TY - JOUR
AU1 - Rakita, Robert M.
AU2 - Van C. Quan,
AU3 - Jacques-Palaz, Karen
AU4 - Singh, Kavindra V.
AU5 - Arduino, Roberto C.
AU6 - Mee, Mee
AU7 - Murray, Barbara E.
AB - Abstract Many clinical isolates of Enterococcus faecium are resistant to neutrophil (PMN)-mediated phagocytosis and killing in the presence of normal human serum. We have now examined the ability of specific polyclonal rabbit antibodies to promote opsonization and killing of phagocytosis-resistant E. faecium. Immune rabbit serum generated against formalin-killed E. faecium TX0016, a phagocytosis-resistant strain, markedly promoted binding of TX0016 organisms to PMNs and PMN-mediated killing. These effects were dramatically reduced by (a) adsorption of immune serum with E. faecium TX0016, but not by adsorption with a strain of E. faecium susceptible to phagocytosis, and (b) incubation of immune serum with carbohydrate purified from TX0016, but not by incubation with a surface protein extract from TX0016. IgG purified from immune serum was unable by itself to promote bacterial binding to PMNs. However, specific IgG was able to promote binding to PMNs and PMN-mediated killing in the presence of normal human serum as a complement source, as were F(ab′)2 and Fab fragments produced from it, and the alternative pathway of complement was sufficient to promote IgG- and F(ab′)2-mediated opsonization. PMN complement receptor type 3, but not complement receptor type 1, was involved in bacterial binding to PMNs induced by the combination of F(ab′)2 fragments and normal human serum. These results suggest that opsonization by antibodies potentially directed against bacterial carbohydrate, in conjunction with complement activation, has an important role in the host defense against phagocytosis-resistant E. faecium. Enterococcus faecium, Phagocytosis, Neutrophil, Complement, Antibody 1 Introduction Enterococci are important causes of endocarditis and nosocomial infections, including urinary tract infections and bacteremia [1,2]. The development of resistance to multiple antimicrobial agents, most recently vancomycin [3], has severely compromised the choice of therapies, and the incidence of vancomycin resistance has dramatically increased [4–6]. While Enterococcus faecalis has been the most frequent species associated with clinical infections, the spread of vancomycin resistance has been associated with a shift in species selection, with Enterococcus faecium predominating over E. faecalis 10:1 among vancomycin-resistant enterococci [5–7]. Neutrophils (PMNs) play an important role in the human host defense system's response to bacterial infections. We [8] and others [9,10] have demonstrated that complement is of major importance in the opsonization of E. faecalis for PMN-mediated killing, and that antibodies from patients infected with E. faecalis and from rabbits immunized with E. faecalis can also function as opsonins. However, while all strains of E. faecalis are readily phagocytosed by PMNs in the presence of complement, we have also previously demonstrated that 50% of unrelated clinical isolates of E. faecium are resistant to PMN-mediated phagocytosis and killing, and that a bacterial surface carbohydrate is most likely responsible for this effect [11]. To investigate the mechanisms involved in host defense against phagocytosis-resistant E. faecium, we generated polyclonal rabbit antiserum against one phagocytosis-resistant strain of E. faecium. We found that antibodies from this antiserum, along with F(ab′)2 and Fab fragments produced from them, were unable to alter phagocytosis resistance by themselves. However, both intact antibodies and fragments could promote phagocytosis and PMN-mediated killing when combined with a complement source, and this bacterial-PMN binding was mediated by PMN complement receptor type 3 (CR3). These antibodies appeared to be directed against bacterial carbohydrates, as their opsonic capability was removed by incubation with carbohydrates purified from this same strain of E. faecium. 2 Materials and methods 2.1 Reagents Reagents were obtained from Sigma (St. Louis, MO) unless otherwise indicated. Antibodies directed against PMN complement receptors included: M1/70 (anti-CD11b, rat IgG2b F(ab′)2 fragments, obtained from American Type Culture Collection (ATCC), Rockville, MD) used at 5 µg ml−1[12]; 60.1 (anti-CD11b, mouse IgG1 F(ab′)2 fragments, courtesy of Repligen Corporation, Cambridge, MA) used at 5 µg ml−1[13]; Mo1 clone 44 (anti-CD11b, IgG2a, mouse ascites, courtesy of R. Todd, Ann Arbor, MI) used at a 1:10 dilution [14]; R15.7 (anti-CD18, mouse IgG1, courtesy of R. Rothlein, Boehringer Ingelheim Pharmaceuticals, Ridgefield, CT) used at 15 µg ml−1[15]; W6/32 (anti-HLA, used as a control, obtained from ATCC) used at 10 µg ml−1[16]; polyclonal rabbit IgG directed against PMN complement receptor type 1 (CR1) and mAb 543 (anti-CR1) [17], both generously provided by I. Gigli and used at 20 µg ml−1. The concentrations used were known to be saturating for human PMNs by flow cytometry studies. 2.2 Bacterial strains E. faecium TX0016 was originally isolated from the blood of a patient with endocarditis [8] and is resistant to PMN-mediated phagocytosis and killing after opsonization with 10% normal human serum [11]. E. faecium RLA-3 [18] is susceptible to phagocytosis [11] and was used as an adsorption control. E. faecium MCV161, TH1-7, 4586, SH-11, TX0015, RLA-6, FA191, and X25645 (see [11] for strain descriptions) are resistant to phagocytosis after opsonization with 10% normal human serum and were used to examine cross reactivity of immune rabbit sera. Bacteria were maintained on brain heart infusion (BHI) agar plates, grown overnight in BHI broth, then diluted 1:100 into fresh BHI broth and grown to exponential phase for 4 h at 37°C with tumbling. Bacteria were then washed twice and resuspended in Hanks' balanced salt solution (HBSS) and the bacterial density was adjusted spectrophotometrically. 