Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 4 (6), 555-66

Apolipoprotein B Is an Innate Barrier Against Invasive Staphylococcus Aureus Infection

Affiliations

Apolipoprotein B Is an Innate Barrier Against Invasive Staphylococcus Aureus Infection

M Michal Peterson et al. Cell Host Microbe.

Abstract

Staphylococcus aureus is both a colonizer of humans and a cause of severe invasive infections. Although the genetic basis for phenotype switching from colonizing to invasive has received significant study, knowledge of host factors that antagonize the switch is limited. We show that VLDL and LDL lipoproteins interfere with this switch by antagonizing the S. aureus agr quorum-sensing system that upregulates genes required for invasive infection. The mechanism of antagonism entails binding of the major structural protein of these lipoproteins, apolipoprotein B, to an S. aureus autoinducing pheromone, preventing attachment of this pheromone to the bacteria and subsequent signaling through its receptor, AgrC. Mice deficient in plasma apolipoprotein B, either genetically or pharmacologically, are more susceptible to invasive agr+ bacterial infection, but not to infection with an agr deletion mutant. Therefore, apolipoprotein B at homeostatic levels in blood is an essential innate defense effector against invasive S. aureus infection.

Figures

Figure 1
Figure 1. Serum lipoproteins antagonize agr signaling
(A) agr:P3 promoter activation by 100 nM AIP 1 is inhibited in S. aureus (ALC 1743 with agr:P3-gfp, 2 × 107/ml) during 3 hrs of culture with either broth or the indicated dilutions of pooled human serum (PHS) as compared to lipoprotein deficient sera (LPDS). Magnitude of promoter activation was measured as fluorescence induction (mean channel of fluorescence, MCF). Inset: Log CFU indicated that bacterial growth was equivalent under both conditions. Data are represented as the mean ± SEM, n=3 performed in duplicate. (B) Effect of VLDL, LDL, or HDL at equivalent cholesterol concentration (0.4 μM) on 100 nM AIP1-induced agr:P3 activation during 3 hr of culture of ALC1743. Data are represented as the mean ± SEM, n=3 performed in duplicate. (C) Relative quantification of RNA III transcript to 16s RNA produced by USA300 clinical isolate UAMS 1378 cultured at 2 × 107/ml with 100 nM AIP1 with or without 2 μg/ml VLDL for 1 hr. Data are represented as the mean ± SEM, n=4 performed in triplicate. (D) Dose-dependent inhibition of AIP1-induced agr:P3 activation by purified apolipoprotein B but not other serum lipoproteins. Concentrations tested: apo B: 2–8μg/ml (4–16nM), apo A-I: 4–8μg/ml (143–288nM), apoC-1: 4–8μg/ml (0.61–1.2 μM), and apoE: 4–8μg/ml (117–234 nM). (E) Goat IgG anti-apoB (5 μg/ml) vs control goat IgG reversed VLDL (5 μg/ml) and purified apo B (5 μg/ml) inhibition of AIP1-induced agr:P3 activation in ALC1743. Data are represented as the mean ± SEM, n=3 performed in duplicate. (F) mAB C1.4 IgG1 anti-apo B (5 μg/ml) vs. control mouse IgG1 reversed VLDL (5 μg/ml) inhibition of 100 nM AIP1-induced relative RNA III transcript produced by USA300 clinical isolate LAC cultured at 2 × 107/ml for 1 hr. Data are represented as the mean ± SEM, n=3 performed in duplicate.
Figure 2
Figure 2. Apolipoprotein B and its associated lipid particles inhibits AIP binding to S. aureus and interacts with AIP by Surface Plasmon Resonance
