The human nasopharyngeal epithelial cell line Detroit 562 (CCL-138 was obtained from the American Type Culture Collection (ATCC)).

Clinical isolates of S. pneumoniae were selected from two nationwide surveillance programmes (Mühlemann et al., 2003; Kronenberg et al., 2006), from anonymous donors and have been used in previous studies (Hathaway et al., 2004, 2012).

Strain D39 (serotype 2) pneumococci were used as the parental strain for the production of mutants. Construction of the individual, ΔAmiA, ΔAliA, ΔAliB, and triple mutant ΔAmiA/ΔAliA/ΔAliB was described previously (Alloing et al., 1994; Kerr et al., 2004). Clinical isolates 106.66 (serotype 6B) and 110.58 (non-encapsulated, classic lineage) (Hilty et al., 2014) were also used. Bacteria, stored at -80°C using Protect bacterial preservers (Technical Service Consultants, Heywood, United Kingdom), were grown on Columbia sheep blood agar (CSBA) plates at 37°C, 5% CO₂. Overnight cultures were prepared with 3–10 colonies in 5 mL brain heart infusion broth (Becton Dickinson and Company, le Pont de Claix, France) containing 5% foetal calf serum (FCS; Biochrom KG, Berlin, Germany; BHI+FCS medium).

Strains were streaked onto CSBA plates and incubated at 37°C in a 5% CO₂-enriched atmosphere overnight then sub-cultured in BHI+FCS medium to OD_600nm 0.5, centrifuged at 3000 ×g for 5 min and resuspended in 5 mL chemically defined medium (CDM) which contains 5.5 mM glucose (Schaffner et al., 2014). Growth was monitored in sterile flat-bottomed 96-well microtiter plates (Nunclon Surface, Nunc, Denmark) based on the method of Brewster (2003). In brief, 200 μL bacteria culture was grown per well at 37°C and OD_450nm measured every 30 min using a VersaMax microplate reader (Molecular Devices) over 22 h with or without peptide ligands (synthesised by PolyPeptide Group, Strasbourg, France). The plate was shaken automatically for 5 s before each reading. The problem of condensation affecting the readings was avoided by pre-treating the lids of the 96-well plates with 3 mL 0.05% Triton X-100 in 20% ethanol and allowing them to air-dry before use (Hathaway et al., 2012).

D39 and its mutants ΔAmiA, ΔAliA, ΔAliB, and ΔAmiA/ΔAliA/ΔAliB were streaked onto CSBA plates and incubated at 37°C in a 5% CO₂-enriched atmosphere overnight then subcultured in BHI+FCS medium to OD_600nm 0.5, centrifuged at 3000 ×g for 5 min and resuspended in 5 mL CDM. 250 μL bacterial culture (125 μL of each of the two strains being compared) was transferred to 4.75 mL CDM pre-warmed to 37°C. The initial inoculum was plated out to check that there was a 1:1 mixture of the two bacterial strains. Peptide ligands were added to a final concentration of 0.5 mg/mL, and the culture incubated to an OD_600nm 0.3. Serial dilutions in PBS were plated onto CSBA plates with and without antibiotics to differentiate between wild type (susceptible) and mutants (resistant) strains. ΔAmiA mutant was selected on 10 μg/mL of tetracycline, ΔAliB on 2 μg/mL of chloramphenicol, and both ΔAliA and triple mutant ΔAmiA/ΔAliA/ΔAliB on 1 μg/mL of erythromycin/mL. After overnight incubation, the numbers of colonies were counted and the colony forming units (CFU) for each strain calculated.

