| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print March 30, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Montana State University, Department of Microbiology, Bozeman, Montana, USA
1 Correspondence: Montana State University, Department of Microbiology, 109 Lewis Hall, Bozeman, MT 59715, USA. E-mail: umbaj{at}montana.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
15% of the total cellular hCAP-18 recovered from culture media, and 10% and 75%, respectively, codistributing with plasma membrane alkaline phosphatase and specific granule LF. Surface association of hCAP-18 on fMLF-stimulated neutrophils was confirmed by immunofluorescence microscopy and flow cytometry analysis, which also suggested a significant up-regulation of surface hCAP-18 on cytochalasin B-pretreated, fully degranulated neutrophils. hCAP-18 surface association was labile to 10 mM NaOH treatment but resistant to 1 M NaCl and also partitioned into the detergent phase following Triton X-114 solubilization, possibly suggesting a stable association with one or more integral membrane proteins. We conclude that fMLF stimulation promotes redistribution of hCAP-18 to the surface of human neutrophils.
Key Words: inflammation • leukocyte • exocytosis • cathelicidin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
During inflammation, hCAP-18 is produced and secreted by activated epithelial cells. Neutrophils, the dominant immune cell type involved in innate immune function, are another principal source of hCAP-18, which is stored preformed within specific granules and released into inflamed tissue as a result of granule exocytosis. Once secreted, the hCAP-18 is cleaved and thus activated by proteinase-3, a component of neutrophil primary granules [6 ]. Beyond its core properties of microbicidal [7 ] and LPS-sequestering activity [8 ], LL-37 is implicated in the regulation of innate and humoral immune responses, wound healing, and tissue repair. The capacity of LL-37 to regulate the innate and humoral immune responses is at least partially a result of its ability to bind the surface of host cells. LL-37 binds several cellular receptors specifically, including the chemoattractant receptor formyl-peptide receptor-like 1, expressed by human neutrophils, monocytes, macrophages, T cells, and mast cells [9 , 10 ], and the P2X7 receptor, expressed on LPS-primed human monocytes [11 ]. Such interactions are implicated in the recruitment of immune cells to inflamed tissue [12 ], mast cell degranulation [13 ], modulation of dendritic cell maturation [14 ], induction of endothelial cell angiogenesis [15 ], and the re-epithelialization of skin wounds [16 ].
Investigations into the cellular interactions of hCAP-18 within inflammatory networks provide a major emphasis for research in this area but have focused primarily on LL-37. The aim of this study was to examine the subcellular distributions of hCAP-18 in human neutrophils using two novel, hCAP-18-specific mAb developed in this laboratory. Based on this study, it is apparent that a substantial fraction of hCAP-18 is bound to the plasma membrane and presented on the surface of human neutrophils upon cell activation by N-formylated peptide chemoattractants.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-cyano-4-hydroxycinnamic (-CHC) acid and the stainless steel MALDI plate were from Bruker (Billerica, MA, USA). The Zorbax 300SB-C18 HPLC-Chip used for liquid chromatography (LC)-mass spectrometry (MS)/MS analysis was purchased from Agilent (Santa Clara, CA, USA), and Labsafe Gel Blue was obtained from G Biosciences (St. Louis, MO, USA). Fluoromount-G was obtained from Southern Biotechnology (Birmingham, AL, USA), and the FITC-conjugated goat antimouse antibody was from Bethyl Laboratories (Montgomery, TX, USA). Wheat germ agglutinin (WGA) was obtained from EY Laboratories (San Mateo, CA, USA). Goat serum was obtained from MP Biomedicals (Irvine, CA, USA); bovine rhodopsin mAb K16 [17
] was produced in this laboratory (Montana State University, Bozeman, MT, USA); mAb 7D5 (recognizing an extracellular epitope of flavocytochrome b) was the kind gift of Dr. Michio Nakamura (Institute of Tropical Medicine, Nagasaki University, Nagasaki, Japan); and cytochrome b-specific mAb 44.1 and 54.1 were produced in this laboratory. All additional materials were purchased from Sigma Chemical Co. (St. Louis, MO, USA).