(A) S. aureus (RN 6390, 2 × 107/ml) was incubated with 1 μM FITC-AIP1 alone (vehicle control) or in combination with potential inhibitors including 1 μM native AIP1, VLDL (2 μg/ml), apo B (8 μg/ml) or apo A-1 (8 μg/ml) at 37°c for 3 hr. Specific binding was measured by flow cytometry. Data are represented as the mean ± SEM, n=4. (B) 1 μM FITC-IgG binding to S. aureus RN6390 was not inhibited by VLDL or apo B. Data are represented as the mean ± SEM, n=3. (C) S. aureus (USA300 LAC, 2 × 107/ml) was incubated with 1 μM FITC-AIP1 in vehicle control or in combination with VLDL (2 μg/ml) or apo B (8 μg/ml) at 37°c for 3 hr. Specific binding was measured by flow cytometry. Data are represented as the mean ± SEM, n=2 performed in duplicate. (D) Biacore X 100 analysis in resonance units (RU) of the interaction of 10 nM apo B, apo C1, or apo A1 with either cyclic biotin-AIP1 or linear biotin-AIP1 immobilized on streptavidin chips after 60 s of contact time followed by 60 s of dissociation time at a flow rate of 10 μl/min. Data represented as mean ± SEM, n=3–5.
Figure 3
Figure 3. Interaction of AIP with modified LDL and uptake by macrophage scavenger receptors
(A) LDL modified by acetylation (5 μg/ml) inhibits AIP1-induced agr:P3 promoter activation and the inhibition is reversed by a mAb to an N-terminal epitope of apo B (mAb C1.4) (5 μg/ml) but not by a mAb against a C-terminal epitope (mAb 4C11) (5 μg/ml). Data are represented as the mean ± SEM, n=3 performed in duplicate. (B) Reversibility of AIP1-induced agr:P3 promoter activation inhibited by modified LDL. Reporter bacteria incubated with 50 nM AIP1 and 5 μg/ml AcLDL for 3 hrs first and then washed and incubated subsequently with either broth or 500 nM AIP1 demonstrated both spontaneous and AIP1-induced agr:P3 promoter activation. Data are represented as the mean ± SEM, n=2 performed in duplicate. (C) RAW264 murine macrophages were incubated first with mixtures of fluorescently labeled and unlabeled AcLDL on ice and then with FITC-AIP1 for 10 or 30 minutes at 37°C. Fixed cells were imaged by confocal microscopy. First row: RAW cells incubated with unlabeled AcLDL (10 μg/ml) and then with FITC-AIP1 (green fluorophore, 1 μM). Confocal images show no bleed through to the red channel. Second row: RAW cells incubated with DiI-AcLDL (red fluorophore, 1 μg/ml) and unlabeled AcLDL (9 μg/ml). Confocal images show no bleed through to the green channel. Third row: RAW cells incubated with DiI-AcLDL (1 μg/ml) and unlabeled AcLDL (9 μg/ml) and then FITC-AIP1 (1 μM). Images show co-localization of LDL and AIP. Fourth row: RAW cells incubated with DiI-AcLDL (1 μg/ml) and unlabeled AcLDL (9 μg/ml) and then FITC-avidin (1 μM). The FITC tag does not co-localize with Ac-LDL. (D) Percent co-localization of FITC-AIP1 with AcLDL determined at both early (mean ± SEM, n=38 cells) and later time points (mean ± SEM, n=30 cells) using Slidebook as described in experimental procedures. As controls, AcLDL was treated with mAb anti-apoB (clone C1.4) (5 μg/ml) before interacting with FITC-AIP1 and the percent co-localization (mean ± SEM, n=50 cells) was significantly reduced (p< 0.001). Also, dialyzing the AcLDL/FITC-AIP1 complex prior to incubation with the macrophages resulted in equivalent co-location (mean ± SEM, n=36 cells).
Figure 4
Figure 4. Effect of agr on S. aureus infection of apolipoprotein deficient Pcsk9−/− mice as compared to wild-type mice