Capsule thickness was determined by measuring the zone of exclusion of FITC-dextran as described previously (Gates et al., 2004; Hathaway et al., 2012), using FITC-dextran of 2000 kDa (Sigma-Aldrich) with the following difference, bacteria were cultured overnight in BHI+FCS medium until OD_600nm = 0.5 then 100 μL subcultured into 2 mL of CDM with and without peptide ligands and grown to an OD_600nm = 0.3. After centrifuging at 3000 ×g for 5 min the pellet was resuspended in 500 μL of PBS. 10 μL bacterial suspension was mixed with 2 μL FITC-dextran (10 mg/mL in water), pipetted onto a microscope slide and a coverslip applied firmly. The slides were viewed using a Zeiss Axio Imager M1 fluorescence microscope with a 100× objective and photographed by a Zeiss AxioCam HRc camera. The images were converted to grayscale and analysed with UTHSCSA ImageTool for Windows v3.0 (University of Texas Health Science Center, San Antonio, TX, United States) software. A bright field image was also photographed for each fluorescent image recorded in order to count the number of bacteria. Each image contained between 8 and 150 bacteria, five images were counted per group in each of three independent experiments making a total of 15 images per group. Each data point in Figure 3B represents one image.

Bacteria were cultured overnight in BHI+FCS medium until OD_600nm = 0.5 then 100 μL sub-cultured into 2 mL of CDM with and without peptide ligands and grown to an OD_600nm = 0.3. After centrifuging at 3000 ×g for 5 min the pellet was resuspended in 500 μL of PBS, 25 μL of the suspension was put on microscope slides and Gram stained. The slides were viewed using a Zeiss Axio microscope with a 100× objective and photographed by a Zeiss AxioCam HRc camera. A bright field image was acquired to count the number of bacteria per chain. 5 to 78 bacterial chains were counted per image, five images were counted per group in each of three independent experiments making a total of 15 images per group. Each data point in Figure 4B represents one image.

Adherence to the nasopharyngeal epithelial cell line Detroit 562 (CCL-138, from the ATCC) was performed as described previously (Hathaway et al., 2004) with the following differences: A total of 4 × 10⁵ cells/well were cultured in a 24-well tissue culture plate in Minimum Essential Medium+Earle’s salts (MEM) containing 10% FCS, 2 mM L-glutamine, 1.5 g/L sodium bicarbonate, 0.1 mM non-essential amino acids and 1 mM sodium pyruvate (Gibco, United Kingdom) until reaching complete confluence at 37°C in 5% CO₂. S. pneumoniae strains D39 and its mutants as well as strains 106.66 and 110.58 were grown in BHI+FCS to the logarithmic phase OD_600nm = 0.4 and washed in MEM then resuspended to give 10⁷ bacteria in 0.5 mL MEM, which was added per well of washed Detroit cells. The bacteria were centrifuged onto the cells at 390 ×g for 5 min at room temperature. After 1 h incubation of the cells with the bacteria at 37°C and five washes with PBS, 200 μL of trypsin was then added to each well and incubated for 5 min in 37°C to detach the cells. Adhered bacteria were recovered, serially diluted in 0.85% NaCl and plated onto CSBA. CFU/mL was then calculated after an overnight incubation at 37°C in 5% CO₂.

Biofilm assays were performed as previously described (Marks et al., 2012b) with the following modifications. A total of 4 × 10⁵ Detroit cells/well in MEM salts containing 10% FCS, 2 mM L-glutamine, 1.5 g/L sodium bicarbonate, 0.1 mM non-essential amino acids and 1 mM sodium pyruvate (Gibco, United Kingdom) were cultured in 24-well tissue culture plate until reaching complete confluence at 37°C in 5% CO₂. Strains D39 and its mutants and strains 106.66 and 110.58 were grown in CDM to OD_600nm 0.4 to give 10⁵ bacteria in 0.5 mL per well. 5 μL salmon sperm DNA was added into each well and, where indicated, corresponding peptide ligands, followed by 16 h of static incubation at 37°C, 5% CO₂. The medium containing the planktonic bacteria was removed and each well was washed twice with PBS taking care not to disturb the biofilm. 200 μL PBS was then added per well and the plate was sonicated for 6 s in a water bath and then the biofilm scraped thoroughly from the wells. The contents of the wells were recovered and the CFU determined by plating out serial dilutions CSBA and incubating overnight at 37°C.