2.3 Bacterial carbohydrate purification Carbohydrate was purified from E. faecium TX0016 by a modification of a method previously used for group B streptococci [19]. Bacteria were grown overnight in 1 l BHI broth, washed, then inoculated into 10 l of RPMI 1640 containing 2% glucose and 10 µg ml−1 FeSO4·7H2O and grown for 48 h. Bacteria were pelleted by centrifugation, washed and resuspended with phosphate-buffered saline (PBS), pH 7.4, and sonicated for 4 h with a Branson Sonifier 250 at a power level of 5 and 50% duty cycle with cooling, which resulted in lysis of the vast majority of the bacteria (as determined by observation with light microscopy). Unbroken bacteria were removed by centrifugation and the resultant supernatant was heated at 100°C for 10 min to inactivate any potential glycosidases present. Coagulated material was removed by centrifugation, and additional proteins were precipitated by the dropwise addition of cold ethanol to a concentration of 30% with cooling in a dry ice/ethanol bath with stirring for 10 min and without stirring for an additional 15 min. Precipitated proteins were pelleted, and the supernatant was subsequently brought to 80% ethanol with cooling in a dry ice/ethanol bath as above to precipitate carbohydrates [20]. The 80% ethanol pellet was suspended in 10 mM Tris, pH 7.0 and treated with 0.1 mg ml−1 DNase, 0.5 mg ml−1 RNase and 5 mM MgCl2 overnight at 37°C, followed by 1 mg ml−1 Pronase and 0.1 mg ml−1 proteinase K also overnight at 37°C. Precipitate was redissolved by acidification to pH 2 with two drops of concentrated HCl, and the solution was subsequently brought back to pH 7.0 using 10 N NaOH. The solution was mixed with an equal volume of phenol/chloroform/isoamyl alcohol in a ratio of 25:24:1 with agitation and centrifuged briefly. The upper aqueous layer containing the majority of the carbohydrates was treated with 1 mg ml−1 DNase overnight at 37°C, followed by extraction with phenol/chloroform/isoamyl alcohol, as above. The upper aqueous layer was applied to a Biogel P-100 column (Bio-Rad, Hercules, CA) and eluted with 10 mM Tris, pH 7.0, and fractions at this stage and in subsequent steps were analyzed for carbohydrate by the phenol/H2SO4 method [21] and for protein by the bicinchoninic acid method (Pierce, Rockford, IL) [22]. The major carbohydrate-containing fractions, which eluted rapidly and contained relatively little protein, were pooled, the pH was adjusted to pH 8.5, and this was applied to a DEAE-Sephacel column, washed with 10 mM Tris, pH 8.5, and eluted with a gradient of 0–2.0 M NaCl in 10 mM Tris, pH 8.5. The major carbohydrate-containing fractions, which eluted with approximately 0.35 M NaCl, were pooled, lyophilized, suspended in 10 mM ammonium acetate and desalted by passage through a Biogel P-4 column. Carbohydrate-containing fractions were pooled and reextracted with phenol/chloroform/isoamyl alcohol, as above, then the upper aqueous layer was lyophilized and resuspended in distilled water. The final sample contained approximately 12 mg ml−1 carbohydrate. Purity of the preparation was determined with the use of SDS–PAGE and silver staining (Bio-Rad), which revealed no protein bands. In addition, agarose gel electrophoresis followed by staining with ethidium bromide showed no evidence of DNA or RNA. 2.4 Carbohydrate analysis The monosaccharide composition of the purified carbohydrate was determined by gas chromatography/mass spectrometry analysis of the trimethylsilyl derivatives of the methyl glycosides (courtesy of Roberta K. Merkle at the Complex Carbohydrate Research Center, University of Georgia) [23]. Trimethylsilyl glycosides were prepared by methanolysis of the carbohydrate in 1 M HCl in methanol, followed by N-acetylation with pyridine and acetic anhydride. Samples were then treated with Tri-Sil and analyzed with a Hewlett Packard 5890 gas chromatograph coupled to a 5970 mass spectrometer. 2.5 Bacterial surface protein extraction Extraction of surface proteins from E. faecium TX0016 was performed using Zwittergent 3–12 (Calbiochem, San Diego, CA), as previously described [24]. In brief, overnight grown bacteria were agitated for 1 h at room temperature with 0.2% Zwittergent 3–12 in PBS. Whole bacteria were removed by centrifugation and the resultant supernatant was dialyzed overnight at 4°C against 50 mM Tris, pH 7.5. Surface proteins were lyophilized and resuspended in water to a final concentration of 1.4 mg ml−1 protein. The resultant extract contained <0.2 mg ml−1 carbohydrate. 2.6 Neutrophil isolation PMNs were isolated from EDTA-anticoagulated blood of healthy volunteers by dextran sedimentation, Ficoll-Hypaque centrifugation and hypotonic lysis of residual erythrocytes [25] and resuspended in HBSS. Cells were =95% neutrophils by Diff-Quick (Baxter Scientific Products, Miami, FL) staining, and viability was =96% by trypan blue exclusion. 