Air pouches generated on the backs of B6 x 129 wild-type or Pcsk9−/− mice (n=8 for each group) were infected with either 3.5 × 107 non-fluorescent early exponential phase S. aureus agr:P3-gfp ALC 1743 (agr+) or 3.2 × 107 agr:P3-gfp ALC1753 (agr−). At 28 hr post infection, the following parameters were determined and represented as the mean ± SEM: (A) Morbidity was scored on a 0–14 point scale. (B) Bacterial burden (Log CFU) of pouch lavage and spleen. (C) Quantification of agr:P3-gfp promoter activation by flow cytometry of bacteria isolated from the pouch lavage. (D) MIP-2, a murine inflammatory cytokine was measured in the pouch lavage. (E) H & E stain of the basal section of pouch tissue demonstrating bacteria at the surface of the lumen, the epidermis and dermis with infiltrating leukocytes, and the skeletal muscle. (F) Representative pouch tissue from B6 x 129 mice (left panel) vs Pcsk9−/− mice (right panel) infected with early exponential phase S. aureus (agr:P3-gfp ALC1743). Pouch tissue was fixed, sectioned, and sagittal sections were stained with anti-S. aureus antibody (red fluorophore). Quorum-sensing S. aureus were visible by production of GFP in the pouch. Graph represents the relative density of both total and GFP+ bacteria in the epidermis from at least 19 representative areas. (G) Representative pouch tissue from B6 x 129 mice (left panel) vs Pcsk9−/− mice (right panel) infected with mid-exponential phase S. aureus (agr:P3-gfp ALC1743) and stained with anti-S. aureus (red fluorophore) demonstrated significantly more quorum sensing, biofilm formation, and invasion into the dermis at 28 hours post infection in Pcsk9−/− mice. Graph represents the relative density of both total and GFP+ bacteria in the epidermis from at least 19 representative areas. * p < 0.02; ** p < 0.005; L = lumen, D = dermis, E = epidermis; MS = morbidity score; Arrows indicate point of greatest penetration by quorum sensing bacteria; bar indicates 50 μm.
Figure 5
Figure 5. Effect of agr on USA300 (LAC) S. aureus infection of NADPH oxidase knockout (gp91phox−/−) mice treated with 4APP to inhibit liver secretion of lipoproteins
Mice were treated with 100 μl of 5 mg/ml 4APP or vehicle control (0.025M phosphate buffer) (n=4 for each group) i.p. 48 and 24 hours prior to infection of air pouches with either 2.5 × 107 CFU of USA300 LAC or LACΔ agr. At 25 hours hrs post infection, the following parameters were determined and the data represented as the mean ± SEM: (A) Morbidity score 0–14. (B) Weight loss in grams. (C) Bacterial burden (Log CFU) in pouch lavage and spleen. (D) Representative pouch tissue from control and 4APP-treated mice stained with anti-S. aureus antibody (red fluorophore) demonstrating biofilm-like aggregates (arrowhead) and increased numbers of bacteria in the epidermis in 4APP treated mice infected with wild-type LAC. Bar indicates 50 μm. (E) Quantification of bacterial density in the epidermis of pouch tissue from control and 4APP treated mice from at least 22 representative areas. (F) Representative muscle and deep dermis tissue from control and 4APP-treated mice infected with wild-type LAC stained with anti-S. aureus antibody (red fluorophore) demonstrating bacteria surrounding muscle bundles (arrows). Bar indicates 20 μM. (G) Quantification of bacterial density in muscle and deep dermis tissue of pouches from control and 4APP-treated mice from at least 25 representative areas of 530 μM2. * p < 0.05, ** p < 0.01, *** p < 0.001; L = lumen, D = dermis, and E = epidermis.

Comment in

  • Host Interception of Bacterial Communication Signals
    AR Horswill et al. Cell Host Microbe 4 (6), 507-9. PMID 19064249.
    Pathogenic Staphylococcus aureus employs a cell-to-cell communication system to control virulence factor expression in the host during infection. In this issue, Peterson …

Similar articles

See all similar articles

Cited by 59 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback