Soluble MAC is primarily released from MAC-resistant bacteria that potently convert complement component C5
- Department of Medical Microbiology, University Medical Center Utrecht, Netherlands
- Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, University of Utrecht, Netherlands
- Netherlands Proteomics Center, Netherlands
Figures
Figure 1 with 2 supplements
Soluble membrane attack complex (sMAC) is primarily formed by MAC-resistant Gram-negative bacteria.
Escherichia coli strains (5×107 bacteria/ml) were incubated in 5% pooled human serum. Bacterial viability was determined after 60 min by counting colony forming units (CFUs) and calculating the survival compared to t=0. The horizontal dotted line represents the detection limit of the assay. (b) E. coli strains (5×107 bacteria/ml) were incubated in 10% C5-depleted serum. Bacteria were stained with AF488-labelled mouse monoclonal anti-C3b after 30 min and staining was measured by flow cytometry. The relative binding was calculated by normalizing the geoMFI to the geoMFI of unlabelled bacteria. (c) sMAC was detected in the reaction supernatant (100-fold diluted) by enzyme-linked immunosorbent assay (ELISA) after E. coli strains (5×108 bacteria/ml) were incubated in 5% C6-depleted serum supplemented with C6-biotin for 60 min. Serum without bacteria (serum only) was taken as background control. (d) sMAC was detected by ELISA for a dilution range of reaction supernatant collected in c for E. coli strains MG1655, 552059.1, and 547655.1. Orange strains are MAC-sensitive (MAC-sens), blue strains MAC-resistant (MAC-res), and black strains complement-resistant (comp-res). Flow cytometry data (b) are represented by individual geoMFI values of the bacterial population. Data represent mean ± SD (d) or individual values with mean ± SD (a, b, c) of three independent experiments.
Figure 1—figure supplement 1
Complement dependency of bacterial killing and C3a release in serum for Escherichia coli strains.
MAC-sensitive E. coli strains (5×108 bacteria/ml) were incubated in 5% pooled human serum supplemented with C5 conversion inhibitors 6 µg/ml OmCI and 6 µg/ml eculizumab. Bacterial viability was determined after 60 min by counting colony forming units (CFUs) and calculating the survival compared to t=0. The horizontal dotted line represents the detection limit of the assay. (b) MAC-sensitive (MAC-sens, orange) and MAC-resistant (MAC-res, blue) E. coli strains (5×107 bacteria/ml) were incubated in 5% pooled human serum. Supernatant was collected by centrifugation after 60 min. C3a in the supernatant was quantified by enzyme-linked immunosorbent assay (ELISA). Data represent individual values with mean ± SD of three independent experiments.
Figure 1—figure supplement 2
Validation specificity soluble membrane attack complex (sMAC) enzyme-linked immunosorbent assay (ELISA) and sMAC release by Klebsiella strains.
(a) sMAC was formed in 5% C6-depleted serum supplemented with C6-biotin incubated alone, with 10 µg/ml CVF or 6 µg/ml eculizumab for 60 min. Afterwards, the reaction mix was serially diluted and sMAC was detected by ELISA. Data represent mean ± SD of three independent experiments. (b) Escherichia coli strains (5×108 bacteria/ml) were incubated in 5% C6-depleted serum with C6-biotin, with and without C5 conversion inhibitors OmCI (6 µg/ml) and eculizumab (6 µg/ml). Supernatant was collected by centrifugation after 60 min. sMAC was detected in the reaction supernatant (100-fold diluted) by ELISA. (c) Klebsiella strains (5×108 bacteria/ml) were also incubated in 5% C6-depleted serum with C6-biotin and supernatant was collected by centrifugation after 60 min. sMAC was detected in the reaction supernatant (100-fold diluted) by ELISA. Orange strains are MAC-sensitive (MAC-sens), blue strains MAC-resistant (MAC-res), and black strains complement-resistant (comp-res). Serum without bacteria (serum only) was taken as background control. Data represent individual values with mean ± SD of three (b) or two (c, where a biological duplicate of KP567709.1 was measured in one experiment) independent experiments.