To determine transformation rates of strain D39 and its mutants, strains 106.66 and 110.58, bacteria were grown to an OD_600nm = 0.15 in BHI+FCS. A total of 0.5 mL of the culture was transferred to 9.5 mL CDM prewarmed to 30°C and incubated for 15 min at 30°C in a waterbath. Competence-stimulating peptide 1 (CSP 1; ERLSKFFRFILQRLL) for D39 and its mutants and CSP 2 (EMRISRIILDFLFLRKK) for 106.66 and 110.58 were added to final concentration 100 ng/mL with and without AmiA, AliA, and AliB peptide ligands to a final concentration of 0.5 mg/mL, the culture incubated for 15 min at 30°C. Chromosomal DNA, 1 μg, from streptomycin resistant strain 104.37 was added and the culture incubated for 60 min at 30°C, then 90 min at 37°C. Serial dilutions in PBS were plated onto CSBA plates with and without 200 μg/mL streptomycin. After overnight incubation, the numbers of colonies were counted and the transformation rate calculated.

Strain D39 and its mutants were cultured overnight in BHI+FCS medium until OD_600nm = 0.5 then 100 μL sub-cultured into 10 mL of CDM and grown to an OD_600nm = 0.2, the culture was split into two 5 mL aliquots and 0.5 mg/mL of peptides were added to one sample and incubated for 15 min at 37°C. Both cultures were centrifuged for 5 min at 4000 ×g and the pellets were resuspended in 50 mM Tris–HCl and 8 M urea and sonicated. Sample processing, Liquid Chromatography Mass Spectrometry (LC-MS/MS) and data interpretation was essentially done as described previously (Engel et al., 2014) with the following changes. LC-MS/MS analysis was carried out on an UltiMate 3000 Nano LC coupled to an Orbitrap Fusion Lumos instrument (Thermo Fisher Scientific) acquiring full MS scans in the m/z range 400–1400 in the orbitrap at resolution 120,000 with AGC set to 4e5 and maximal ion injection time of 50 ms. Peptide precursors with charge 2–8 were fragmented once in the ion trap then excluded for 30 s. The ion trap settings were data-dependent MS2 cycle time of 3 s, isolation width of 1.6 m/z, fragmentation HCD mode with 30% normalised collision energy, AGC of 1e4 with maximal ion injection time of 35 ms. The LC-MS/MS data was processed with MaxQuant (version 1.5.4.1) using default settings for peak detection, strict trypsin cleavage rule allowing for up to three missed cleavages, variable oxidation on methionine and acetylation of protein N-termini with strict carbamidomethylation of cysteines. Match between runs was activated with a retention time window of 0.7 min. The fragment spectra were interpreted with the S. pneumoniae strain D39 (serotype 2) ensemble database (version gca_d39.ASM1436v1). LFQ intensities were used for data analysis with imputation of missing values from the low end of the LOG2 transformed intensity distribution of each LC-MS/MS run using Perseus (version 1.5.5.3) as suggested by Lazar et al. (2016).

For proteomics, differential protein expression analysis was performed with t-tests within Perseus, including a permutation-based false discovery rate estimation (cutoff at 5% and using S0 function set at 0.5) to correct for multiple testing. For other experiments two-way ANOVA or student t-tests were performed to obtain p-values using the software GraphPad Prism (Version 7, GraphPad Software, Inc.).

Streptococcus pneumoniae strain D39 (serotype 2) and its mutants in which one or all three of the substrate-binding proteins have been deleted (ΔAmiA, ΔAliA, ΔAliB, and ΔAmiA/ΔAliA/ΔAliB) were grown in CDM in 96 well plates in the absence and presence (at concentrations of between 0.5 and 0.062 mg/mL) of AmiA, AliA, and AliB peptide ligands for 22 h.