Blood-neutrophil isolation
Human neutrophils were isolated by a dextran-based procedure as described previously [18
].
mAb generation
The hCAP-18-specific mAb H7 and N9 were produced together with a number of flavocytochrome b-specific mAb using standard hybridoma technology as described [19
]. BALB/c mice were immunized against Triton X-100-solubilized preparations of integral membrane protein purified from cytochalasin B-pretreated, formyl-Met-Leu-Phe (fMLF)-stimulated human neutrophils by WGA- and heparin agarose-affinity chromatography as described [20
]. Hybridomas were screened by ELISA and/or immunoblot for immunogen-specific, IgG-producing clones. Positive cultures were cloned twice by limiting dilution and grown in RPMI medium containing 5% FCS. For purification, antibody was bound to Gammabind-Sepharose beads and eluted with 100 mM glycine, 150 mM NaCl, pH 2.5, into neutralization buffer and dialyzed against 150 mM NaCl, 10 mM Na2HPO4 (pH 7.4).
N-terminal protein sequence analysis
For protein sequence analysis, antigen was immunopurified using mAb H7-Sepharose beads as described below, and the bead-eluate was separated by SDS-PAGE. After minimal staining with Coomassie, a single 16-kDa protein was visualized. Following electrotransfer onto PVDF membranes, the band was excised and sent to Harvard Microchem (Cambridge, MA, USA) for sequence analysis.
SDS-PAGE/immunoblot analysis
These analyses were performed as described previously [18
].
Immunoprecipitation (IP) assays
mAb H7, N9, and K16 were conjugated separately to cyanogen bromide-activated sepharose beads according to the manufacturer’s directions. IP assays were performed with purified plasma membranes from fMLF-stimulated, cytochalasin B-pretreated (fully degranulated) neutrophils. Antibody beads were incubated with purified plasma membrane fractions from 5 x 107 to 1 x 108 cell equivalents in 1.4 ml (1:1 v/v) of PBS (pH 7.4) with 1% DDM. Incubations were for 1 h at 25°C. Beads were then washed six to eight times in PBS (pH 7.4), 1% BSA, and 0.2% DDM and separated by SDS-PAGE. Analysis of immunoprecipitates from mAb H7 or N9 yielded similar results by direct protein staining or by immunoblot analysis.
Cell treatment
Neutrophils were stimulated with fMLF in the presence or absence of cytochalasin B using the following method. After isolation from blood, neutrophils were suspended in Dulbecco’s PBS supplemented with 0.9 mM MgCl2, 0.5 mM CaCl2, 0.1% BSA, and 0.1% dextrose (DPBS4+) at 108 cells/ml and incubated 7 min (37°C), at which time, 1 µM fMLF was added, and incubation continued for an additional 3 min. Reactions were stopped with a fivefold volume of ice-cold DPBS4+ supplemented with 2 mM EDTA, 100 µM PMSF, and protease inhibitor cocktail (Sigma Chemical Co., Catalog Number P8340), sedimented at 740 g for 10 min (4°C), and washed one to two times in DPBS4+. Fully degranulated neutrophil populations were obtained by adding 2 µg/ml cytochalasin B (which promotes full degranulation by inhibiting cortical F-actin polymerization) with 80 µg catalase and 10,000 activity units bovine liver SOD to the fMLF-stimulation buffer at the beginning of the above 7-min incubation period (37°C). Cells were then fMLF-stimulated as above, and catalase and SOD were added to reduce cellular damage. NADPH oxidase-generated superoxide and its toxic oxygen products were released during the full degranulation of neutrophils. fMLF-stimulated neutrohpils, partially and fully degranulated, were analyzed for hCAP-18 expression by flow cytometry or indirect immunofluorescence microscopy as indicated in Results. Alternatively, hCAP-18 was immunoprecipitated from plasma membrane fractions isolated from fully degranulated neutrophils.
Subcellular fractionation
Plasma membranes from fully degranulated neutrophils were isolated by sedimentation of postnuclear fractions (obtained as described in ref. [18
]) at 100,000 g for 1 h (4°C) using a Beckman Ti-75 rotor. Purified plasma membranes were determined to be granule-free by examining membrane fractions for latent LF activity as described below.
Isopycnic sucrose density gradient sedimentation was performed as described previously [18 ]. Plasma membrane-enriched gradient regions [indicated by peak activity of alkaline phosphatase (AP)] were examined for the presence of intact, hCAP-18-enriched, specific granules by ELISA analysis of latent LF activity as described below. No latent LF activity was detectable in plasma membrane fractions from gradients made from the unstimulated or stimulated neutrophil populations examined in this study.