Figure 2 with 1 supplement
Gram-positive bacteria also form soluble membrane attack complex (sMAC) in serum.
MAC-sensitive Escherichia coli MG1655 and Gram-positive strains (5×108 bacteria/ml) were incubated in 5% C6-depleted serum supplemented with C6-biotin. The supernatant was collected after 60 min by centrifugation. Serum without bacteria (serum only) was taken as background control. sMAC was detected in a dilution range of reaction supernatant by enzyme-linked immunosorbent assay (ELISA). Data represent mean ± SD of three independent experiments.
Figure 2—figure supplement 1
C5a generation by Gram-positive bacteria in serum.
Membrane attack complex (MAC)-sensitive Escherichia coli MG1655 and Gram-positive strains (5×108 bacteria/ml) were incubated in 5% C6-depleted serum supplemented with C6-biotin. The supernatant was collected after 60 min by centrifugation. Serum without bacteria (serum only) was taken as background control. C5a in the supernatant was quantified by enzyme-linked immunosorbent assay (ELISA). Data represent individual values with mean ± SD of three independent experiments.
Figure 3 with 1 supplement
Membrane attack complex (MAC)-resistant Escherichia coli strains potently convert C5 in serum.
E. coli strains (5×108 bacteria/ml) were incubated in 5% pooled human serum and supernatant was collected by centrifugation after 60 min. (a) Representative Western blot for C5 of the supernatant. The upper band represents the α-chain of C5, the middle band of C5b (α’), and the lower band the β-chain of both C5 and C5b. Serum without bacteria (ser) was taken as control for the absence of C5 conversion. The Western blot is a representative of at least three independent experiments. (b) C5a in the supernatant was quantified by enzyme-linked immunosorbent assay (ELISA). Orange strains are MAC-sensitive (MAC-sens) and blue strains are MAC-resistant (MAC-res). E. coli CGSC7740 wildtype without lipopolysaccharide (LPS) O-antigen (O-Ag) (O-Agneg) and wbbL+ with LPS O-Ag (O-Agpos) were also incubated in 5% pooled human serum to collect supernatant. C5a (c) and sMAC (d) in the supernatant were quantified by ELISA. ELISA data represent individual values with mean ± SD of three independent experiments. Statistical analysis was done using an unpaired two-tailed t-test with the mean C5a concentrations of MAC-sensitive strains vs. MAC-resistant strains (b) or a paired two-tailed t-test on individual samples (c and d). Relevant p-values are indicated in the figure.
-
Figure 3—source data 1
Escherichia coli strains (5×108 bacteria/ml) were incubated in 5% pooled human serum and supernatant was collected by centrifugation after 60 min.
The supernatant was analyzed by Western blotting for C5. Serum without bacteria (serum only) was taken as control for the absence of C5 conversion. Ten nM C5 or pC5b6 were loaded as positive controls for C5 and C5b.
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig3-data1-v2.jpg
Figure 3—figure supplement 1
Lipopolysaccharide (LPS) O-antigen (O-Ag) expression of Gram-negative strains and its effect on C3b deposition.
Escherichia. coli strains (a) and Klebsiella strains (b) were typed for the presence of LPS O-Ag via silver staining (methods described in Doorduijn et al., 2021). LPS-core were distinguished from LPS O-Ag based on size. Orange strains are MAC-sensitive (MAC-sens), blue strains MAC-resistant (MAC-res), and black strains complement-resistant (comp-res). The silver stain was previously shown in part (only for silver MAC-res and MAC-sens E. coli strains) in Doorduijn et al., 2021. (c) 5×107 bacteria/ml of E. coli strain CGSC7740 wildtype (O-Agneg) and wbbL+ (O-Agpos) were incubated in a titration of C5-depleted serum for 30 min. Bacteria were washed and stained with AF488-labelled mouse monoclonal anti-C3b for 30 min. Antibody staining was measured by flow cytometry. Flow cytometry data are represented by individual geoMFI values of the bacterial population. Data represent individual values with mean ± SD of two independent experiments.