For strain D39, AmiA, and AliA peptide ligands reduced growth significantly in a dose dependent way (Figures 1A,B) whilst AliB peptide ligand slightly reduced the lag phase and increased the maximum optical density (OD) (Figure 1C). To check that these effects were peptide specific, D39 was grown in the presence of peptides with single amino acid differences from peptides AliA and AliB (Figure 1D). (FSEMQPIVDRQ differs from AliA ligand at the second amino acid and AIQSEKRRKHN differs from AliB ligand at the seventh amino acid, underlined; a single amino acid version of AmiA peptide was insoluble and so could not be used in the growth assay). A single amino acid substitution caused AliA peptide to lose its ability to suppress growth and the substitution in the AliB peptide reduced its ability to boost growth response to the peptide.

Less suppression of growth was seen by AmiA peptide in the ΔAmiA mutant supporting the idea that this is the specific receptor for the peptide (Figure 1E). In contrast, AliA peptide affects the ΔAliA mutant in the same way as it affects the D39 wild type as expected since AliA peptide is known to bind to AmiA receptor (Figure 1F; Nasher et al., 2018b). Unlike the wild type D39 strain, growth of the D39 ΔAliB mutant was reduced by the AliB peptide (Figure 1G). However, none of the three peptide ligands had a great effect on the triple mutant ΔAmiA/ΔAliA/ΔAliB indicating a redundancy of function between the three substrate binding proteins (Figure 1H).

To test whether the peptides had the same effect on growth for clinical isolates, we tested the effects on strain 106.66 (serotype 6B) and non-encapsulated strain 110.58 (Figures 1I,J). As for D39, AmiA and AliA peptide ligands reduced growth whilst AliB peptide ligand slightly shortened the lag phase for these clinical isolates.

AmiA and AliA peptides have a bacteriostatic, rather than bacteriolytic, effect as shown by plating out and counting CFU at different timepoints during exponential growth (Supplementary Figure S1).

Given the effects on growth by the peptides in Figure 1, next we determined whether possession of the oligopeptide binding proteins affected the outcome of a competition assay when wild type strain D39 and each of its mutants were cultured together in the presence of the peptides. Figure 2A shows that D39 is outcompeted by both the mutant lacking AmiA (ΔAmiA) and the triple mutant ΔAmiA/ΔAliA/ΔAliB when the AmiA peptide is present but not when the AmiA peptide is absent (Figure 2B). The same pattern is seen for AliA (Figure 2C) with mutants lacking the receptor for the peptide having a growth advantage in the competition assay when the peptide was present (Figure 2C) but not when the peptide was absent (Figure 2D). In contrast, the wild type D39 strain had a competitive advantage over the mutant lacking AliB (ΔAliB) and over the triple mutant (Figure 2E). D39 also had a growth advantage over the ΔAliB mutant in the absence of peptide, although to a lesser extent, and no advantage over the triple mutant (Figure 2F).

(See Supplementary Figure S2 for all data points at OD_600nm = 0.1, 0.2, and 0.3).

The results of the competition assays support those of the growth curves: peptides AmiA and AliA inhibit growth but AliB enhances growth.

Since the polysaccharide capsule is a major virulence factor of S. pneumoniae, we investigated whether any of the three peptides affected capsule thickness. Figure 3A shows an example of a FITC-dextran exclusion assay using strains D39 (serotype 2) and 106.66 (serotype 6B). For both strains there was a clear diminution of capsule size following incubation with AmiA peptide. The results for all three peptides for both strains are summarised in Figure 3B. The reduction in capsule thickness in response to AmiA peptide was significant for both strains. AliA peptide also reduced capsule thickness in both strains, although the difference was only significant for strain D39. In contrast, AliB peptide ligand caused a slight, non-significant, increase in capsule thickness in strain D39 only (Figure 3B). None of the three peptide ligands affected capsule thickness significantly in the triple mutant ΔAmiA/ΔAliA/ΔAliB (Supplementary Figure S3). In the absence of peptide ligands, mutants had thinner capsule than the parent strain D39. AliA peptide ligand did not affect capsule thickness in the ΔAliA mutant, nor did AliB peptide ligand affect capsule thickness in the ΔAliB mutant. AmiA peptide significantly increased capsule thickness in the ΔAmiA mutant but only to a value equivalent to the capsule thickness of D39 in the absence of peptide (Supplementary Figure S3).

