C-reactive Protein: A Physiological Activator of Interleukin 6 Receptor Shedding

Culture reagents were obtained from GIBCO BRL, and purified human CRP from Calbiochem-Novabiochem Corp. Peptides corresponding to CRP amino acid residues 77–82 (VGGSEI), 174–185 (IYLGGPFSPNVL), and 201–206 (KPQLWP) were from Sigma Chemical Co. Biotinylated anti– human IL-6R antibody (BAF-227) was from R&D Systems. Anti-DS–sIL-6R mAb (2F3) was generated as described previously (20). Dr. R.A. Black (Immunex Corp.) provided the TNF-α–protease inhibitor, TAPI. Lymphoprep was from Nycomed Pharma, and ImmunoPure 3,3′, 5,5′-tetramethylbenzidine (TMB) from Pierce Chemical Co.

Venous blood (20 ml) was obtained by antecubital venipuncture from nonsmoking healthy individuals (aged 26–54), mixed with an equal volume of 2% (wt/ vol) dextran/0.8% (wt/vol) trisodium citrate in PBS (pH 7.4), and erythrocytes were allowed to sediment. Plasma was collected, underlayered with Lymphoprep (2:1 [vol/vol] plasma/Lymphoprep), and centrifuged at 4°C for 20 min at 800 g. The neutrophil-containing pellet was collected, and contaminating erythrocytes were removed by hypotonic lysis. Neutrophil preparations were found to be >95% pure as assessed by differential Wright staining. Before use, neutrophils were resuspended in serum-free RPMI 1640 containing 2 mM l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin.

Neutrophils (2 × 10⁶ cells) were treated as described in the figure legends. Culture medium was harvested and stored at −80°C until required. Concentration of sIL-6R was determined using an ELISA procedure. Microtiter 96-well plates were coated with 10 μg/ml anti– human IL-6R mAb (mAb 17.1; reference 21) and blocked at 4°C with 0.5% BSA. sIL-6R standards and unknowns were added and incubated at room temperature for 2 h. To detect bound sIL-6R, biotinylated anti–human IL-6R antibody (50 ng/ml BAF-227) was added for 2 h at room temperature, followed by a 20-min incubation with horseradish peroxidase–conjugated streptavidin. Plates were washed between each step with PBS containing 0.1% Tween 20. Peroxidase activity was determined using TMB as a substrate. The reaction was stopped with 1.8 M H₂SO₄, and absorbance was measured at 450 nm. To detect DS–sIL-6R, the capture antibody was replaced with 20 μg/ml anti-DS–sIL-6R antibody (mAb 2F3), and ELISA was performed as described using baculovirus-expressed DS–sIL-6R as a standard (20). The lower limit of detection for sIL-6R and DS–sIL-6R was 10 and 50 pg/ml, respectively.

Loss of IL-6R expression from the neutrophil cell surface after stimulation was monitored by cytofluorometry (FACScan^®; Becton Dickinson) as described (20). Values are expressed as the percent reduction in mean fluorescence units (MFU) from nonstimulated control cells: MFU = (FU_experimental − FU_{autofluorescence})/(FU_control − FU_{autofluorescence}).

Statistical analysis was performed using Student's t test incorporated into the SigmaPlot (version 2.01) graphics program. A P < 0.05 indicated a statistically significant difference.

Examination of human neutrophils obtained from 10 independent donors showed that CRP activates sIL-6R production (Fig. 1). In each case, basal sIL-6R release was significantly increased (P < 0.0001) after exposure to 50 μg/ml CRP, with the extent of sIL-6R production ranging between 86 and 234 pg/ml after CRP stimulation compared with 21–84 pg/ml for controls. On average, CRP resulted in a 3.06 ± 1.03–fold induction of sIL-6R levels (Fig. 1 B). In contrast, activation of human neutrophils with IL-4 or IL-10 had no effect on sIL-6R generation (data not shown). As shown in Fig. 2 A, production of sIL-6R increased rapidly, with optimal release occurring between 30 and 60 min after CRP addition. Generation of sIL-6R was also dose-dependent, with 50 μg/ml CRP inducing a maximal response (Fig. 2 B). Release of the sIL-6R in response to a single exposure to CRP was transient, since levels returned to baseline within 4–5 h after stimulation. Addition of a second CRP dose 2 h after the initial CRP stimulation did not further enhance production (data not shown).