AP and LF activity
AP and LF activities were used as plasma membrane and specific granule marker activities, respectively, and measured in fractions collected from isopycnic sucrose density gradients as described previously [18
]. Plasma membrane fractions isolated by isopycnic sucrose gradient density fractionation or by 100,000 g sedimentation were examined for hCAP-18-enriched specific granule contamination by ELISAs designed to detect latent LF activity. Exocytosed and surface-localized LF, such as that observed on the fMLF-stimulated neutrophils examined in this study (see 
Fig. 5B
), are accessible to anti-LF antibody in ELISAs performed in the absence of detergent. By contrast, specific granule-sequestered LF is inaccessible to antibody in the absence of detergent and can only be evaluated after vesicle disruption (e.g., detergent solubilization). This inaccessible fraction of LF is thus referred to as latent, and latent LF activity was measured as the activity difference in LF detected by ELISAs performed in the presence versus absence of 0.2% Triton X-100.
|
|
|
|
|
Phage display analysis
The epitope structure of mAb H7 and N9 was examined by selecting peptide sequences, which mimic the hCAP-18 binding site from the J404 nonapeptide library, a collection of 5 x 108 unique, M13-based bacteriophage members. Each clone of the library displays up to five copies of a unique, nine-residue peptide displayed at the N-terminus of the pIII capsid protein and contains the nucleotide segment encoding the unique peptide within the genome. Three sequential rounds of selection and amplification of phage display peptide library clones were carried out in these studies. For the initial round of selection, 0.4 mg antibody was allowed to react with 1012 PFU overnight at 4°C, and then antibody was captured on 0.1 ml of the Gammabind-Sepharose matrix for 1 h at 25°C. Following extensive washing of the matrix to remove unbound phage clones, adherent phage displaying immunoreactive peptides were eluted in low pH and amplified in host K91 cells as described [22
]. After sequential rounds of selection and amplification, the sequences of displayed peptides were deduced by nucleotide sequence analysis of the isolated clones as described [23
].
Isolation of hCAP-18 and its prodomain from intact, degranulated neutrophil supernatants
Freshly isolated neutrophils were stimulated with 1 µM fMLF in the presence of cytochalasin B. Following termination of cellular activation, supernatants were isolated and concentrated through an Amicon ultrafiltration membrane with a nominal molecular weight cutoff of 30,000. Filtrates were then separated by SDS-PAGE and immunoblotted as described below.
Indirect immunofluorescence
Freshly isolated neutrophils were treated with 3 mM DFP in DPBS (pH 7.4) for 15 min at 4°C. Cells were stimulated with fMLF, as described above, or left unstimulated. Cells were then washed in the DPBS and fixed with 3% paraformaldehyde (in PBS, pH 7.4) for 15 min at 25°C and then washed several times in the same buffer. Fixed cells were resuspended with PBS (pH 7.4) and 0.2% gelatin, with or without 0.01% saponin (referred to as blocking buffer) and incubated overnight at 4°C. Antibody labeling was performed in 100 µl reaction volumes using 106 neutrophils. Primary labeling used 1 µg each mAb H7 and flavocytochrome b-specific control mAb 7D5 (against extracellular epitope) and 44.1 (recognizing intracellular epitope) used to control for surface and cytoplasmic staining, respectively. mAb H7 labeling experiments were performed alternatively, with or without saponin permealization to label for surface and intracellular antigen, respectively. Control labeling experiments with mAb 7D5 were performed without saponin, and those with mAb 44.1 were performed with the detergent. Irrelevant mAb control or controls without primary antibody indicated no fluorescence and are not shown. After primary labeling, reactions were quenched with 4.5 ml ice-cold blocking buffer, with or without saponin, as indicated in Results, followed by two additional, low-speed washings. After primary labeling, the goat antimouse antibody, anti-IgG-FITC (diluted 1/400 in blocking buffer±saponin), was added and reactions quenched and washed as above. All incubations were for 30 min at 37°C. Stained cells were finally resuspended in 3.5 ml PBS and then distributed in 100 µl aliquots onto glass slides using a Cytospin 2 centrifuge. Cells were then mounted using 50 µl Fluoromount-G and viewed microscopically using a Zeiss Axioskop 2 Plus instrument equipped with a CD camera to allow video imaging via microcomputer and AxioVision Rel. 4.4 software.