2.7 Sera For generation of immune rabbit serum [8], E. faecium TX0016 was grown in 20 ml of 50% heat inactivated horse serum in BHI for 18 h. Bacteria were harvested by centrifugation at 6000×g for 10 min and suspended in 20 ml of 0.6% formalin in 0.9% NaCl. Cells were incubated overnight at 4°C, washed twice, and resuspended in 2 ml of 0.9% NaCl. A New Zealand White rabbit (Bethyl Laboratories, Montgomery, TX) was injected intravenously twice weekly for 4 months with 0.5 ml of the formalin-killed whole-cell suspension diluted 1:10 in normal saline, resulting in approximately 2×109 bacteria per injection. Blood was obtained before (preimmune rabbit serum, PRS) and after (immune rabbit serum, IRS) immunization, and serum was shipped to us, aliquoted and stored at −70°C. IRS was examined for reactivity with TX0016 carbohydrate by Ouchterlony gel immunodiffusion [26] and for reactivity with TX0016 surface proteins by Western blot [24], as previously described. For Western blots, membranes were incubated with IRS at a dilution of 1:200. For normal human serum (NHS), blood was obtained from healthy adult volunteers and allowed to clot in glass tubes for 20 min at room temperature. After centrifugation at 1500×g for 20 min at 4°C, NHS was used in experiments the same day with autologous PMNs. In some experiments, serum was heated at 56°C for 30 min to inactivate complement or treated with 10 mM MgCl2 and 10 mM EGTA to inactivate the classical complement pathway while maintaining alternative pathway activity. Immune rabbit serum was also generated by immunizing with formalin-treated E. faecium MCV161 or TH1-7, as described above. 2.8 Adsorption of sera Antibodies were adsorbed from immune rabbit serum by incubation with heat-killed E. faecium TX0016 (same strain as used for immunization), or E. faecium RLA-3 as a control. Bacteria were grown in BHI broth at 37°C overnight with end-over-end rotation, centrifuged, resuspended in distilled water, and boiled for 10 min. After incubation with approximately 2×109 heat-killed bacteria ml−1 of serum at 4°C for 2 h, the adsorbed serum was centrifuged at 10 000×g for 10 min at 4°C to pellet bacteria and the serum was stored at −70°C. Alternatively, antibodies were adsorbed by incubation with E. faecium TX0016 purified carbohydrate or surface proteins. Immune rabbit serum was incubated with 0.66 mg ml−1 purified carbohydrate or surface proteins for 1 h at 4°C. This mixture was then used immediately for opsonization of bacteria. To assess the effect of adsorption with purified carbohydrate or surface proteins, Western blots were performed, as noted above [24], using the Zwittergent surface protein extract as the antigen and 1:200 dilutions of IRS, IRS+purified surface proteins, IRS+purified carbohydrate, or PRS as the antibody source. 2.9 Preparation of IgG, F(ab′)2 and Fab fragments IgG was prepared from anti-TX0016 rabbit antiserum by binding to a protein A-Sepharose column and subsequent elution, following the manufacturer's instructions (Pierce). F(ab′)2 and Fab fragments were generated by digestion of purified IgG with Sepharose-immobilized pepsin and papain, respectively (ImmunoPure preparation kits, Pierce), with removal of undigested IgG and Fc fragments by binding to a protein A-Sepharose column. Fragments were dialyzed against 10 mM ammonium acetate, lyophilized, and resuspended in distilled water. The quality of purification of IgG and the complete digestion of IgG and purification of fragments was assessed by SDS–PAGE and silver staining (Bio-Rad). 2.10 Bacterial binding to PMNs Bacterial binding to PMNs was examined using a fluorescence microscopy assay, as previously described [11]. In brief, bacteria were labeled by incubation with 0.1% fluorescein isothiocyanate in 50 mM sodium carbonate buffer, pH 9.6, for 30 min at 37°C while protected from light. Enterococci were washed twice, suspended in HBSS, then opsonized with 10% serum, IgG (final concentration 0.025–6.2 mg protein ml−1), F(ab′)2 fragments (0.01–0.6 mg ml−1) or Fab fragments (0.03–0.1 mg ml−1), or combinations thereof, at 37°C for 15 min. After opsonization, bacteria were washed and resuspended in HBSS plus 2 mM Ca2+ and 2 mM Mg2+. 200 µl opsonized fluorescein-labeled enterococci (2×108 ml−1) were mixed with 200 µl neutrophils (2×107/ml) and incubated for 30 min at 37°C. Aliquots of 100 µl were removed and 5 µl of ethidium bromide was added to a final concentration of 50 µg ml−1. 10 µl of the mixture was placed on a slide with a coverslip and the samples were viewed using a Nikon Optiphot fluorescence microscope. Twenty-five consecutive individual PMNs per sample were examined and the number of bound bacteria was measured as the combined number of ingested and attached organisms per PMN. For experiments examining the effect of antibodies against PMN complement receptors, antibodies or buffer controls were added to PMNs 20 min prior to the addition of opsonized, labeled bacteria. 2.11 PMN-mediated killing Enterococci were opsonized with 10% serum, specific anti-TX0016 IgG or non-specific rabbit IgG (purchased from Sigma) (both at 2.0 mg ml−1), or combinations thereof, at 37°C for 15 min. After opsonization, bacteria were washed and resuspended in HBSS plus 2 mM Ca2+ and 2 mM Mg2+. 100 µl opsonized enterococci (2×108/ml) were mixed with 100 ml neutrophils (2×107 ml−1) and incubated at 37°C with end-over-end rotation. At the indicated times, 20-µl aliquots were diluted with 180 µl of distilled water for 10 min to lyse the PMNs and release the viable intracellular bacteria, followed by serial dilution in 0.1 M Na2SO4. Colony counts were determined by the pour plate method using BHI agar. 2.12 Statistical analysis Statistical significance was assessed using Student's two-tailed t-test, and differences were considered significant for P<0.05. 