758 6 163 pg/ml for LPS in saline). NH4Cl also significantly ( p 5 0.0005) reduced the release of TNF-a in the HBP 1 LPS group (2012 6 191 pg/ml for HBP 1 LPS in NH4Cl compared with 2937 6 355 pg/ml for HBP 1 LPS in saline). However, 30 mM of NH4Cl did not block the effect of HBP to enhance the LPS-induced release of TNF-a from isolated monocytes. Combined, these data indicate that NH4Cl inhibits intracellular trafficking of HBP, and that the effects of HBP to enhance the LPS-induced TNF-a release depends, in part, on NH4Cl-sensitive, but also non-NH4Cl-sensitive pathways. Effect of bafilomycin A1 In the fourth set of experiments, we specifically blocked the vacuolar type ATPase (19) with bafilomycin A1 to inhibit endosomal acidification (30) and to reduce the delivery of internalized molecules from mature multivesicular endosomes to lysosomes (28). Pretreatment with bafilomycin A1 significantly reduced FITC-HBP fluorescence in CD14-positive cells in a concentration-dependent manner (Fig. 6A). Fluorescence values at the lowest concentration of bafilomycin A1 (0.1 mM in 0.03% DMSO) were 210 6 7 for the bafilomycin A1treated group compared with 258 6 10 for the DMSO control group. At the highest concentration of bafilomycin A1 (3 mM in 0.9% DMSO), FITC-HBP fluorescence values were 79 6 4 for the bafilomycin A1-treated group compared with 218 6 10 for the DMSO control group. Interestingly, DMSO alone reduced FITC-HBP fluorescence somewhat compared with values for saline (250 6 12), but this was statistically significant only at the highest DMSO concentration (0.9% DMSO).

4244

FIGURE 6. Effect of bafilomycin A1 on FITC-HBP affinity for CD14positive monocytes (A) and on TNF-a production from isolated monocytes (B). A, Whole blood from healthy donors was pretreated for 60 min with saline (white bar), increasing concentrations of bafilomycin A1 (0.1, 0.3, 1, and 3 mM; black bars), or corresponding DMSO concentrations (gray bars), and then incubated for 60 min with FITC-labeled HBP (10 mg/ml). CD14-positive monocytes were tagged with the monoclonal anti-CD14 Ab Mo2-PE. Mean FITC-HBP fluorescence intensity in CD14-positive monocytes was measured by flow cytometry. Values are mean 6 SEM (n 5 5 donors). p , 0.05 by ANOVA and Fisher’s PLSD posthoc test. p, Compared with corresponding DMSO control; †, compared with saline. B, Human monocytes were isolated by density gradient centrifugation (21). After pretreatment with bafilomycin A1 (1 mM) or 0.1% DMSO control for 60 min, cells were incubated for 24 h with saline (S, white bars); HBP (10 mg/ml) (H, gray bars); LPS (10 ng/ml) (L, hatched bars); or a combination of HBP with LPS (H1L, black bars). Values are mean 6 SEM (n 5 5 donors). p , 0.05 by ANOVA and Fisher’s PLSD posthoc test. Significant differences within the DMSO control or bafilomycin A1 groups: p, vs saline; †, vs saline, HBP, and LPS. Significant differences between the control and bafilomycin A1 groups: ¶, vs HBP 1 LPS.

