[Driver Solution X100c
Laurice Whack
Streptococcus pneumoniae, the main causative agent of pneumonia, triggers inflammation and tissue damage by producing a pore-forming toxin, pneumolysin (Ply). Ply-induced inflammation drives pneumococcal transmission from nasopharynx (its primary reservoir), but also contributes to host mortality, limiting its occupiable habitats. Here, we uncovered the structural basis for loss of pore-forming activity of a Ply variant, present in Serotype 1 ST306, and observed that this enabled adoption of an intracellular lifestyle, attenuating inflammatory responses and prolonging host tolerance of pneumococcus in the lower airways. This commensal-like lifestyle, resembling that of members of the mitis group of Streptococci, might have evolved within ST306 by loss of function ply mutations, compensating for limited nasopharyngeal carriage capacity by facilitating adaptation to an alternate niche.
Copyright: 2020 Badgujar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All files regarding Ply-NH crystal structure are available in PDB database (accession number 6JMP). All other relevant data are within the manuscript and its Supporting Information files.
Streptococcus pneumoniae (the pneumococcus or SPN) is a Gram positive, alpha-hemolytic bacterium and is the leading cause of community-acquired pneumonia, pediatric empyema and bacterial meningitis. SPN has a characteristic asymptomatic colonization phase in the human nasopharynx, its predominant niche and the primary reservoir for onward transmission, but can act as an opportunistic pathogen within other host sites [3]. Nasopharyngeal colonization is a prerequisite for the development of pneumococcal disease, but the relative invasiveness of SPN varies between serotypes [4], 100 different types of which have been classified, based on the composition of capsular polysaccharide [5,6]. Although pneumococcal virulence factors induce cytotoxicity, the host inflammatory response triggered against these factors is the major mediator of pathology and lethality associated with invasive pneumococcal disease (IPD). The key trigger of this dysregulated host inflammation is pneumolysin (Ply), a pore forming toxin belonging to the family of cholesterol dependent cytolysins (CDC) (S1 Fig), produced by all SPN strains. In addition to the extensive cellular damage resulting from pore forming activity on host cell membranes, Ply modulates host inflammatory and immune responses. At low levels, Ply can stimulate tolerogenic host responses via interaction with mannose receptor C type 1 (MRC-1) [7], but at the high concentrations achieved during IPD, this results in excessive inflammation, driven via its interactions with Toll-like receptors (TLR4) [8] and activation of the NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome [9]. Ply also induces necroptosis of respiratory epithelial cells and this mode of cell death has been linked to release of IL-1α from host cells [10,11]. This IL-1 signaling has been identified to play a key role in inflammatory clearance of SPN from the nasopharynx [12] with Ply-deficient strain exhibiting slower clearance than their wild type counterparts [13]. On the other hand, inflammation triggered by the pore-forming activity of Ply in the upper respiratory tract was found to be vital for pneumococcal shedding and transmission [14]. Thus, Ply is a versatile factor, the activities of which can have different (and sometimes opposing) consequences depending on the concentration or niche in which it is produced, variously aiding or hindering pneumococcal fitness within the host. The critical contribution of Ply to stimulation of the inflammation required to achieve transmission provides an explanation for why SPN lineages have not lost this toxin over the course of evolution, despite its potential to kill the host, leading to loss of reservoir.
To determine the molecular basis for the inability of Ply-NH to form pores, each step of pore formation, namely, (a) binding of monomers to membrane cholesterol, (b) oligomerization to form pre-pore and (c) pre-pore to pore transition (TMH formation) was examined. Comparison of the cholesterol binding ability of Ply-NH with that of Ply-H was performed by pre-incubating Ply variants with cholesterol, followed by analyzing its interaction with RBC ghost membrane. Western blot analysis showed that Ply-NH interacts with the membrane in a similar manner to Ply-H and the interaction of both proteins with membranes was reduced upon cholesterol pre-treatment, suggesting that the mutations in Ply-NH do not alter its ability to bind cholesterol-containing membranes (Fig 1C). Transmission electron microscopy, used to visualize Ply assembly on membranes, demonstrated that Ply-H predominantly formed oligomeric rings (Fig 1D). Interestingly, Ply-NH also formed rings of similar size to Ply-H, which was further confirmed by SDS-PAGE, following treatment of eukaryotic membranes with the Ply variants (Fig 1E).