3 Results E. faecium TX0016 was completely resistant to PMN-mediated phagocytosis after opsonization with pre-immune rabbit serum (Fig. 1A). However, immune rabbit serum generated against formalin-killed whole TX0016 was able to overcome this resistance to phagocytosis (Fig. 1B), with a marked increase in bacterial binding to PMNs. Quantitation of attached or ingested bacteria revealed that the number of bacteria bound to PMNs increased from 0±0 bacteria/PMN (mean±S.E.M.) with PRS to 9.7±0.4 bacteria/PMN with IRS (P<0.001) (Fig. 2A). Heat treatment of IRS (56°C for 30 min) almost completely eliminated its ability to promote PMN binding (0.8 bacteria/PMN with heat-treated IRS), suggesting that complement was required for this activity. However, the opsonic activity of IRS was also dependent on serum components specifically directed against TX0016 organisms, as adsorption with heat-killed TX0016 markedly reduced the ability of IRS to promote PMN binding, while adsorption with the phagocytosis-susceptible strain of E. faecium RLA3 had little effect (Fig. 2A). Figure 1 View largeDownload slide Bacterial binding to PMNs — fluorescence microscopy. E. faecium TX0016 organisms were labelled with fluorescein isothiocyanate, opsonized with 10% preimmune or immune rabbit serum and exposed to PMNs, and representative fluorescence micrographs are shown. Magnification ×950. A: Preimmune rabbit serum; no bacteria are bound to PMNs. B: Immune rabbit serum; many bacteria are bound to PMNs. Figure 1 View largeDownload slide Bacterial binding to PMNs — fluorescence microscopy. E. faecium TX0016 organisms were labelled with fluorescein isothiocyanate, opsonized with 10% preimmune or immune rabbit serum and exposed to PMNs, and representative fluorescence micrographs are shown. Magnification ×950. A: Preimmune rabbit serum; no bacteria are bound to PMNs. B: Immune rabbit serum; many bacteria are bound to PMNs. Figure 2 View largeDownload slide Promotion of bacterial binding by IRS. Bacteria were opsonized with (A) anti-TX0016 immune serum (IRS), pre-immune serum (PRS), or immune serum adsorbed with either TX0016 or a phagocytosis-susceptible strain of E. faecium (RLA-3), or (B) IRS alone or with purified carbohydrate or a surface protein Zwittergent extract from TX0016, and exposed to PMNs as in Fig. 1, and data are presented as the number of bacteria bound per PMN (mean+S.E.M., n=3–7). *P<0.001 vs. IRS. Figure 2 View largeDownload slide Promotion of bacterial binding by IRS. Bacteria were opsonized with (A) anti-TX0016 immune serum (IRS), pre-immune serum (PRS), or immune serum adsorbed with either TX0016 or a phagocytosis-susceptible strain of E. faecium (RLA-3), or (B) IRS alone or with purified carbohydrate or a surface protein Zwittergent extract from TX0016, and exposed to PMNs as in Fig. 1, and data are presented as the number of bacteria bound per PMN (mean+S.E.M., n=3–7). *P<0.001 vs. IRS. IRS contained antibodies that reacted with multiple TX0016 surface proteins, as determined by Western blot (Fig. 3), and with TX0016 carbohydrate, as determined by Ouchterlony gel immunodiffusion (data not shown). However, while incubation of IRS with 0.66 mg ml−1 carbohydrate purified from TX0016 completely eliminated its ability to promote bacterial binding to PMNs, incubation with the same concentration of surface proteins extracted from TX0016 reduced bacterial-PMN binding only slightly (Fig. 2B, P>0.05). Incubation of IRS with purified carbohydrate did not appreciably affect its reactivity with TX0016 surface proteins, as determined by Western blot (Fig. 3), while incubation with purified surface proteins resulted in markedly fewer bands on Western blot. Monosaccharide analysis of carbohydrate purified from TX0016 revealed that it was composed of approximately 47% fucose and 41% glucose, with lesser amounts of other monosaccharides (6% galactose, 3% rhamnose, 2%N-acetylgalactosamine, 1%N-acetylglucosamine). Figure 3 View largeDownload slide Western blot using immune serum±adsorption. The antigen consisted of a surface protein extract from TX0016. Lane 1, IRS incubated with 0.66 mg ml−1 purified carbohydrate; lane 2, IRS alone; lane 3, IRS incubated with 0.66 mg ml−1 purified surface proteins; lane 4, PRS; lane 5, molecular mass standards. Figure 3 View largeDownload slide Western blot using immune serum±adsorption. The antigen consisted of a surface protein extract from TX0016. Lane 1, IRS incubated with 0.66 mg ml−1 purified carbohydrate; lane 2, IRS alone; lane 3, IRS incubated with 0.66 mg ml−1 purified surface proteins; lane 4, PRS; lane 5, molecular mass standards. IRS also promoted PMN-mediated killing of TX0016. While opsonization with either NHS or PRS had no significant ability to promote PMN-mediated loss of bacterial viability over a 2-h span, opsonization with IRS resulted in a 86±5% decline in bacterial viability after 2 h exposure to PMNs (Fig. 4). As with bacterial binding to PMNs, promotion of PMN-mediated killing was also markedly reduced by incubation of IRS with TX0016 carbohydrate, but not by incubation with TX0016 surface proteins. Figure 4 View largeDownload slide Promotion of PMN-mediated killing by IRS. Bacteria were opsonized with the indicated serum (in the absence of an additional complement source) and exposed to PMNs, and colony counts were determined at the indicated time points. Data are presented as the percent of the initial viability (mean+S.E.M., n=3−6). *P<0.05 vs. PRS; †P<0.05 vs. IRS+CHO. Figure 4 View largeDownload slide Promotion of PMN-mediated killing by IRS. Bacteria were opsonized with the indicated serum (in the absence of an additional complement source) and exposed to PMNs, and colony counts were determined at the indicated time points. Data are presented as the percent of the initial viability (mean+S.E.M., n=3−6). *P<0.05 vs. PRS; †P<0.05 vs. IRS+CHO. To investigate the mechanism of this opsonization, we purified IgG from IRS and demonstrated that these antibodies by themselves could not promote phagocytosis of E. faecium TX0016 (Fig. 5). However, when IgG was present in conjunction with NHS as a complement source, bacterial adherence to PMNs was markedly promoted (Fig. 5). The added promotional effect of NHS was likely due to the presence of active complement, as heat treatment of NHS eliminated its ability to increase bacterial adhesion promoted by IgG (1.0 bacteria/PMN with IgG+heat-treated NHS). An intact alternative complement pathway