Flow cytometry analysis
After blood isolation, neutrophils were resuspended in DPBS4+, distributed at 106 cells/tube, and labeled with 1 µg mAb H7, 1 µg control antibody, or no primary antibody for 30 min on ice. Reactions were quenched by adding a 50-fold vol of ice-cold DPBS4+, followed by sedimentation at 360 g for 8 min (4°C). Neutrophils were then labeled with secondary, FITC-conjugated goat antimouse antibody, washed as described above, resuspended in PBS, and analyzed by flow cytometry. The negative controls used in these experiments included secondary labeling (without primary labeling), primary labeling with irrelevant mAb K16, and primary labeling with mAb 44.1, specific for the cytoplasmic domain of the flavocytochrome b subunit p22phox (a cytoplasmic control to insure the plasma membrane integrity of cells). Negative controls were virtually identical in relative intensity, so only data from the mAb 44.1 control are included in the histogram (see Fig. 6
).
|
Triton X-114 phase separation
NaCl-washed plasma membranes (1 M; isolated from fully degranulated neutrophils) were subjected to Triton X-114 phase analysis as described by Borregaard and Tauber [24
].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
16 kDa when immunoblotted against flavocytochrome b-enriched immunogen preparations of detergent-solubilized and WGA and heparin agarose affinity-purified membrane fractions (data not shown). To determine the identity of this mAb-reactive, 16-kDa band, the corresponding antigen was immunopurified using mAb H7. The SDS-PAGE profile of proteins purified from mAb H7- or irrelevant mAb K16-coated beads is shown in Figure 1A
. Two bands are evident, including a 16-kDa band present only in mAb H7-bound material and a 14-kDa contaminant present in the eluted material of both mAb. For antigen identification, the specific 16-kDa band was electrotransferred, stained, and excised for N-terminal sequence analysis. A corresponding immunoblot is shown in Figure 1B
. Table 1
shows exact alignment of the eight N-terminal residues (starting at Ser 50), which were sequenced by this method with the corresponding residues of hCAP-18 primary structure. It is important that previous studies using N-terminal sequence analysis and MALDI peptide mass mapping have suggested that 31Gln of hCAP-18 contains a pyrrolidone carboxylic acid modification, which prohibits N-terminal sequence analysis [25
, 26
]. Taken together, our N-terminal sequence analysis is consistent with hCAP-18 having a blocked N-terminus and appears to have detected a proteolyzed form of hCAP-18 containing Ser 50 as the N-terminal residue.
|
|
Epitope mapping of mAb H7 and N9
To characterize these novel, hCAP-18-specific mAb further, the composition and structure of the amino acid sequence motif targeted by mAb H7 and N9 were examined using phage display analysis. Sequence analysis of phage clones complementary to the epitope of both mAb revealed virtually identical consensus motifs (Fig. 3A)
. These findings suggested that mAb H7 and N9 recognized a similar or identical probable epitope sequence in hCAP-18 primary structure. A consensus sequence of NARGY was obtained upon initial comparison of phage clones. This was later modified to NARGZ (where Z is an aromatic residue) to facilitate precise alignment in hCAP-18 primary structure (Fig. 3B)
. The NARGZ consensus was assigned to residues 115, 117–119, and 121, located at the C-terminal end of the prodomain (Fig. 3B)
, and is consistent with the secondary structure or surface folding, found in numerous other protein epitopes [22
, 27
]. In light of this phage display data and the fact that indistinguishable results were obtained with mAb H7 and N9 in each of the experiments described below, only data from mAb H7 are presented in the remainder of this study.
To confirm mAb H7 recognition of the prodomain of hCAP-18, immunoblots were performed against supernatants collected from fully degranulated neutrophils. By stimulating cells with fMLF in the presence of cytochalasin B, degranulation is potentiated by inhibiting actin polymerization within the cortical barrier of plasma membranes. Within supernatants collected from fully degranulated cells, hCAP-18 becomes accessible to proteinase-3 (normally sequestered in primary granules), which catalyzes LL-37 cleavage [6 ]. A representative immunoblot from three independent experiments is shown in Figure 4 , indicating mAb H7 recognition of a 16-kDa band (Lanes A–C), identified above as the hCAP-18 and a 14-kDa band (Lanes B and C). The latter band is within the molecular weight range (12–14 kDa) expected for the prodomain [28 ]. No detectable signal was observed in the molecular weight range (3–4 kDa) of LL-37 on immunoblots from high-percentage polyacrylamide gels (data not shown). The 16- and 14-kDa bands labeled in Figure 4 were subsequently coimmunopurified from degranulated neutrophil supernatants and their identity confirmed by MALDI peptide mass mapping (data not shown). Lysozyme was also detected as a component of the 14-kDa band in these experiments (data not shown). These data support results from phage display analysis indicating mAb recognition of the hCAP-18 prodomain.