-
Figure 3—figure supplement 1—source data 1
Escherichia coli strains were typed for the presence of lipopolysaccharide (LPS) O-antigen (O-Ag) via silver staining (methods described in Doorduijn et al., 2021).
Orange strains are MAC-sensitive (MAC-sens), blue strains MAC-resistant (MAC-res), and black strains complement-resistant (comp-res). A dual-color marker was loaded as control (on the left). The silver stain was previously shown in part (only for silver MAC-resistant and MAC-sensitive E. coli strains) in Doorduijn et al., 2021.
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig3-figsupp1-data1-v2.jpg
-
Figure 3—figure supplement 1—source data 2
Klebsiella strains were typed for the presence of lipopolysaccharide (LPS) O-antigen (O-Ag) via silver staining (methods described in Doorduijn et al., 2021).
Orange strains are MAC-sensitive (MAC-sens), blue strains MAC-resistant (MAC-res), and black strains complement-resistant (comp-res). A dual-color marker was loaded as control (on the left).
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig3-figsupp1-data2-v2.jpg
Figure 4
Binding of C8 and C9 triggers release of membrane attack complex (MAC) precursors from MAC-resistant Escherichia coli.
(a) Schematic overview of how E. coli (orange rods) MG1655 (MAC-sensitive) and 552059.1 (MAC-resistant) were labelled with convertases (green ovals) in 10% C5-depleted serum. Next, bacteria were washed and bacteria (5×108 bacteria/ml for b, 1×108 bacteria/ml for c and d) were incubated with alternative pathway (AP) convertase components (5 µg/ml FB and 0.5 µg/ml FD) and 100 nM C5 and C6 (1); 100 nM C5, C6, and C7 (2); 100 nM C5, C6, C7, and C8 (3) or 100 nM C5, C6, C7, C8, and 1000 nM C9 (4). The supernatant was collected after 60 min by centrifugation. (b) Western blot for C5 of the supernatant. The Western blot is a representative of at least three independent experiments. The upper band represents the α-chain of C5, the middle band of C5b (α’), and the lower band the β-chain of both C5 and C5b. C5b6 in the supernatant of MAC-sensitive MG1655 (c) and MAC-resistant 552059.1 (d) was quantified by enzyme-linked immunosorbent assay (ELISA). Dotted line represents the background OD450. ELISA data represent individual values with mean ± SD of three independent experiments. Statistical analysis was done using an ordinary one-way ANOVA with Tukey’s multiple comparisons test (c and d) and relevant p-values are indicated in the figure.
-
Figure 4—source data 1
Escherichia coli MG1655 and 552059.1 were labelled with convertases in 10% C5-depleted serum.
Next, bacteria were washed and 5×108 bacteria/ml were incubated with alternative pathway (AP) convertase components (5 µg/ml FB and 0.5 µg/ml FD) and 100 nM C5 and C6 (2); 100 nM C5, C6, and C7 (3); 100 nM C5, C6, C7, and C8 (4) or 100 nM C5, C6, C7, C8, and 1000 nM C9 (5). The supernatant was collected after 60 min by centrifugation and analyzed by Western blotting for C5. Ten nM C5 or pC5b6 were loaded as positive controls for C5 and C5b.
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig4-data1-v2.jpg
Figure 5 with 2 supplements
Release of membrane attack complex (MAC) precursors from Escherichia coli triggers lysis of bystander human erythrocytes, but is prevented by serum regulators vitronectin (Vn) and clusterin (Clu).