was sufficient, as Mg-EGTA treatment of NHS did not significantly alter its ability to increase bacterial adhesion promoted by IgG (data not shown). Specific anti-TX0016 IgG, when combined with NHS, also promoted PMN-mediated killing (Fig. 6), while non-specific rabbit IgG was unable to promote killing, even in the presence of NHS. Figure 5 View largeDownload slide Promotion of bacterial binding by IgG, F(ab′)2 or Fab fragments. Bacteria were opsonized with IgG purified from IRS (final IgG concentration 6.2 mg ml−1), F(ab′)2 fragments (0.1 mg ml−1) or Fab fragments (0.1 mg ml−1)±10% NHS as a complement source, and the number of bacteria per PMN was determined as in Fig. 2 (n=3–7). *P<0.001 vs. NHS alone. Figure 5 View largeDownload slide Promotion of bacterial binding by IgG, F(ab′)2 or Fab fragments. Bacteria were opsonized with IgG purified from IRS (final IgG concentration 6.2 mg ml−1), F(ab′)2 fragments (0.1 mg ml−1) or Fab fragments (0.1 mg ml−1)±10% NHS as a complement source, and the number of bacteria per PMN was determined as in Fig. 2 (n=3–7). *P<0.001 vs. NHS alone. Figure 6 View largeDownload slide Promotion of PMN-mediated killing by specific anti-TX0016 IgG. Bacteria were opsonized with either specific anti-TX0016 IgG or non-specific rabbit IgG (both at 2.0 mg ml−1) ±10% NHS and exposed to PMNs. Data are presented as in Fig. 4. *P<0.05 vs. Nl IgG. Figure 6 View largeDownload slide Promotion of PMN-mediated killing by specific anti-TX0016 IgG. Bacteria were opsonized with either specific anti-TX0016 IgG or non-specific rabbit IgG (both at 2.0 mg ml−1) ±10% NHS and exposed to PMNs. Data are presented as in Fig. 4. *P<0.05 vs. Nl IgG. F(ab′)2 fragments or Fab fragments generated from specific anti-TX0016 IgG were similarly unable to promote PMN adherence by themselves but could do so when combined with NHS (Fig. 5), and Mg-EGTA-treated NHS was also able to promote binding to PMNs induced by F(ab′)2 fragments (Fig. 7). Bacterial binding to PMNs declined as the concentration of either IgG or F(ab′)2 fragments was lowered in combination with Mg-EGTA-treated NHS (Fig. 7). However, the maximum binding promoted by F(ab′)2 fragments and Mg-EGTA-treated NHS was less than that promoted by an equimolar concentration of IgG- and Mg-EGTA-treated NHS (Fig. 7), and higher concentrations of F(ab′)2 fragments (0.6 mg ml−1) yielded no further increase in bacterial binding to PMNs (data not shown). Figure 7 View largeDownload slide Bacterial binding is proportional to F(ab′)2 and IgG concentrations. Bacteria were opsonized with either F(ab′)2 fragments or IgG at the indicated concentrations along with 10% Mg-EGTA-treated NHS and the number of bacteria per PMN was determined (mean+S.D., n=2). Figure 7 View largeDownload slide Bacterial binding is proportional to F(ab′)2 and IgG concentrations. Bacteria were opsonized with either F(ab′)2 fragments or IgG at the indicated concentrations along with 10% Mg-EGTA-treated NHS and the number of bacteria per PMN was determined (mean+S.D., n=2). As these results suggested that deposition of complement was likely involved in the ability of specific anti-TX0016 IgG, F(ab′)2 and Fab fragments to overcome resistance to phagocytosis, we examined the importance of PMN complement receptors in this process by pre-incubating PMNs with antibodies directed against components of these receptors. Pre-incubation of PMNs with either of two antibodies (M1/70 and Mo1 clone 44) against CD11b, one of the components of complement receptor type 3, inhibited bacterial binding to PMNs promoted by the combination of F(ab′)2 fragments and NHS by 62% and 70%, respectively (Fig. 8). However, two other anti-CR3 antibodies, 60.1 against CD11b and R15.7 against CD18, had little effect (2% inhibition with 60.1 and 10% increase in bacterial-PMN binding with R15.7). Preincubation of PMNs with polyclonal rabbit IgG against complement receptor type 1 did not significantly affect bacterial binding promoted by F(ab′)2 and NHS (Fig. 8), nor did the anti-CR1 mAb 543 (only 18% inhibition), and addition of polyclonal anti-CR1 to M1/70 yielded no significant additional inhibition beyond that seen with M1/70 alone (P>0.05, data not shown). Figure 8 View largeDownload slide Inhibition of bacterial binding by antibodies against PMN complement receptors. Bacteria opsonized with 0.1 mg ml−1 F(ab′)2 fragments+10% NHS were exposed to PMNs pre-incubated with the indicated antibody, and the number of bacteria per PMN was determined (mean+S.E.M., n=3–7). Antibodies: Ctrl, HBSS control; M1/70, anti-CD11b; Mo1 clone 44, anti-CD11b; 60.1, anti-CD11b; W6/32, anti-HLA; anti-CR1, polyclonal rabbit IgG against CR1. *P<0.01 vs. HBSS control. Figure 8 View largeDownload slide Inhibition of bacterial binding by antibodies against PMN complement receptors. Bacteria opsonized with 0.1 mg ml−1 F(ab′)2 fragments+10% NHS were exposed to PMNs pre-incubated with the indicated antibody, and the number of bacteria per PMN was determined (mean+S.E.M., n=3–7). Antibodies: Ctrl, HBSS control; M1/70, anti-CD11b; Mo1 clone 44, anti-CD11b; 60.1, anti-CD11b; W6/32, anti-HLA; anti-CR1, polyclonal rabbit IgG against CR1. *P<0.01 vs. HBSS control. To examine whether the activity of IRS directed against TX0016 could be generalized to other phagocytosis-resistant strains of E. faecium, we generated IRS against two other phagocytosis-resistant strains, MCV161 and TH1-7, and defined the ability of each IRS to promote phagocytosis of a variety of other strains. Anti-TX0016 IRS promoted phagocytosis of the original strain TX0016, and also SH-11 and 4586, but not MCV161, TH1-7, FA191, TX0015, RLA-6, and X25645. Anti-MCV161 IRS likewise promoted phagocytosis of the original strain MCV161, and also 4586, RLA-6, and TX0015, but not TX0016. Anti-TH1-7 IRS promoted phagocytosis of TX0015, RLA-6, and MCV161, but not TX0016, nor interestingly the original strain TH1-7. 