When performing the capsule thickness experiments, we noticed that the peptide ligands appeared to affect chain length of the encapsulated pneumococci. We therefore quantified the effects of the peptide ligands on chain length (Figure 4). All three peptide ligands increased the average number of bacteria per chain, but by far the greatest effect was seen in presence of AliB peptide ligand on strains D39 and 106.66 (Figure 4A). The peptides increased chain length in the encapsulated strains D39 (serotype 2) and 106.66 (serotype 6B) but not the non-encapsulated strain 110.58 (Figure 4B). None of the three peptides had any significant effect on chain length in any of the four mutant strains (Supplementary Figure S4). By replacing the AliB peptide by the one with the single amino acid substitution (AIQSEKRRKHN) we show that the effect on chain length is specific to the AliB peptide (Figure 4C).

We found no effect of the peptide ligands on adherence (see Supplementary Figure S5) but nevertheless we looked for an effect on biofilm formation. Peptide ligands for AmiA and AliA significantly increased biofilm formation for all three strains; D39, 106.66, and 110.58 (Figure 5). AliB peptide ligand caused a significant increase in biofilm in strains D39 and 110.58, although to a lesser extent than the other two peptides, but not in strain 106.66 (Figure 5). AmiA peptide increased biofilm production in mutant ΔAmiA/ΔAliA/ΔAliB but not in mutant ΔAmiA. AliA and AliB peptide ligands caused no changes in biofilm formation in any of the mutants tested (Supplementary Figure S6).

Streptococcus pneumoniae is naturally competent for genetic transformation and being able to take up DNA in a hostile environment may confer a survival advantage. The effect of the peptides on CSP-mediated transformation rate was therefore tested in the strain D39, its mutants and strains 106.66 and 110.58. All three peptide ligands significantly reduced the transformation rate in strains D39, 106.66, and 110.58 (Figure 6). No significant effect was observed in the mutants (Supplementary Figure S7A). It must be noted that the peptide concentration is high (0.5 mg/mL) relative to the concentration of CSP (100 ng/mL) and that at such concentrations the AliA and AliB peptides with single amino acid substitutions and an irrelevant peptide reduced transformation rate (Supplementary Figure S7B), although in contrast to AmiA, AliA, and AliB, the results were not statistically significant.

To gain insight into how the peptides caused these phenotype changes, we also looked at the effect of AmiA, AliA, and AliB peptide ligands on the pneumococcal proteome. Proteomic analysis was performed by LC-MS/MS on strain D39 and its mutants ΔAmiA, ΔAliA, ΔAliB, and triple mutant ΔAmiAΔAliAΔAliB in the presence and absence of peptide ligands. Table 1 shows all proteins significantly upregulated and downregulated in the wild type following a 15 min exposure to each of the peptide ligands. Although there was some overlap of the effect on protein expression between the three peptides, we noted that only AliB peptide increased expression of alanine racemase, a protein that catalyses the racemisation of L-alanine and D-alanine. D-alanine is needed for biosynthesis of peptidoglycan in the cell wall, essential for AliB peptide-induced growth. AmiA peptide appeared to switch off expression of several proteins associated with metabolism such as Vex2, the ATP-binding protein of an ABC transporter for export of lipoproteins during cell division, Cof family protein involved in metabolism and RecN, a protein that repairs DNA during replication. Two proteins uniquely switched off by AliA were a replicative DNA helicase and penicillin-binding protein 2X which is required for building cell wall. These results seem to support the role of AmiA and AliA proteins in reducing growth. Of interest was also the CinA protein, which is associated with competence and was switched on by AmiA and AliA but not AliB.

See Supplementary Table S1 for upregulated and downregulated proteins in the single and triple mutants in the presence of the peptides and Supplementary Table S2 for the complete set of proteins.