Pretreatment of isolated monocytes with 1 mM of bafilomycin A1 (Fig. 6B) significantly ( p 5 0.0001) reduced the ability of HBP to enhance the LPS-induced production of TNF-a (2704 6 477 pg/ml for HBP 1 LPS in bafilomycin A1 compared with 4238 6 300 pg/ml HBP 1 LPS in DMSO control). In both the DMSO control and bafilomycin A1-treated groups, HBP alone induced the release of TNF-a, confirming a previously established variability in HBP responsiveness (13), which could be similar to the genetically determined variability of LPS-induced TNF-a production (31). However, HBP still enhanced the LPS-induced release of TNF-a (Fig. 6B) in the bafilomycin A1-treated group (2704 6 477 pg/ml for HBP 1 LPS in bafilomycin A1 compared with 1788 6 217 pg/ml for LPS in bafilomycin A1). More importantly, bafilomycin A1 had no effect on the ability of LPS alone to stimulate the monocyte release of TNF-a (1549 6 197 pg/ml for LPS in DMSO control compared with 1788 6 217 pg/ml for LPS in bafilomycin A1). These data demonstrate that endosomal acidification is an important step for the effect of HBP to enhance the LPS-induced release of TNF-a from isolated monocytes.

Discussion HBP is a multifunctional protein released from activated neutrophils. We and others (4, 13) have previously shown that HBP enhances the LPS-induced TNF-a release from human monocytes. In the current studies, we further established that the internalization of HBP in human monocytes is an important mechanism for the effect of HBP to enhance LPS-induced TNF-a release. We conclude that HBP must enter the cell to generate the enhancement of the LPS-induced release of TNF-a from isolated monocytes. We demonstrated this by using four substances to inhibit endocytosis at different levels of intracellular trafficking: colchicine, cytochalasin D, NH4Cl, and bafilomycin A1. Intracellular F-actin

ENDOCYTOSIS OF HEPARIN-BINDING PROTEIN (CAP37) polymerization is a critical step for the effect of HBP to enhance this LPS-induced TNF-a release. Colchicine is a potent drug that interferes with microtubule assembly, both in vivo and in vitro (26). Colchicine poisons microtubule assembly by initially binding to the soluble 6S dimer and then subsequently attaching to the end of the growing microtubule as a colchicine-dimer complex during the normal process of microtubule assembly. This integration of colchicine aborts further polymerization and microtubule assembly (26). In our experiments, colchicine concentrations ranging from 0.1 mM (26) to 10 mM (32) showed that HBP internalization was dependent on intact microtubular assembly. Our study also demonstrated that colchicine blocked the effect of HBP to enhance LPS-induced TNF-a release. Furthermore, intact microtubular assembly was important for LPS-induced release of TNF-a from monocytes. This effect of colchicine to reduce TNF-a release was not surprising because microtubules are involved in many intracellular transport mechanisms, including exocytosis of macromolecules. Treatment with colchicine led to a 90% reduction of TNF-a release, but interestingly, LPS still activated monocytes to produce six to seven times more TNF-a than saline. These data suggest that colchicine did not completely abrogate the ability of monocytes to release TNF-a. Cytochalasins are a family of substances that block actin-filament elongation by binding to high affinity sites that are associated with F-actin (27). In the experiments of Flanagan and Lin (27), a small amount of filamentous F-actin was added to a solution of globular G-actin assay. The resulting rapid polymerization was blocked by cytochalasin D at concentrations of 100 nM or more. Other investigators (32) who studied F-actin polymerization in cells used 2000-fold higher concentrations of cytochalasin D (200 mM). We used cytochalasin D concentrations between 30 mM and 1000 mM to test for a decrease in FITC-HBP fluorescence from CD14-positive monocytes. Our 30-mM concentration of cytochalasin D did not inhibit HBP fluorescence from monocytes, indicating that internalization of HBP was not reduced by these low concentrations of cytochalasin D. However, this 30-mM concentration of cytochalasin D did inhibit F-actin polymerization as shown in the phagocytosis assay and abrogated the effect of HBP to enhance the LPS-induced TNF-a release from monocytes. More importantly, 30 mM of cytochalasin D did not reduce the release of TNF-a produced by LPS alone, indicating that LPS-induced cell signaling, and the HBP enhancement of that response occurs by different mechanisms. These results demonstrate that there is an F-actin-dependent trafficking of HBP, which is necessary to enhance LPS-induced TNF-a release from monocytes. In our study, we used two treatments for endosomal alkalization: NH4Cl and bafilomycin A1. Endosomal alkalization alters the kinetics of endocytic uptake. For FITC-albumin endocytosis, this kinetic alteration has been shown by Gekle et al. (18) in intact proximal tubule-derived opossum kidney cells. These authors have demonstrated that endosomal pH is an important determinant for the kinetics of receptor-mediated endocytic uptake of albumin in the proximal tubule but not for fluid-phase endocytosis. Hence, endosomal alkalization disturbs intracellular ligand handling and receptor trafficking, reducing endocytic capacity and affinity. NH4Cl has been used by many investigators to study endocytosis (28, 29). Gekle et al. (18) demonstrated that during prolonged exposure of cells to NH4Cl, endosomal pH was elevated, but cytoplasmic pH was not significantly different from control. This selective alkalization of endosomes by NH4Cl is thought to inhibit protein sorting and trafficking and to inhibit intracellular dissociation of receptor-ligand complexes. This would decrease the production and recycling of free receptors for continuation of receptor-mediated endocytosis (33).

