Molecular mimicry of host sialylated glycans allows a bacterial pathogen to engage neutrophil Siglec-9 and dampen the innate immune response

Abstract

Human neutrophil Siglec-9 is a lectin that recognizes sialic acids (Sias) via an amino-terminal V-set Ig domain and possesses tyrosine-based inhibitory motifs in its cytoplasmic tail. We hypothesized that Siglec-9 recognizes host Sias as "self," including in cis interactions with Sias on the neutrophil's own surface, thereby dampening unwanted neutrophil reactivity. Here we show that neutrophils presented with immobilized multimerized Siaalpha2-3Galbeta1-4GlcNAc units engage them in trans via Siglec-9. The sialylated capsular polysaccharide of group B Streptococcus (GBS) also presents terminal Siaalpha2-3Galbeta1-4GlcNAc units, and similarly engages neutrophil Siglec-9, dampening neutrophil responses in a Sia- and Siglec-9-dependent manner. Reduction in the neutrophil oxidative burst, diminished formation of neutrophil extracellular DNA traps, and increased bacterial survival are also facilitated by GBS sialylated capsular polysaccharide interactions with Siglec-9. Thus, GBS can impair neutrophil defense functions by coopting a host inhibitory receptor via sialoglycan molecular mimicry, a novel mechanism of bacterial immune evasion.

Figure 1

Human neutrophil cell surface Siglec-9 interacts with sialic acids presented in trans. (A) Neutrophils bind to immobilized α2-3–linked Sias. Wells were coated with polyacrylamide arrays carrying multiple copies of Siaα2-3Galβ1-4GlcNAc or Galβ1-4GlcNAc (100 μL of 1 mg/mL stock diluted 1:100; GlycoTech, Gaithersburg, MD). Fluorescently labeled neutrophils were added, allowed to adhere, and unbound cells then washed away. The percentage of adherent neutrophils was determined by dividing the final FI by initial FI and multiplying by 100. (B) Validation of IgG anti–Siglec-9 Sia binding site (BSAb) and Sia nonbinding site (NBSAb) antibodies as functional reagents; 96-well plates were coated with 100 μL protein A in carbonate buffer (2 μg/mL) overnight at 4°C. Recombinant soluble Siglec-9-Fc chimera was then added in 100 μL enzyme-linked immunosorbent assay (ELISA) buffer (2 μg/mL) overnight at 4°C. The wells were washed, pretreated with BSAb, NBSAb, or PBS for 30 minutes at room temperature, and then washed 3 times with ELISA buffer. Wells were then incubated with biotinylated Siaα2-3Galβ1-4GlcNAc polyacrylamide array probes for 1 hour at room temperature, washed 3 times, incubated with streptavidin-alkaline phosphatase diluted 1:1000 in ELISA buffer, and washed again, and wells were developed with 100 μL substrate/well. (C) Neutrophil binding to immobilized α2-3–linked sialic acids requires Siglec-9. Neutrophils were incubated with BSAb, NBSAb, or PBS before carrying out the binding assay, as described in panel A. (D) Neutrophils bind to the α2-3–linked Sias of the GBS capsular polysaccharide via Siglec-9. Cell wall extracts were isolated from GBS strain COH1 using mutanolysin (Sigma-Aldrich, St Louis, MO) digestion as described. Further purification of CPS from cell wall preparations was performed using ion exchange and size exclusion chromatography as described. The purified CPS was deposited on microtiter wells as in panel A. Neutrophils pretreated with the BSAb or NBSAb were added and allowed to adhere before washing away nonadherent cells. (E) Neutrophils bind to GBS cell surface extracts via Siglec-9. The unpurified type III GBS (COH1) cell surface extract, which includes the sialylated CPS, was immobilized in ELISA wells and studied for binding of neutrophils pretreated with the BSAb or NBSAb as in panel A. All results are representative of at least 3 experiments performed in triplicate. Neutrophil viability was approximately 85% at the end of the 30-minute time period of the assays, and the antibodies had no major effects on viability or activation (data not shown). Error bars represent SEM.

Figure 2

GBS engagement of Siglec-9 attenuates human neutrophil immune functions. (A) GBS attenuates neutrophil oxidative burst via engagement of Siglec-9. BSAb- or NBSAb-treated neutrophils were labeled with dichlorofluorescin-diacetate at a final concentration of 20 μM and incubated for 20 minutes at 37°C. GBS COH1 WT was added at a multiplicity of infection (MOI) of 10 and incubated for 30 minutes at 37°C. Oxidative burst was measured using FACS analysis. MFI indicates mean fluorescence intensity. (B) GBS reduces neutrophil release of granule proteases by engagement of Siglec-9. Neutrophils were pretreated with BSAb or NBSAb (10⁶ cells in 50 μL mixed with GBS COH1 bacteria at an MOI of 10 in triplicates) and incubated at 37°C for 30 minutes. Tubes were centrifuged at 1000g for 5 minutes and the supernatant collected into wells of a 96-well plate; 0.5 μL of MeOSuc-Ala-Ala-Pro-ValNmec dissolved in dimethyl sulfoxide at 20 mM was added to each well. After incubating at room temperature for 20 minutes, hydrolysis of the substrate was monitored spectrofluorometrically by change in absorbance at 405 nm. Baseline release of elastase by unstimulated neutrophils is set to 100%, and the graph shows a relative percentage release. (C,D) Production of NETs is attenuated GBS CPS engagement of Siglec-9. Antibody-treated neutrophils were incubated with bacteria at MOI 100 for 30 minutes at 37°C. NETs were stained using Cytox Orange and counted across the central transect of each well under the fluorescent microscope. The y-axis shows the number of NETs counted per transect of the wells. (E) Siglec-9 engagement by GBS blunts the up-regulation of mRNA expression of the inhibitory cytokine IL-10. GBS COH1 WT was added to antibody-pretreated neutrophils at an MOI 10 in a 6-well plate for 30 minutes at 37°C, and expression of IL-10 mRNA quantified with reverse-transcribed polymerase chain reaction primers forward (AACCTCCTCTCTGCCATCAA) and reverse (GGAAGACCCCTCCCAGATAG). (F) GBS engagement of Siglec-9 via its capsular Sia promotes resistance to neutrophil total killing. The y-axis shows percentage survival, ie, recovered colony-forming units (cfu) divided by inoculum cfu multiplied by 100. (G) GBS engagement of Siglec-9 via CPS Sia promotes resistance to neutrophil extracellular killing. Total and extracellular neutrophil killing assays were as described. Cytochalasin D was added to antibody-treated neutrophils and mixed with GBS COH1 WT, GBS COH1 deltaNeuA (Sia minus), or group A Streptococcus M1 strain at MOI 10, and incubated at 37°C. Part of each sample was plated at 5, 10, 20, and 30 minutes, and the surviving bacteria cfu counted the next day. The graph shows surviving cfu percentage at 30 minutes. Similar differences were seen at other time points.