Subcellular distribution of hCAP-18 in unstimulated and fMLF-stimulated neutrophils
The immunogen preparation used to elicit mAb H7 and N9 consisted of cell surface-transmembrane membrane proteins purified from fMLF-stimulated, cytochalasin B-pretreated, fully degranulated neutrophils [20
]. The generation of mAb that recognize hCAP-18, using membrane fractions as the source of immunogen, suggested the specific association of hCAP-18 with the cell surface of neutrophils following their activation. This observation was of interest, as the association of hCAP-18 with neutrophil membranes had yet to be described. Fully degranulated cells, however, are nonphysiological, so that the surface translocation of hCAP-18 observed in this population might be an in vitro artifact. Therefore, additional experiments were performed to determine surface translocation of hCAP-18 in response to fMLF stimulation without cytochalasin B pretreatment. In these studies, we determined the subcellular distribution pattern of hCAP-18 in unstimulated (Fig. 5A
) versus fMLF-stimulated (Fig. 5B)
neutrophils in parallel experiments (n=3) using the same number of cells per experiment. In these experiments, postnuclear fractions were prepared and sedimented on sucrose density gradients as described in Materials and Methods. The data shown in Figure 5A
indicate the cosedimentation of the 16-kDa hCAP-18 with the specific granule marker LF at 42% sucrose in unstimulated neutrophils. In this set of experiments, hCAP-18 could not be detected in the peak plasma membrane fraction (identified by AP) at 30% sucrose and thus, suggested the absence of hCAP-18 from the surface of unstimulated cells.
We observed that fMLF stimulation resulted in losses of 20–30% and 25–38% for specific granule hCAP-18 and LF, respectively, relative to unstimulated neutrophils (n=3, data not shown). When the subcellular distribution of hCAP-18 was examined in fMLF-stimulated neutrophils, a bimodal sedimentation pattern was detected (Fig. 5B) with peak activities in the specific granule (LF) and plasma membrane (AP) compartments. The latter peak suggested that following exocytosis, hCAP-18 partially redistributes to the cell surface. An average 10.3% of the total cellular hCAP-18 could be recovered from neutrophil plasma membranes after fMLF stimulation, with 16.3% recovered from supernatants and 73.3% remaining associated with intact granules. It is interesting that the amount of cellular hCAP-18 recovered from the cell surface increased by several fold (comprising >50% of the total cellular hCAP-18) when neutrophils were fully degranulated by fMLF stimulation in the presence of cytochalasin B prior to fractionation on sucrose density gradients (data not shown; see below). We also found that the proform of hCAP-18, shown to have a relative molecular weight of 13,630 ± 120 (n=3; see Fig. 4 above), was not detected in the plasma membrane fractions from partially or fully degranulated cells. This was evident from five independent determinations of hCAP-18, 15,950 ± 180 (n=5), including immunoprecipations and direct immunoblot analyses of membrane fractions. With regard to fMLF-stimulated (partially degranulated) neutrophils, we also noted that degranulated LF appeared to redistribute to the plasma membrane in a pattern similar to hCAP-18, as evidenced by the cosedimentation of LF and AP (Fig. 5B) . These latter observations are consistent with previous reports demonstrating the stable association of LF with plasma membranes of agonist-stimulated neutrophils [29 , 30 ].
To confirm sucrose density gradient studies outlined above, the subcellular localization pattern of hCAP-18 in unstimulated and fMLF-stimulated neutrophils was examined further by indirect immunofluorescence microscopy. Figure 5C shows representative DIC and fluorescence micrographs of neutrophils labeled with mAb H7 (–fMLF, +fMLF) or control mAb (Control) in the absence (left) or presence (right) of permeabilizing detergent, saponin, from five independent experiments per neutrophil population. Labeling with control mAb 7D5 (Control, left; specific for an extracellular epitope of flavocytochrome b) in the absence of saponin produced the typical, ringed fluorescence pattern expected for the labeling of surface-exposed epitopes. Labeling with control mAb 44.1 (Control, right; specific for cytoplasmic epitope of flavocytochrome b) in the presence of saponin produced the expected pattern for cytoplasmic staining with prominent nuclear exclusion. In unstimulated neutrophils (–fMLF), fluorescence signal with mAb H7 was only observed when cells were labeled in the presence of saponin (right) and resulted in typical cytoplasmic staining with nuclear exclusion. This indicated that hCAP-18 is predominately intracellular in unstimulated neutrophils and consistent with the observed single peak of hCAP-18 and LF cosedimentation in this population (Fig. 5A) . By contrast, fMLF-stimulated neutrophils (+fMLF) exhibited a characteristic, ringed fluorescence when labeled with mAb H7 in the absence of saponin (left). These results indicate the localization of hCAP-18 on the cell surface of this population and are consistent with the codistribution of hCAP-18 and AP in sucrose density gradients (Fig. 5B) . A much more intense signal was obtained when fMLF-stimulated cells (+fMLF) were labeled in the presence of saponin (right), indicating that hCAP-18 remains predominantly intracellular in this population, also consistent with our sucrose density sedimentation studies. No labeling was observed by control, irrelevant, isotype-matched mAb K16 in the presence or absence of saponin for unstimulated or fMLF-stimulated cells. This control confirms that the nonspecific binding of lysozyme by IgG1 antibodies observed in Figure 1 is insignificant for immunofluorescence studies.