Neutrophil stimulation has been shown to activate the cleavage of native CRP into biologically active peptide fragments (22). In particular, peptides corresponding to amino acid residues 77–82, 174–185, and 201–206 profoundly influence neutrophil responses (23, 24). Accordingly, human neutrophils were incubated with each of these peptides and their capacity to augment sIL-6R production was determined. As shown in Fig. 3, (174–185)CRP and (201–206)CRP stimulated sIL-6R production in a dose-dependent manner, whereas peptide (77–82)CRP had little or no effect.

The sIL-6R can be released through proteolytic shedding of the cognate IL-6R or secreted as the product of differential IL-6R mRNA splicing (11–14). Flow cytometry using neutrophils from three separate donors showed that a 30-min exposure to 100 μg/ml native CRP resulted in a 44 ± 2.5% loss of the cognate IL-6R. Similarly, 100 μg/ml (174–185)CRP and (201–206)CRP stimulated a 33 ± 6.2% and 24 ± 0.3% reduction in the surface expression of IL-6R, respectively, whereas peptide (77–82)CRP had little effect (data not shown). This indicates a role for CRP in the activation of IL-6R shedding. To verify this conclusion, the concentration of DS–sIL-6R was determined in conditioned media from CRP-activated neutrophils, using an antibody specific for the unique COOH-terminal amino acid sequence (GSRRRGSCGL) of DS–sIL-6R (20). As shown in Table I, no detectable level of DS–sIL-6R could be identified either before or after CRP stimulation. Interestingly, no correlation could be established between elevated systemic CRP concentrations and DS–sIL-6R levels in patients suffering from various clinical disorders (Horiuchi, S., and N. Yamamoto, unpublished data).

Hydroxamic acid–based metalloprotease inhibitors such as TAPI are known to prevent shedding of various cell surface proteins (25–27), including the IL-6R (12, 20). Surprisingly, the CRP-induced release of sIL-6R by neutrophils was only partially blocked by TAPI (∼20–25%; Fig. 4). Consistent with previous reports (12, 20), TAPI inhibited 70– 75% of the phorbol ester–stimulated sIL-6R production by monocytic THP-1 cells (data not shown). Thus, the mechanism responsible for CRP-induced release of the cognate IL-6R from human neutrophils may be distinct from that described for monocytic cells.

Elevated levels of the sIL-6R have been associated with the pathology of several disease states. This implies that production of the sIL-6R is increased as part of the inflammatory response. However, little is known regarding the factors that might regulate sIL-6R generation. In this study, physiological concentrations of native CRP and biologically relevant CRP-derived peptides were found to stimulate sIL-6R production by human neutrophils. Release of this soluble receptor was rapidly induced after CRP treatment and occurred via shedding of the cognate IL-6R from the cell surface. C-reactive protein represents the first known endogenous activator of this process. The observation that release of sIL-6R is only partially prevented by the hydroxamic acid–based metalloprotease inhibitor TAPI is of particular interest, since IL-6R shedding in response to phorbol esters and ionomycin has been shown to be prevented by this agent (12, 20). Thus, in neutrophils, shedding of the IL-6R presumably occurs through a unique proteolytic mechanism. Indeed, under certain experimental conditions the phorbol ester–induced shedding of L-selectin (CD62L) from human neutrophils has also been found to be only partially susceptible to a TAPI homologue (27).