(a) Schematic overview of the bystander lysis assay. E. coli strains (orange rods) were labelled with convertases (green ovals) in 10% C5-depleted serum and washed. Next, convertase-labelled bacteria (3.3×108 per ml) were incubated with: human erythrocytes (1×108 per ml), alternative pathway (AP) convertase components (5 nM FB and 20 nM FD) and MAC proteins (100 nM C5, 100 nM C6, 100 nM C7, 100 nM C8, and 500 nM C9). The supernatant was collected after 60 min by centrifugation and analyzed for the presence of hemoglobulin. The percentage of lysed erythrocytes was calculated by setting a buffer-only control at 0% lysis and MilliQ control at 100% lysis. (b) Bystander erythrocyte lysis for MAC-sensitive (MAC-sens) and MAC-resistant (MAC-res) E. coli strains. (c) Bystander erythrocyte lysis for convertase-labelled MAC-resistant E. coli 552059.1 incubated with 10% pooled human serum, MAC proteins (30 nM C5, 30 nM C6, 30 nM C7, 30 nM C8, and 300 nM C9) or MAC components with 133 nM Vn, 133 nM Clu, or 20 µg/ml C5 conversion inhibitor OmCI. Data represent individual values with mean ± SD of three independent experiments. Statistical analysis was done using an ordinary one-way ANOVA with Tukey’s multiple comparisons test (c) and relevant p-values are indicated in the figure (all conditions compared with MAC proteins only).
Figure 5—figure supplement 1
Soluble membrane attack complex (sMAC) release with purified MAC proteins from convertase-labelled Escherichia coli strains.
MAC-sensitive (MAC-sens, in orange) and MAC-resistant (MAC-res, in blue) E. coli strains were labelled with convertases in 10% C5-depleted serum and washed. Next, convertase-labelled bacteria (3.3×108 per ml) were incubated with alternative pathway (AP) convertase components (50 and 20 nM FD) and MAC components (100 nM C5, 100 nM C6, 100 nM C7, 100 nM C8, and 500 nM C9). The supernatant was collected by centrifugation after 60 min. sMAC was detected in the reaction supernatant (100-fold diluted) by enzyme-linked immunosorbent assay (ELISA). MAC components without bacteria (MAC components only) were taken as background control. Data represent individual values with mean ± SD of three independent experiments.
Figure 5—figure supplement 2
Effect of vitronectin (Vn) and clusterin (Clu) on C9 polymerization and target-specific membrane attack complex (MAC) assembly.
(a) MAC-resistant Escherichia coli 552059.1 was labelled with convertases in 10% C5-depleted serum and washed. Next, convertase-labelled bacteria (3.3×108 per ml) were incubated with: alternative pathway (AP) convertase components (50 nM FB and 20 nM FD), MAC components C5-C9 (30 nM C5, 30 nM C6, 30 nM C7, 30 nM C8, and 300 nM Cy5-labelled C9) with or without 133 nM Vn or 133 nM Clu. SDS-PAGE was performed to distinguish monomeric-C9 from polymeric-C9 by in-gel Cy5 fluorescence. (b, c) MAC-sensitive E. coli MG1655 were labelled with convertases in 10% C5-depleted serum and washed. Next, convertase-labelled bacteria (5×107 per ml) were incubated with: inner membrane (IM) damage marker 2.5 µM Sytox blue, a concentration range of MAC components (C5:C6:C7:C8:C9 as 1:1:1:1:3) alone and with 100 nM Vn or Clu. After 30 min, bacteria were analyzed by flow cytometry for binding of Cy5-labelled C9 (b) and Sytox fluorescence as read-out for IM damage (c). Flow cytometry data are represented by geoMFI values of the bacterial population, in the case of Sytox the relative geoMFI was calculated compared to bacteria in buffer and Sytox only. The SDS-PAGE image are representative for at least three independent experiments. Data represent mean ± SD of three independent experiments.
-
Figure 5—figure supplement 2—source data 1
Membrane attack complex (MAC)-resistant Escherichia coli 552059.1 was labelled with convertases in 10% C5-depleted serum and washed.