Surface expression of hCAP-18 on neutrophils following stimulation in the presence and absence of cytochalasin B
To determine if differences existed in the levels of surface hCAP-18 between fMLF-stimulated, partially and fully degranulated cells, neutrophils were stimulated with fMLF in the absence (for partial degranulation) or the presence (for full degranulation) of cytochalasin B and examined by flow cytometry for surface hCAP-18 expression. The histogram shown in Figure 6
(upper) is representative of results obtained from three independent experiments and indicates a progressive right-shift in the signal intensity of surface expression of hCAP-18 on partially versus fully degranulated cells. The averaged increases in relative fluorescence intensity are indicated in the bar graph of Figure 6
(n=3), showing that full degranulation increases surface hCAP-18 five- to sixfold. Together, these experiments suggest that hCAP-18 progressively accumulates on the surface of stimulated neutrophils in parallel with specific granule exocytosis.
Properties of hCAP-18 surface expression
The capacity of hCAP-18 to associate with stimulated neutrophil plasma membranes suggests its capacity for surface electrostatic interactions, bilayer insertion, or specific binding to one or more membrane-bound proteins. We performed additional experiments to obtain insight into the nature of hCAP-18 plasma membrane interaction. The relative strength and/or nature of hCAP-18 membrane association were examined by washing membranes in various buffer solutions. High pH or high ionic strength buffers can remove peripherally associated membrane proteins from purified plasma membrane fractions with varying degrees of efficiency (see below, refs. [31
, 32
]) and are therefore useful in distinguishing peripheral versus transmembrane associations. For these studies, plasma membrane fractions were purified from fully degranulated neutrophils and washed consecutively with DPBS alone, 1 M NaCl, and 10 mM NaOH. After each wash, aliquots were removed and assayed for hCAP-18 content by immunoblot analysis. We noted that a relatively small fraction of the surface hCAP-18 (8%) could be removed after washing with DPBS (Table 3
). However, the major fraction of surface hCAP-18 retained after washing in DPBS (92%) also remained membrane-associated after washing with 1 M NaCl (Table 3)
. These data suggested that hCAP-18 may interact with the cell surface through low- and high-affinity associations. Whereas some high-affinity, peripheral associations may withstand 1 M NaCl treatment, 10 mM NaOH is a much harsher treatment, which removes such strongly bound peripheral associations completely [31
] and can disrupt receptor-ligand interactions through protein denaturation. Neither treatment affects the associations of integral or hydrophobically associated proteins [31
, 32
]. Accordingly, upon further treatment with 10 mM NaOH, the resulting membranes were nearly devoid of surface hCAP-18 (Table 3)
. Taken together, these data may suggest that hCAP-18 principally interacts with the cell surface through highly stable peripheral associations with the membrane bilayer or through a receptor-mediated interaction.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
16,000. These antigens were identified initially as an N-terminally truncated hCAP-18 fragment-detected N-terminal sequence analysis. MALDI and nanospray LC-MS/MS analysis of tryptic digests subsequently confirmed that the antigen contained peptides spanning residues 31Gln–167Arg, representing 77% of sequence coverage. This coverage suggests that but for three C-terminal residues (168TES), the entire 16,016-kDa, 140-amino acid residue hCAP-18 antigen, including peptides derived from the LL-37 microbicidal peptide domain, is represented. Phage display epitope mapping, indicating that mAb H7 and N9 recognize a 5-amino acid consensus sequence, NARGZ, suggested the epitope most probably incudes the primary sequence 115NXARGXZ121 (where X and Z indicate omitted and aromatic amino acids, respectively), present in the C-terminal region of the prodomain. It is interesting that the relative molecular mass of 16 kDa observed in this study for hCAP-18 matches the 16,016-Da mass predicted for the 140 amino acid residue protein shown in Figure 3B
, beginning at 31Gln. The SDS-PAGE relative molecular mass we observed differs slightly from the 18-kDa estimate reported in other studies [33
34
35
36
37
]. However, as MALDI and LC-MS/MS analysis of the mAb H7 affinity-purified hCAP-18 demonstrated the presence of the LL-37 domain, this discrepancy is most likely a result of limitations in SDS-PAGE for quantitative evaluation of molecular weight [38