4 Discussion In the present study, we have demonstrated that antibodies generated against whole, formalin—killed E. faecium TX0016 can promote opsonization of this phagocytosis-resistant organism with binding to PMNs and subsequent PMN-mediated killing. These antibodies are likely directed against bacterial carbohydrates, as their opsonic activity for both binding to PMNs and PMN-mediated killing could be abrogated by incubation with a carbohydrate purified from TX0016, but not by incubation with a surface protein extract. As (a) the carbohydrate preparation appeared to be free of contaminating proteins, as determined by SDS–PAGE and silver staining, and (b) incubation of immune serum with purified carbohydrate revealed no diminution in reactivity with surface proteins by Western blot [24], while incubation with a surface protein extract resulted in marked loss of immunoreactivity, it is very likely that these opsonizing antibodies are directed against TX0016 carbohydrates and not against surface proteins. While a strain of E. faecalis has been shown by Pazur and colleagues to contain a diheteroglycan composed of a main chain of trisaccharide units (glucose-β(1-6)-glucose-β(1-4)-galactose) joined by β(1,4) linkages with lactosyl and cellobiosyl side chains [27], and a tetraheteroglycan composed of galactose, rhamnose, N-acetylgalactosamine, and β-d-glucose-1-phosphate [28], no characterization of surface carbohydrates of E. faecium has been previously available. Binding of the Fc portion of specific IgG with PMN Fc receptors does not appear to be sufficient for bacterial-PMN adhesion, as whole IgG by itself was unable to promote bacterial binding to or killing by PMNs. However, whole IgG, F(ab′)2 fragments, or Fab fragments were able to promote PMN binding when a complement source was present, and the alternative pathway of complement was sufficient. CR3 most likely is the PMN receptor involved, as either of two mAbs against CD11b could inhibit bacterial binding, while antibodies against CR1 had little effect. These results contrast with those seen in a study using different capsular serotypes of group B streptococci, where both CR1 and CR3 were shown to be important in binding of opsonized bacteria to PMNs [29]. While we found that another anti-CD11b antibody and an anti-CD18 antibody did not block bacterial binding, differences in inhibitory activity of different anti-CR3 antibodies are not surprising, as CR3 is a large multi-functional molecule, these antibodies bind to different epitopes thereon, and different anti-CD11b antibodies may have differing abilities to block binding of iC3b-opsonized particles [14,29,30]. M1/70 and Mo1 clone 44, the two anti-CR3 antibodies with inhibitory activity in the present study, block both iC3b and lectin binding sites on CR3, and this may be important in their ability to inhibit the binding of TX0016 to PMNs. As CR3 is the primary binding site for iC3b-opsonized particles [31], while CR1 serves as the receptor for C3b-opsonized particles, these results also suggest that iC3b is the critical opsonin in this process. However, we did not directly measure complement deposition, and further experiments are planned to investigate the mechanism of complement-mediated opsonization. Bacteria use a variety of mechanisms to resist opsonization [32,33], and that used by E. faecium TX0016 has not been completely delineated. For example, a bacterial surface component may inhibit amplification of the complement cascade, as is true for sialic acid on the surface of group B streptococci [34] which renders bound C3b accessible to control by factor H and promotes dissolution of the C3b convertase [35]. Antibodies could promote adherence of E. faecium TX0016 to PMNs by directly binding to such sites on the bacterial surface and interfering with this inhibition, thus converting the bacteria from a non-activating to an activating surface for the alternative complement pathway, as occurs with antibodies against sialylated group B streptococci [36]. Alternatively, antibodies could bind to other sites and themselves promote complement activation [37]. Our results using F(ab′)2 fragments do not distinguish between these two possibilities, as these fragments themselves can promote activation of the alternative complement pathway [37–39]. We also examined the effect of Fab fragments, as classically these have been thought to be unable to promote complement activation and opsonization [40]. Using this assumption, our results demonstrating promotion of bacterial adherence by Fab fragments in the presence of NHS would suggest that these fragments are directly binding to sites on the bacterial surface that intrinsically interfere with complement activation. However, some studies have suggested that Fab fragments themselves may promote alternative complement pathway activation [41] and cytotoxicity [42]. Thus we cannot yet completely define the mechanism of antibody-mediated opsonization for E. faecium TX0016. It is not known whether the results obtained in this study can be generalized to other phagocytosis-resistant strains of E. faecium. Preliminary experiments suggest that there is some cross reactivity in the ability of IRS generated against one phagocytosis-resistant strain to promote phagocytosis of other strains. However, this cross reactivity is limited and quite variable, even in the small number of immune sera and strains tested here. This suggests either that the mechanism used for resistance to phagocytosis varies amongst strains of E. faecium or more likely that opsonic antibodies are directed against discrete targets on different strains. The latter system could be analogous