In general, plasma CRP levels correlate with severity of inflammatory diseases. During the onset of inflammation or tissue injury, plasma concentrations of CRP are dramatically elevated from ∼1 μg/ml in healthy individuals to as much as 500 μg/ml during the acute phase response (28). In vitro studies have shown that control of this response is primarily regulated by IL-6 (29). More recently, human CRP-transgenic mice were used to verify in vivo that IL-6 is absolutely required for the induced expression of CRP during an inflammatory acute phase response (30). Our current findings show that this relationship between IL-6 and CRP is more complex than previously thought, since IL-6R shedding in response to CRP likely contributes to formation of the agonistic sIL-6R/IL-6 complex. Thus, CRP acts not only as an acute phase reactant, but it may have a profound effect on distal IL-6–mediated events that occur during the inflammatory process. Indeed, CRP levels in several diseases have been found to correlate with those of sIL-6R (31–33).

It is now recognized that CRP plays a significant role in host defense against pathogens (34). C-reactive protein also binds to specific receptors on human neutrophils and diminishes neutrophil responses, such as chemotaxis (35) and the activation of superoxide generation and degranulation by chemoattractants (36). In addition, CRP prevents neutrophil adhesion to endothelial cells via induction of L-selectin (CD62L) shedding (37). Consistent with these findings, in vivo studies have shown that CRP abates neutrophil recruitment in models of inflammation (38, 39). Taken together, these data indicate that CRP also performs an antiinflammatory function. It is therefore noteworthy that the sIL-6R/IL-6 complex has been shown to regulate proinflammatory activation of endothelial cells and to promote neutrophil recruitment (7, 8). In agreement with these findings, it has been observed that the extent of neutrophil infiltration into arthritic joints correlates with elevated sIL-6R levels in synovial fluid (40). It is conceivable that CRP may perform a pivotal role during inflammation by modulating the rate of neutrophil recruitment. It is also highly likely that CRP represents only one endogenous activator of IL-6R shedding, whereas release of DS–sIL-6R may also contribute to the overall properties of sIL-6R (13, 14, 18).

Previous studies have shown that peptides spanning residues 77–82 and 201–206 of the native CRP molecule block neutrophil superoxide generation and chemotaxis (23, 24), whereas peptide fragment 177–182 enhances cytokine/chemokine production and the tumoricidal activity of monocytic cells (41). Structure/function investigations of native CRP (for a review, see reference 34) reveal that amino acids 77–82 reside within the phosphocholine (PCh)-binding site of the CRP molecule, whereas residues 174–185 and 201–206 form parts of the walls of a deep cleft on the opposite face of the CRP protomer. The shallow end of this cleft represents the C1q-binding site of CRP (34), whereas residues 175–179 are important for Fcγ-R1 binding (42). Interestingly, in the present study, CRP peptides 174–185 and 201–206 effectively augmented sIL-6R production by human neutrophils. However, release was not observed in response to residues 77– 82. Similarly, CRP peptides 174–185 and 201–206, but not 77–82, were found to mediate L-selectin shedding (37). These data argue that the ability of CRP to stimulate IL-6R and L-selectin shedding from neutrophils involves interaction via the C1q/Fcg-R1 binding motif of CRP, and does not involve the PCh-binding site. Support for this concept is derived from the fact that disruption of the Ca²⁺-dependent interaction of PCh with CRP (34) through the addition of EDTA had no effect on the CRP-induced release of sIL-6R (data not shown).

Although neutrophils express relatively high levels of the cognate IL-6R, IL-6 signaling in these cells is poorly defined and appears to evoke only weak biological activities (43, 44). However, shedding of the IL-6R from human neutrophils has been shown to activate endothelial cells (8). As a result, expression of the IL-6R on neutrophils may primarily serve as an inducible source of sIL-6R. Thus, the activated shedding of the IL-6R from neutrophils may indirectly propagate the inflammatory response via stimulation of resident tissue cells by the sIL-6R/IL-6 complex.

The authors wish to thank Drs. B.A. Freeman, J.S. Hagood, and V.B. O'Donnell for their help and advice. We are also grateful to Dr. R.A. Black for kindly providing TAPI, and Mrs. T. Rogers for FACScan^® analysis.

This work was funded in part by an American Heart Association Research Fellowship awarded to S.A. Jones.