Next, convertase-labelled bacteria (3.3×108 per ml) were incubated with: alternative pathway (AP) convertase components (50 nM FB and 20 nM FD), MAC components C5-C9 (30 nM C5, 30 nM C6, 30 nM C7, 30 nM C8, and 300 nM Cy5-labelled C9, named MAC comp) with or without 133 nM vitronectin (Vn) or 133 nM clusterin (Clu). SDS-PAGE was performed to distinguish monomeric-C9 from polymeric-C9 by in-gel Cy5 fluorescence.
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig5-figsupp2-data1-v2.jpg
Figure 6 with 1 supplement
Isolation of soluble membrane attack complex (sMAC) using His-tagged C6 and analysis by blue-native PAGE (BN-PAGE) and liquid chromatography-tandem mass spectrometry (LC-MS/MS).
sMAC was generated by incubating MAC-resistant (MAC-res) Escherichia coli 552059.1 in C6-depleted serum supplemented with His-tagged C6 (His-C6). sMAC in the supernatant was captured with HisTrap beads and eluted. (a) Concentrated eluate was separated by size exclusion chromatography (SEC) on a Superose 6 column and OD280 was measured to determine protein content (OD280). Eluate from serum without bacteria (nonactivated serum) and 50 µg commercially available sMAC (Complement Technology) were analyzed as controls. (b) BN-PAGE was performed with pooled (B5+B6 and B7+B8) or individual (B9-B12) SEC fractions and analyzed by Western blotting for C6 (above) and C9 (below). Black fractions represent SEC fractions commercially available sMAC, blue fractions represent SEC fractions of serum incubated with MAC-res E. coli. Two µg of purified sMAC and His-C6 were loaded as control. (c) The protein abundance of all sMAC components (iBAQ value) was determined by LC-MS/MS for individual SEC fractions (B5–B12). (d) The ratio of all individual sMAC components to C5 was determined in a pooled sample of fraction B5-B12. The mean ratio of individual samples was indicated for relevant components above the dot plots. (e) In addition, the ratio of C8 and C9 to C5 was determined for each individual fraction separately for serum incubated with MAC-res E. coli and commercially available sMAC (f). The dotted line (d, e, and f) represents a ratio of 1. The SEC profile and Western blot are representative of three independent experiments. LC-MS/MS data represent three individual digests of the same fraction with mean ± SD that are representative for two independent experiments.
-
Figure 6—source data 1
Soluble membrane attack complex (sMAC) was generated by incubating MAC-resistant (MAC-res) Escherichia coli 552059.1 in C6-depleted serum supplemented with His-tagged C6 (His-C6).
sMAC in the supernatant was captured with HisTrap beads and eluted. Concentrated eluate was fractionated by size exclusion chromatography (SEC) on a Superose 6 column. Fifty µg commercially available sMAC (Complement Technology) was also analyzed as control. Blue-native PAGE (BN-PAGE) was performed with pooled (B5+B6 and B7+B8) or individual (B9-B12) SEC fractions and analyzed by Western blotting for C6 (left) and C9 (right). Black fractions represent SEC fractions commercially available sMAC, blue fractions represent SEC fractions of serum incubated with MAC-res E. coli. Two µg of purified sMAC and His-C6 or His-C9 were loaded as control.