to that of the different capsular serotypes present on group B streptococci or Streptococcus pneumoniae. Resistance to phagocytosis is a likely virulence factor used by some strains of E. faecium, as it is for a variety of other organisms. Due to the increasing prevalence of E. faecium infections, along with the spread of multiply antibiotic-resistant isolates, the need to develop alternative treatment strategies will grow. These studies demonstrating that antibodies can overcome resistance to PMN-mediated phagocytosis and killing, most likely by promoting deposition of complement, suggest that treatment strategies that improve antibody-mediated opsonization, such as passive or active immunization, might be beneficial in preventing or altering the course of infections due to these bacteria. Acknowledgements This work was supported by Grants-in-Aid from the American Heart Association, Texas Affiliate and National (to R.M.R.) and NIH Grant AI42399 (to B.E.M.). Monosaccharide analysis at the Complex Carbohydrate Research Center was supported in part by the NIH-funded Resource Center for Biomedical Complex Carbohydrates (NIH Grant P41-RR05351). We thank M. Mariscalco, R. Todd, R. Rothlein, and I. Gigli for providing antibodies; R. Jordon for use of the fluorescence microscope; F. Wold for assistance with carbohydrate purification; R. Merkle for monosaccharide analysis; and I. Gigli for helpful comments. A portion of this work was presented at the 1995 American Federation for Clinical Research meeting, San Diego, CA (J. Invest. Med. (1995) 43 (Suppl. 2), 288A). References [1] Murray B.E. ( 1990) The life and times of the Enterococcus. Clin. Microbiol. Rev.  3, 46– 65. Google Scholar CrossRef Search ADS PubMed  [2] Schaberg D.R. Culver D.H. Gaynes R.P. ( 1991) Major trends in the microbial etiology of nosocomial infection. Am. J. Med.  91, 72S– 75S. Google Scholar CrossRef Search ADS PubMed  [3] Leclercq R. Courvalin P. ( 1997) Resistance to glycopeptides in enterococci. Clin. Infect. Dis.  24, 545– 556. Google Scholar CrossRef Search ADS PubMed  [4] Murray B.E. ( 1997) Vancomycin-resistant enterococci. Am. J. Med.  102, 284– 293. Google Scholar CrossRef Search ADS PubMed  [5] Jones R.N. Marshall S.A. Pfaller M.A. Wilke W.W. Hollis R.J. Erwin M.E. Edmond M.B. Wenzel R.P. ( 1997) Nosocomial enterococcal blood stream infections in the SCOPE program: antimicrobial resistance, species occurrence, molecular testing, results, and laboratory testing accuracy. Diagn. Microbiol. Infect. Dis.  29, 95– 102. Google Scholar CrossRef Search ADS PubMed  [6] Clark N. Cooksey R. Hill B. Swenson J. Tenover F. ( 1993) Characterization of glycopeptide-resistant enterococci from US hospitals. Antimicrob. Agents Chemother.  37, 2311– 2317. Google Scholar CrossRef Search ADS PubMed  [7] Huycke M.M. Sahm D.F. Gilmore M.S. ( 1998) Multiple-drug resistant enterococci: the nature of the problem and an agenda for the future. Emerging Infect. Dis.  4, 239– 249. Google Scholar CrossRef Search ADS PubMed  [8] Arduino R.C. Murray B.E. Rakita R.M. ( 1994) Roles of antibodies and complement in phagocytic killing of Enterococci. Infect. Immun.  62, 987– 993. Google Scholar PubMed  [9] Harvey B.S. Baker C.J. Edwards M.S. ( 1992) Contributions of complement and immunoglobulin to neutrophil-mediated killing of enterococci. Infect. Immun.  60, 3635– 3640. Google Scholar PubMed  [10] Gaglani M.J. Baker C.J. Edwards M.S. ( 1997) Contribution of antibody to neutrophil-mediated killing of Enterococcus faecalis. J. Clin. Immunol.  17, 478– 484. Google Scholar CrossRef Search ADS PubMed  [11] Arduino R.C. Jacques-Palaz K. Murray B.E. Rakita R.M. ( 1994) Resistance of Enterococcus faecium to neutrophil-mediated phagocytosis. Infect. Immun.  62, 5587– 5594. Google Scholar PubMed  [12] Anderson D.C. Miller L.J. Schmalstieg F.C. Rothlein R. Springer T.A. ( 1986) Contributions of the Mac-1 glycoprotein family to adherence-dependent granulocytic functions: structure-function assessments employing subunit-specific monoclonal antibodies. J. Immunol.  137, 15– 27. Google Scholar PubMed  [13] Wallis W.J. Hickstein D.D. Schwartz B.R. June C.H. Ochs H.D. Beatty P.G. Klebanoff S.J. Harlan J.M. ( 1986) Monoclonal antibody-defined functional epitopes on the adhesion-promoting glycoprotein complex (CDw18) of human neutrophils. Blood  67, 1007– 1013. Google Scholar PubMed  [14] Dana N. Styrt B. Griffin J.D. Todd R.F. Klempner M.S. III Arnaout M.A. ( 1986) Two functional domains in the phagocyte membrane glycoprotein Mo1 identified with monoclonal antibodies. J. Immunol.  137, 3259– 3263. Google Scholar PubMed  [15] Entman M.L. Youker K. Shoji T. Kukielka G. Shappell S.B. Taylor A.A. Smith C.W. ( 1992) Neutrophil induced oxidative injury of cardiac myocytes. A compartmented system requiring CD11b/CD18-ICAM-1 adherence. J. Clin. Invest.  90, 1335– 1345. Google Scholar CrossRef Search ADS PubMed  [16] Barnstable C.J. Bodmer W.F. Brown G. Galfre G. Milstein C. Williams A.F. Ziegler A. ( 1978) Production of monoclonal antibodies to group A erythrocytes, HLA and other human cell surface antigens-new tools for genetic analysis. Cell  14, 9– 20. Google Scholar CrossRef Search ADS PubMed  [17] Tausk F.A. McCutchan J.A. Spechko P. Schreiber R.D. Gigli I. ( 1986) Altered erythrocyte C3b receptor expression, immune complexes, and complement activation in homosexual men in varying risk groups for acquired immune deficiency syndrome. J. Clin. Invest.  78, 977– 982. Google Scholar CrossRef Search ADS PubMed  [18] Miranda A.G. Singh K.V. Murray B.E. ( 1991) DNA fingerprinting of Enterococcus faecium by pulsed-field gel electrophoresis may be a useful epidemiologic tool. J. Clin. Microbiol.  