39
40
41
]. As discussed previously [42 ], residues 1–30 of human neutrophil hCAP-18 provide a signal sequence, which is processed proteolytically during biosynthesis and targeting of the protein to specific granules. However, the N-terminal sequence determined for our affinity-purified hCAP-18 was found to occur 20 amino acids downstream of 31Gln (the predicted N-terminus of the mature protein without signal sequence; see Fig. 3B ). Examination of the immunopurified antigen by nanoelectrospray MS confirmed 31Gln cyclization to pyrrolidone glutamic acid at the N-terminus of hCAP-18, previously implicated by MALDI analyses performed by others [25 , 26 ]. This modification inhibits sequential Edman degradation [43 ], thus explaining our exclusive detection of a proteolyzed form of hCAP-18 lacking residues 31Gln–49Arg by N-terminal sequence analysis (Table 1) . It is interesting that Malm et al. [35 ] reported observing a doublet hCAP-18 band in immunoblots of SDS-PAGE-separated human seminal plasma, possibly suggesting conservation of this species in other tissues. In addition, the cyclization of N-terminal 31Gln to pyrrolidone glutamic acid, which has also been observed to occur in nonhuman cathelicidins [44 ], is known to alter the functional and/or regulatory state of other proteins and/or peptides [45 , 46 ], and may play a role in human immune function [47 ]. Thus, it will be interesting to determine whether the structural alterations of hCAP-18 observed in this study are important for its biological function in neutrophils.
Given that hCAP-18 is a well-characterized component of neutrophil-specific granules [25 ], results from subcellular fractionation studies further supported the identity of the H7/N9 antigen as hCAP-18 by showing the expected codistribution of this molecule with specific granule LF in unstimulated neutrophils. However, further analysis of hCAP-18 distribution in fMLF-stimulated, partially degranulated neutrophils showed cosedimentation of hCAP-18 with specific granule LF and the plasma membrane marker AP. In conjunction with the fact that mAb H7 and N9 were generated using a detergent-solubilized membrane fraction as the immunogen, the above observation suggested a novel capacity of soluble hCAP-18 to interact with intact neutrophils. In the present study, surface hCAP-18 association on fMLF-stimulated neutrophils was confirmed separately by indirect immunofluorescence microscopy and flow cytometry analysis and in immunopurification assays, by using purified plasma membrane fractions from fully degranulated cells.
hCAP-18 surface expression on intact, fMLF-stimulated, partially and fully degranulated neutrophils shown by flow cytometry in Figure 6
indicates that surface hCAP-18 expression increased in parallel with the exocytosis of hCAP-18-enriched specific granules. Neutrophils contain 0.63 µg hCAP-18/106 cells [34
]. From our subcellular fractionation studies (Fig. 5)
, we estimate that fMLF stimulation induces 10% of this amount or 0.06 µg hCAP-18/106 cells to be bound on the cell surface. Full degranulation results in a five- to sixfold increase or 0.3 µg hCAP-18/106 cells and represents 50% of the content of specific granules. This amount is more than three times the amount observed to be bound to the surface of human spermatozoa [35
]. Malm et al. [35
] suggest that such amounts are suggestive of a microbicidal function for surface-bound hCAP-18 on spermatozoa, if only from the release of LL-37 by exposure to the serine protease acrosin released upon fertilization. Given that large amounts of proteinase-3 are released from neutrophil primary granules upon full degranulation, the plasma membrane-bound hCAP-18 could also serve as a targeted source of hCAP-18 for the production of LL-37.
It is interesting that we are unable to detect significant amounts of hCAP-18 on the surface of unstimulated neutrophils by indirect immunofluorescence or subcellular fractionation. A recent study reported hCAP-18 to be a significant component of normal human plasma in complex with lipoproteins at a concentration of 1.2 µg/ml [48 ] or approximately one-third the amount stored in neutrophils in the same volume of whole blood. This consideration suggests that unstimulated neutrophil plasma membranes are incapable of stably interacting with hCAP-18 in the blood. This property could be a result of differences in membrane composition between unstimulated and stimulated neutrophils or the inability of the lipoprotein-bound hCAP-18, which is constitutive in human plasma, to interact with the surface of circulating, quiescent neutrophils.