The Journal of Immunology Bafilomycin A1 is a specific inhibitor of the vacuolar type ATPase (19) and is structurally related to macrolide antibiotics that are isolated from Streptomyces. Bafilomycin A1 inhibits acidification and protein degradation in late endosomes and lysosomes (30), and thereby reduces the delivery of internalized molecules from mature multivesicular endosomes to lysosomes (28). NH4Cl produced a modest, but significant reduction in the effect of HBP to enhance LPS-induced TNF-a release, indicating a role for endosomal acidification in this process. Our results from the bafilomycin A1 experiments showed that specific blockade of endosomal acidification reduced the internalization of HBP as well as reduced the effect of HBP to enhance LPS-induced TNF-a production. Moreover, bafilomycin A1 did not reduce LPS-induced TNF-a release. This suggests that endosomal acidification is not a mechanism used for LPS-induced TNF-a release from monocytes. However, the physiological acidification of the early endosome is important in protein sorting, protein trafficking, and receptor recycling (18, 33). Therefore, we propose that bafilomycin A1 reduced the effect of HBP to enhance LPS-induced TNF-a release by altering the intracellular trafficking and delivery of HBP to an activating compartment. The NH4Cl and bafilomycin A1 data (Figs. 5 and 6) show that HBP internalization is only partly dependent on endosomal acidification and that altered endosomal acidification significantly reduces, but does not completely abrogate, the effect of HBP to increase LPS-induced TNF-a release. These data indicate that endosomal acidification is not the only determining factor, suggesting that other factors (e.g., cellular motor elements) are equal or even more important for this process to occur. The colchicine experiments (Fig. 2.) demonstrate that microtubule assembly modulates the effect of HBP to enhance the LPS response and that microtubular assembly is very important for the effect of LPS alone. However, with cytochalasin D, internalization of HBP is dissociated from the effect of HBP to enhance the LPS response (Figs. 3 and 4). Cytochalasin D did not prevent internalization, but it did abrogate the effect of HBP to enhance LPS-induced TNF-a release. This occurred without alteration of the response of LPS alone. Together, our data demonstrate that the magnitude of inhibition of internalization is not correlated with the reduction of LPS-induced TNF-a response. Pugin and colleagues (34) recently demonstrated that cytochalasin D inhibited both LPS and CD14 internalization but did not prevent LPS-dependent activation, confirming our results and again indicating that these two processes are also dissociated. In summary, our experiments with different inhibitors of endocytosis provide evidence that: 1) internalization of HBP in monocytes is essential for HBP to enhance LPS-induced TNF-a release; 2) the trafficking of HBP to an activating compartment depends on intact F-actin polymerization and endosomal acidification; and 3) the mechanisms by which HBP enhances LPS-induced TNF-a release are different from the mechanisms induced by LPS alone.

Acknowledgments We thank Mrs. L. E. Gordon and Mrs. A. J. Roll for excellent technical assistance.