To decipher the molecular reasons behind loss of pore forming ability in Ply-NH, the crystal structure was solved (2.2 resolution) and was found to consist of 4 distinct domains, D1-D4 (Figs 2A and S4 and S1 Table). This is the first reported structure of a non-pore forming CDC. D1 is present at the N-terminal region and comprises of 6 anti-parallel β strands, loops and 5 α helices. D2 consists of five-stranded anti-parallel β sheets, which form the backbone of the structure and connect D4 with D1. D3 consists of a single antiparallel β-sheet with two α helices on either side. D4, which is connected to D2 via the Arg-Asn-Gly flexible linker, is comprised of two anti-parallel β sheets, with the conserved undecapeptide at the end of the loop, which is required for binding to cholesterol.
The overall structural fold of Ply-NH was found to be similar to that of other reported CDCs, such as anthrolysin (3CQF) [23], intermedilysin (4BIK) [24], perfringolysin (PFO; 1PFO) [25] and pneumolysin (Ply-H; 5CR6 and 4QQA) [20,26]. Superposition of Ply-NH structure with the recently reported structures of Ply-H (5CR6 and 4QQA) [20,26] produced root mean square deviation (r.m.s.d.) of 2.4 and 1.2 , respectively, over 471 Cα atoms. Superposition of specific domains, D1-3 of Ply-NH and Ply-H (5CR6) yielded r.m.s.d. of 0.75 and alignment of only D4 yielded r.m.s.d. of 0.23 . Comparison with one Ply-H structure, 5CR6, showed higher r.m.s.d. compared to the other (4QQA) because of a relative 10 movement of D4 with respect to the rest of the molecule.
Monomers in the Ply-NH crystal were found to be tightly packed and this crystallographic arrangement resembles the monomer-monomer interaction interface that may form during formation of the pre-pore complex. Although both sides of the monomer show charge complementarities, the overall surface of the structure is highly electronegative (Fig 2B). However, D3 did not show any charge complementarity. In the well-studied and closely related CDC perfringolysin of Clostridium perfringens, the α helices present in D3 are reported to undergo a conformational change to form TMH1 and TMH2, thereby inserting into the membrane [27]. Since most of the substitutions and deletions are present in the D3 domain of Ply-NH, we hypothesized that its non-hemolytic nature might be a consequence of the conformational changes associated with these mutations.
Another important substitution in Ply-NH is threonine to isoleucine at position 172. The I172 in Ply-NH is well defined in the electron density (S4 Fig) and is found to be located at the tip of TMH1 (Fig 2F). Presence of the polar side chain of T172 in the non-polar pocket of Ply-H likely makes this region quite unstable, allowing the smooth disengagement from β3 and β4 upon binding of this toxin to the membrane. On the contrary, I172 in Ply-NH is found to be stabilized through the hydrophobic interactions involving the side chains of F169, L176, Y247, V288 and L290 (Fig 2F). The interactions from Y247, V288 and L290 might be essential to prevent the disengagement of these α helices joining β3 and β4, to form TMHs for membrane insertion and subsequent pore formation.
We next mutated H150 and I172 back to their respective Ply-H residues (H150Y and I172T) in the Ply-NH background. Though individual mutations showed some gain of activity, the double mutant (Ply-NHH150Y+I172T) regained most of the hemolytic activity (Fig 2G). This clearly implicates Y150H and T172I as the major mutations responsible for the loss of pore forming ability of Ply-NH, by preventing disengagement of α helices joining β3 and β4 and subsequent TMH formation.
In order to track the fate of SPN strains harboring Ply variants, following entry into host cells, we performed penicillin-gentamicin protection assays using R6:Ply-H, R6:Ply-NH and R6:Ply-DM (a recombinant R6 strain where Ply-H is replaced with the Ply-NHH150Y+I172T allele which contains reversion mutations in the form of H150Y and I172T) in A549 cells. Our findings demonstrate significantly improved survival of R6:Ply-NH compared to R6:Ply-H at all time points, whereas R6:Ply-DM behaved similarly to R6:Ply-H, indicating that loss of pore forming ability is beneficial for prolonged intracellular persistence of SPN (Fig 5A). Ability of R6:Ply-NH to persist longer than R6:Ply-H was also observed inside THP-1 macrophages (Fig 5B).
Since Ply is a key inducer of host inflammation, tissue injury, morbidity and mortality, we analyzed the inflammatory response in the respiratory tract of mice infected with different SPN strains. Consistent with the findings of others [36], mice infected with D39:Ply-NH and ST306 had significantly reduced lung inflammation, evident by lower expression of pro-inflammatory cytokines KC and interleukin-6 (S10C and S10D Fig) and reduced infiltration of leukocytes (S10E Fig), principally neutrophils (PMN) (S10F Fig), compared to D39:Ply-H infected animals. Histo-pathological analysis of lung sections from infected mice revealed similar trends (Fig 6C).
795a8134c1