The expression of low-density lipoprotein receptors (LDLR; ref. [49 ]) by human neutrophils, considered together with the observed association of hCAP-18 with plasma lipoproteins [26 ], suggests that plasma lipoproteins might mediate the binding of hCAP-18 to the surface of human neutrophils. However, it is unlikely that plasma lipoproteins contribute to the surface association of hCAP-18 observed in the present study, given that our experiments were performed in the absence of plasma lipoprotein. In addition, as we could not detect hCAP-18 on the surface of unstimulated neutrophils, which reportedly express LDLR [49 ] and are exposed to lipoprotein and hCAP-18 prior to blood withdrawal, it seems unlikely that plasma lipoprotein or neutrophil LDLR expression contributes to hCAP-18 surface association observed here. It is interesting, however, that results from the present and previous studies do not exclude the possibility that neutrophils express a hCAP-18-specific receptor, which is expressed differentially on the surface of activated human neutrophils.
The positive correlation of surface hCAP-18 expression with the extent of neutrophil activation observed in Figure 6 is interesting considering that neutrophil activation and the expression level of hCAP-18 within tissues are highly regulated responses to inflammation. Several previous reports have documented the ability of human neutrophils to occupy functionally diverse, cellular activation states [50 51 52 53 54 ]. This phenomenon is associated with a graded as opposed to an all-or-nothing cellular response to microenvironmental stimuli, which in turn suggests a functional basis for the cell surface localization of hCAP-18 observed in this study. Specifically, plasma membrane-bound hCAP-18 could contribute to the microbicidal efficacy of neutrophils within inflamed tissue. For example, surface hCAP-18 would be ideally situated for the in situ killing and degradation of microbes during their phagocytic uptake. Such proximity could confer a spatial specificity on hCAP-18 activation by extracellular proteases by concentrating antimicrobial action at this interface, thus facilitating microbial destruction and simultaneously minimizing damage to surrounding tissues. In this context, it is possible that hCAP-18 plasma membrane association is mediated by one or more membrane proteins, which upon binding hCAP-18, contribute to cell-signaling processes involved in phagocytosis and/or other mechanical functions (e.g., substrate adherence) in a manner analogous to interactions involving plasma membrane-bound LF [55 , 56 ].
Reports documenting expression of hCAP-18 on cell surface membranes have, up to now, been confined to germ-line cells, which can be coated with as many as 6.6 million copies of hCAP-18 per cell [33 , 35 ]. Although a function for hCAP-18 in reproductive biology has not yet been identified, the possibility that the active maintenance of hCAP-18 at the cellular microenvironmental interface could supplement the defensive capacity of such cells has been postulated [5 ]. In addition, previous studies have indicated that neutrophil hCAP-18 is likely soluble within specific granules [25 , 34 , 57 , 58 ] and that as much as 25% of this cellular fraction can be lost during fMLF stimulation in the absence of phagocytosis [52 ]. This suggests that significant quantities of hCAP-18 might be released during the process of neutrophil extravasation and their subsequent transit into infected tissues, which could in turn limit the use of this antimicrobial protein by neutrophils at the infection site and subject uninfected tissues to damage. In this regard, the substantial, progressive cell surface association of hCAP-18 with agonist-stimulated neutrophils might also serve as a mechanism to maximize the recovery and transport of soluble hCAP-18 or its fragments into sites of microbial infection, as well as enhance the efficacy and specificity of their antimicrobial action. In spite of the implied uncertainties of this study, which does not directly address the antimicrobial action of surface-associated hCAP-18, we speculate that surface localization enhances the efficiency of delivery to inflammatory sites, where there could be increased coordinate action [59 ] of hCAP-18 or its proteolytic fragments with other defenses unleashed in the phagosome or extracellular milieu. Collectively, the findings presented in this study offer a basis for additional investigation into the biology and function of hCAP-18 and introduce two highly specific anti-hCAP-18 mAb for use in future research.
|
ACKNOWLEDGEMENTS |
|---|
Received September 22, 2006; revised January 29, 2007; accepted January 29, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-helical antimicrobial peptides Biopolymers 55,4-30[CrossRef][Medline]-blocked Biochem. J. 217,253-257[Medline]