References 1. Flodgaard, H., E. Ostergaard, S. Bayne, A. Svendsen, J. Thomsen, M. Engels, and A. Wollmer. 1991. Covalent structure of two novel neutrophile leucocyte-derived proteins of porcine and human origin: neutrophile elastase homologues with strong monocyte and fibroblast chemotactic activities. Eur. J. Biochem. 197:535. 2. Pereira, H. A. 1995. CAP37, a neutrophil-derived multifunctional inflammatory mediator. J. Leukocyte Biol. 57:805. 3. Pereira, H. A. 1996. CAP37, a neutrophil granule-derived protein stimulates protein kinase C activity in endothelial cells. J. Leukocyte Biol. 60:415.

4245 4. Rasmussen, P. B., S. Bjorn, S. Hastrup, P. F. Nielsen, K. Norris, L. Thim, F. C. Wiberg, and H. Flodgaard. 1996. Characterization of recombinant human HBP/CAP37/azurocidin, a pleiotropic mediator of inflammation-enhancing LPSinduced cytokine release from monocytes. FEBS Lett. 390:109. 5. Glauser, M. P., G. Zanetti, J. D. Baumgartner, and J. Cohen. 1991. Septic shock: pathogenesis. Lancet 338:732. 6. Rietschel, E. T., T. Kirikae, F. U. Schade, U. Mamat, G. Schmidt, H. Loppnow, A. J. Ulmer, U. Zahringer, U. Seydel, and F. Di Padova. 1994. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J. 8:217. 7. Ulevitch, R. J., and P. S. Tobias. 1995. Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin. Annu. Rev. Immunol. 13:437. 8. Polk, H. C., Jr. 1986. The enhancement of host defenses against infection: search for the Holy Grail? Surgery 99:1. 9. Cheadle, W. G., K. Hanasawa, R. N. Gallinaro, T. Nimmanwudipong, M. Kodama, and H. C. Polk, Jr. 1991. Endotoxin filtration and immune stimulation improve survival from gram-negative sepsis. Surgery 110:785. 10. Cheadle, W. G., M. Mercer-Jones, M. Heinzelmann, and H. C. Polk, Jr. 1996. Sepsis and septic complications in the surgical patient: who is at risk? Shock 1(Suppl.):6. 11. Polk, H. C., Jr., M. Heinzelmann, M. A. Mercer Jones, M. A. Malangoni, and W. G. Cheadle. 1997. Pneumonia in the surgical patient. Curr. Probl. Surg. 34:117. 12. Mercer-Jones, M., M. Heinzelmann, J. Peyton, M. Cook, H. Flodgaard, and W. G. Cheadle. 1996. Monocyte recruitment increases survival in fecal peritonitis. Surg. Forum 47:106. 13. Heinzelmann, M., M. Mercer-Jones, H. Flodgaard, and F. N. Miller. 1998. Heparin binding protein (CAP37) is internalized in monocytes and increases lipopolysaccharide-induced monocyte activation. J. Immunol. 160:5530. 14. Heinzelmann, M., H. C. Polk, Jr., and F. N. Miller. 1998. Modulation of lipopolysaccharide-induced monocyte activation by heparin-binding Protein and fucoidan. Infect. Immun. 66:5842. 15. Krieger, M., and J. Herz. 1994. Structures and functions of multiligand lipoprotein receptors: macrophage scavenger receptors and LDL receptor-related protein (LRP). Annu. Rev. Biochem. 63:601. 16. Robinson, M. S., C. Watts, and M. Zerial. 1996. Membrane dynamics in endocytosis. Cell 84:13. 17. Riezman, H., P. G. Woodman, G. van Meer, and M. Marsh. 1997. Molecular mechanisms of endocytosis. Cell 91:731. 18. Gekle, M., S. Mildenberger, R. Freudinger, and S. Silbernagl. 1995. Endosomal alkalinization reduces Jmax and Km of albumin receptor-mediated endocytosis in OK cells. Am. J. Physiol. 268:F899. 