- https://cdn.elifesciences.org/articles/77503/elife-77503-fig6-data1-v2.jpg
Figure 6—figure supplement 1
Validation isolation soluble membrane attack complex (sMAC) with HisTrap beads and liquid chromatography-tandem mass spectrometry (LC-MS/MS) size exclusion chromatography (SEC) fractions.
sMAC was detected by enzyme-linked immunosorbent assay (ELISA) at different steps during the isolation procedure using HisTrap beads. His-tagged sMAC was generated by incubating Escherichia coli strain 552059.1 (5×108 per ml) in 10% C6-depleted serum with 50 nM His-tagged C6 (His-C6) for 60 min at 37°C. Supernatant was collected by centrifugation (serum supernatant) and His-tagged sMAC was captured with HisTrap beads for 90 min at 4°C. Supernatant was collected from the beads (supernatant HisTrap beads) and sMAC was eluted from beads with 250 mM imidazole (eluate HisTrap beads) for 30 min at 4°C. (a) sMAC was detected by ELISA for a dilution range of samples. Data represent mean ± SD of four independent experiments. (b–e) LC-MS/MS was performed on individual SEC fractions of concentrated bead eluate for serum incubated with MAC-res E. coli and 50 μg commercially available sMAC. The protein abundance of individual sMAC components (iBAQ value) and ratio of individual sMAC components (compared to C5) were represented for serum incubated with E. coli (b, d) and commercial sMAC (c, e). The dotted line (d and e) represents a ratio of 1. LC-MS/MS data represent three individual digests of the same fraction with mean ± SD that are representative for two independent experiments.
Figure 7
Soluble membrane attack complex (sMAC) that is released from bacteria is a heterogeneous protein complex with different stoichiometries.
Mass spectrometry (MS) profiling of sMAC components in serum supernatant incubated for 3 hr at 37°C with (bottom, activated) and without (top, nonactivated) Staphylococcus aureus Wood46. Serum was separated by size exclusion chromatography (SEC) and the protein abundance (normalized iBAQ values) in each fraction was determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS). The gray boxes indicate the elution of sMAC components in nonactivated and activated serum. The arrows on the top indicates elution of molecular weight (kDa) markers.
Tables
Table 1
Bacterial strains used in this study.
| Strain | Origin | Amount of detectable LPS O-Ag |
|---|---|---|
| Escherichia coli MG1655 | Laboratory strain | Absent (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli BW25113 | Laboratory strain | Absent (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli MC1061 | Laboratory strain | Absent (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli 547563.1 | Clinical isolate* | Low† (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli 552059.1 | Clinical isolate* | High (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli 552060.1 | Clinical isolate* | High (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli 567705.1 | Clinical isolate* | High (Doorduijn et al., 2021) (also in Figure 3—figure supplement 1) |
| Escherichia coli 566989.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Escherichia coli 552912.1 | Clinical isolate* | No (Figure 3—figure supplement 1) |
| Escherichia coli 552866.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Escherichia coli 547654.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Escherichia coli 547655.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Klebsiella variicola 402 | Clinical isolate* | Low† (Figure 3—figure supplement 1) |
| Klebsiella pneumoniae 567880.1 | Clinical isolate* | Low† (Figure 3—figure supplement 1) |
| Klebsiella pneumoniae 567702.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Klebsiella pneumoniae 567709.1 | Clinical isolate* | High (Figure 3—figure supplement 1) |
| Escherichia coli CGSC7740 | Laboratory strain | Absent (Doorduijn et al., 2021) |
| Escherichia coli CGSC7740 wbbL+ | Laboratory strain | High (Doorduijn et al., 2021) |
| Staphylococcus aureus SH1000 | Laboratory strain | n.a. |
| Staphylococcus aureus Wood46 | Laboratory strain | n.a. |
| Staphylococcus aureus Newman | Laboratory strain | n.a. |
| Staphylococcus epidermidis KV103 | Clinical isolate* | n.a. |
| Streptococcus agalactiae COH-1 | Clinical isolate* | n.a. |
-
*
All clinical isolates were obtained from the clinical Medical Microbiology department at the University Medical Center Utrecht.
-
†
Detection of O-Ag was limited, but not absent. n.a. means O-Ag expression is not applicable, because these are Gram-positive bacteria that inherently do not express LPS.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Soluble MAC is primarily released from MAC-resistant bacteria that potently convert complement component C5
eLife 11:e77503.
https://doi.org/10.7554/eLife.77503