29, 2752– 2757. Google Scholar PubMed  [19] Wessels M.R. Rubens C.E. Benedi V. Kasper D.L. ( 1989) Definition of a bacterial virulence factor: Sialylation of the group B streptococcal capsule. Proc. Natl. Acad. Sci. USA  86, 8983– 8987. Google Scholar CrossRef Search ADS   [20] Wessels M.R. Paoletti L.C. Kasper D.L. DiFabio J.L. Michon F. Holme K. Jennings H.J. ( 1990) Immunogenicity in animals of a polysaccharide-protein conjugate vaccine against type III group B Streptococcus. J. Clin. Invest.  86, 1428– 1433. Google Scholar CrossRef Search ADS PubMed  [21] Dubois M.K. Gilles K.A. Hamilton J.K. Rebers P.A. Smith S. ( 1956) Colorimetric method for determination of sugars and related substances. Anal. Chem.  28, 350– 356. Google Scholar CrossRef Search ADS   [22] Hinson D.L. Webber R.J. ( 1988) Miniaturization of the BCA protein assay. BioTechniques  6, 14– 16. Google Scholar PubMed  [23] Merkle R.K. Poppe I. ( 1994) Carbohydrate composition analysis of glycoconjugates by gas-liquid chromatography/mass spectrometry. Methods Enzymol.  230, 1– 15. Google Scholar CrossRef Search ADS PubMed  [24] Sulaiman A. Rakita R.M. Arduino R.C. Patterson J.E. Steckelberg J.M. Singh K.V. Murray B.E. ( 1996) Serology of enterococcal infections using Western blot. Eur. J. Clin. Microbiol. Infect. Dis.  15, 826– 829. Google Scholar CrossRef Search ADS PubMed  [25] Rosen H. Michel B.R. Chait A. ( 1991) Phagocytosis of opsonized oil droplets by neutrophils. J. Immunol. Methods  144, 117– 125. Google Scholar CrossRef Search ADS PubMed  [26] Tan E.M. Kunkel H.G. ( 1966) Characteristics of a soluble nuclear antigen precipitating with sera of patients with systemic lupus erythematosus. J. Immunol.  96, 464– 471. Google Scholar PubMed  [27] Pazur J.H. Cepure A. Kane J.A. Hellerquist C.G. ( 1973) Glycans from streptococcal cell walls: the molecular structure of an antigenic diheteroglycan of glucose and galactose from Streptococcus faecalis. J. Biol. Chem.  248, 279– 284. Google Scholar PubMed  [28] Pazur J.H. ( 1982) β-d-Glucose-1-phosphate: a structural unit and an immunological determinant of a glycan from streptococcal cell walls. J. Biol. Chem.  257, 589– 591. Google Scholar PubMed  [29] Smith C.L. Baker C.J. Anderson D.C. Edwards M.S. ( 1990) Role of complement receptors in opsonophagocytosis of Group B Streptococci by adult and neonatal neutrophils. J. Infect. Dis.  162, 489– 495. Google Scholar CrossRef Search ADS PubMed  [30] Diamond M.S. Garcia-Aguilar J. Bickford J.K. Corbi A.L. Springer T.A. ( 1993) The I domain is a major recognition site on the leukocyte integrin Mac-1 (CD11b/CD18) for four distinct adhesion ligands. J. Cell Biol.  120, 1031– 1043. Google Scholar CrossRef Search ADS PubMed  [31] Wright S.D. ( 1992) Receptors for complement and the biology of phagocytosis. in: Inflammation: Basic Principles and Clinical Correlates , 2nd edn. ( Gallin J.I. Goldstein I.M. Snyderman R., Eds.), pp. 477– 495. Raven Press, New York. [32] Moffitt M. Frank M. ( 1994) Complement resistance in microbes. Springer Semin. Immunopathol.  15, 327– 344. [33] Horstmann R.D. ( 1992) Target recognition failure by the nonspecific defense system: surface constituents of pathogens interfere with the alternative pathway of complement activation. Infect. Immun.  60, 721– 727. Google Scholar PubMed  [34] Edwards M.S. Kasper D.L. Jennings H.J. Baker C.J. Nicholson-Weller A. ( 1982) Capsular sialic acid prevents activation of the alternative complement pathway by type III, group B streptococci. J. Immunol.  128, 1278– 1283. Google Scholar PubMed  [35] Fearon D.T. ( 1978) Regulation by membrane sialic acid of b1H-dependent decay-dissociation of amplification C3 convertase of the alternative complement pathway. Proc. Natl. Acad. Sci. USA  75, 1971– 1975. Google Scholar CrossRef Search ADS   [36] Edwards M.S. Nicholson-Weller A. Baker C.J. Kasper D.L. ( 1980) The role of specific antibody in alternative complement pathway-mediated opsonophagocytosis of type III, group B Streptococcus. J. Exp. Med.  151, 1275– 1287. Google Scholar CrossRef Search ADS PubMed  [37] Ratnoff W.D. Fearon D.T. Austen K.F. ( 1983) The role of antibody in the activation of the alternative complement pathway. Springer Semin. Immunopathol.  6, 361– 371. [38] Reid K.B.M. ( 1971) Complement fixation by the F(ab′)2-fragment of pepsin-treated rabbit antibody. Immunology  20, 649– 658. Google Scholar PubMed  [39] Winkelstein J.A. Shin H.S. ( 1974) The role of immunoglobulin in the interaction of pneumococci and the properdin pathway: evidence for its specificity and lack of requirement for the Fc portion of the molecule. J. Immunol.  112, 1635– 1642. Google Scholar PubMed  [40] Trautmann M. Triest K. Hofstaetter T. Seiler F.R. Hahn H. ( 1985) Influence of different immunoglobulin G preparations on phagocytosis of Pseudomonas aeruginosa by polymorphonuclear granulocytes. Zbl. Bakteriol. Hyg. A  259, 104– 117. [41] Albar J.P. Juarez C. Vivanco-Martinez F. Bragado B. Ortiz F. ( 1981) Structural requirements of rabbit IgG F(ab′)2 fragment for activation of the complement system through the alternative pathway — I. disulfide bonds. Mol. Immunol.  18, 925– 934. Google Scholar CrossRef Search ADS PubMed  [42] Ehrnst A. ( 1978) Separate pathways of C activation by measles virus cytotoxic antibodies: subclass analysis and capacity of F(ab) molecules to activate C via the alternative pathway. J. Immunol.  121, 1206– 1212. Google Scholar PubMed  © 2000 Federation of European Microbiological Societies.
TI - Specific antibody promotes opsonization and PMN-mediated killing of phagocytosis-resistant Enterococcus faecium
JF - Journal of the Endocrine Society
DO - 10.1111/j.1574-695X.2000.tb01489.x
DA - 2000-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/specific-antibody-promotes-opsonization-and-pmn-mediated-killing-of-uNOv2E2dkD
SP - 291
EP - 299
VL - 28
IS - 4
DP - DeepDyve
ER -