19. Bowman, E. J., A. Siebers, and K. Altendorf. 1988. Bafilomycins: a class of inhibitors of membrane ATPases from microorganisms, animal cells, and plant cells. Proc. Natl. Acad. Sci. USA 85:7972. 20. Hunziker, W., and H. J. Geuze. 1996. Intracellular trafficking of lysosomal membrane proteins. BioEssays 18:379. 21. Boyum, A. 1968. A one-stage procedure for isolation of granulocytes and lymphocytes from human blood: general sedimentation properties of white blood cells in a 1g gravity field. Scand. J. Clin. Lab. Invest. 97:51. 22. Perticarari, S., G. Presani, and E. Banfi. 1994. A new flow cytometric assay for the evaluation of phagocytosis and the oxidative burst in whole blood. J. Immunol. Methods 170:117. 23. White Owen, C., J. W. Alexander, R. M. Sramkoski, and G. F. Babcock. 1992. Rapid whole-blood microassay using flow cytometry for measuring neutrophil phagocytosis. J. Clin. Microbiol. 30:2071. 24. Hed, J. 1986. Methods for distinguishing ingested from adhering particles. Methods Enzymol. 132:198. 25. Szollosi, J., L. Tron, S. Damjanovich, S. H. Helliwell, J. D. Arndt, and T. M. Jovin. 1984. Fluorescence energy transfer measurements on cell surfaces: a critical comparison of steady-state fluorimetric and flow cytometric methods. Cytometry 5:210. 26. Margolis, R. L., and L. Wilson. 1977. Addition of colchicine-tubulin complex to microtubule ends: the mechanism of substoichiometric colchicine poisoning. Proc. Natl. Acad. Sci. USA 74:3466. 27. Flanagan, M. D., and S. Lin. 1980. Cytochalasins block actin filament elongation by binding to high affinity sites associated with F-actin. J. Biol. Chem. 255:835. 28. Pless, D. D., and R. B. Wellner. 1996. In vitro fusion of endocytic vesicles: effects of reagents that alter endosomal pH. J. Cell. Biochem. 62:27. 29. Venisse, A., J. J. Fournie, and G. Puzo. 1995. Mannosylated lipoarabinomannan interacts with phagocytes. Eur. J. Biochem. 231:440. 30. Yoshimori, T., A. Yamamoto, Y. Moriyama, M. Futai, and Y. Tashiro. 1991. Bafilomycin A1, a specific inhibitor of vacuolar-type H(1)-ATPase, inhibits acidification and protein degradation in lysosomes of cultured cells. J. Biol. Chem. 266:17707. 31. Garcia Merino, A., C. A. Alper, K. Usuku, D. Marcus Bagley, R. Lincoln, Z. Awdeh, E. J. Yunis, G. S. Eisenbarth, S. J. Brink, and S. L. Hauser. 1996. Tumor necrosis factor (TNF) microsatellite haplotypes in relation to extended haplotypes, susceptibility to diseases associated with the major histocompatibility complex and TNF secretion. Hum. Immunol. 50:11. 32. Chazaud, B., M. P. Muriel, C. Bauvy, P. Codogno, M. Aubery, and M. Decastel. 1994. Requirement of either the NH4Cl-sensitive or the cytochalasin D-sensitive pathway for ricin toxicity depends upon the enterocytic state of differentiation of HT-29 cells. Eur. J. Cell Biol. 64:15. 33. Tietze, C., P. Schlesinger, and P. Stahl. 1982. Mannose-specific endocytosis receptor of alveolar macrophages: demonstration of two functionally distinct intracellular pools of receptor and their roles in receptor recycling. J. Cell Biol. 92:417. 34. Poussin, C., M. Foti, J. L. Carpentier, and J. Pugin. 1998. CD14-dependent endotoxin internalization via a macropinocytic pathway. J. Biol. Chem. 273:20285.