Search

Detection and identification of pathogenic bacteria isolated from biological samples (blood, urine, sputum, etc.) are crucial steps in accelerated clinical diagnosis. However, accurate and rapid identification remain difficult to achieve due to the challenge of having to analyse complex and large samples. Current solutions (mass spectrometry, automated biochemical testing, etc.) propose a trade-off between time and accuracy, achieving satisfactory results at the expense of time-consuming processes, which can also be intrusive, destructive and costly. Moreover, those techniques tend to require an overnight subculture on solid agar medium delaying bacteria identification by 12-48 hours, thus preventing rapid prescription of appropriate treatment as it hinders antibiotic susceptibility testing. In this study, lens-free imaging is presented as a possible solution to achieve a quick and accurate wide range, non-destructive, label-free pathogenic bacteria detection and identification in real-time using micro colonies (10-500 μm) kinetic growth pattern combined with a two-stage deep learning architecture. Bacterial colonies growth time-lapses were acquired thanks to a live-cell lens-free imaging system and a thin-layer agar media made of 20 μl BHI (Brain Heart Infusion) to train our deep learning networks. Our architecture proposal achieved interesting results on a dataset constituted of seven different pathogenic bacteria-Staphylococcus aureus (S. aureus), Enterococcus faecium (E. faecium), Enterococcus faecalis (E. faecalis), Staphylococcus epidermidis (S. epidermidis), Streptococcus pneumoniae R6 (S. pneumoniae), Streptococcus pyogenes (S. pyogenes), Lactococcus Lactis (L. Lactis). At T = 8h, our detection network reached an average 96.0% detection rate while our classification network precision and sensitivity averaged around 93.1% and 94.0% respectively, both were tested on 1908 colonies. Our classification network even obtained a perfect score for E. faecalis (60 colonies) and very high score for S. epidermidis at 99.7% (647 colonies). Our method achieved those results thanks to a novel technique coupling convolutional and recurrent neural networks together to extract spatio-temporal patterns from unreconstructed lens-free microscopy time-lapses.

Detection and identification of pathogenic bacteria isolated from biological samples (blood, urine, sputum, etc.) are crucial steps in accelerated clinical diagnosis. However, accurate and rapid identification remain difficult to achieve due to the challenge of having to analyse complex and large samples. Current solutions (mass spectrometry, automated biochemical testing, etc.) propose a trade-off between time and accuracy, achieving satisfactory results at the expense of time-consuming processes, which can also be intrusive, destructive and costly. Moreover, those techniques tend to require an overnight subculture on solid agar medium delaying bacteria identification by 12–48 hours, thus preventing rapid prescription of appropriate treatment as it hinders antibiotic susceptibility testing. In this study, lens-free imaging is presented as a possible solution to achieve a quick and accurate wide range, non-destructive, label-free pathogenic bacteria detection and identification in real-time using micro colonies (10–500 μm) kinetic growth pattern combined with a two-stage deep learning architecture. Bacterial colonies growth time-lapses were acquired thanks to a live-cell lens-free imaging system and a thin-layer agar media made of 20 μl BHI (Brain Heart Infusion) to train our deep learning networks. Our architecture proposal achieved interesting results on a dataset constituted of seven different pathogenic bacteria—Staphylococcus aureus (S. aureus), Enterococcus faecium (E. faecium), Enterococcus faecalis (E. faecalis), Staphylococcus epidermidis (S. epidermidis), Streptococcus pneumoniae R6 (S. pneumoniae), Streptococcus pyogenes (S. pyogenes), Lactococcus Lactis (L. Lactis). At T = 8h, our detection network reached an average 96.0% detection rate while our classification network precision and sensitivity averaged around 93.1% and 94.0% respectively, both were tested on 1908 colonies. Our classification network even obtained a perfect score for E. faecalis (60 colonies) and very high score for S. epidermidis at 99.7% (647 colonies). Our method achieved those results thanks to a novel technique coupling convolutional and recurrent neural networks together to extract spatio-temporal patterns from unreconstructed lens-free microscopy time-lapses. Author summary In order to combat the spread of resistances and to meet the specific needs of infection diagnosis in rural and remote settings, there is a strong need for new diagnostic tools that are faster, space efficient and less expensive than conventional ones while still being accurate. To answer those criteria, optical identification methods such as lens-free microscopy seem best-suited. Indeed, lens-free microscopy is an imaging technique that allows for continuous image acquisition during incubation, as it is non-destructive and compact enough to be integrated into an incubator. Here, we propose the combined use of lens-free microscopy imaging and deep learning to train neural networks able to detect and identify pathogenic species thanks to bacteria growth time-lapses on thin-layer agar and a novel deep learning architecture combining recurrent and convolutional neural networks. We achieved high quality detection and identification from 8 hours onwards on a dataset composed of 7 different pathogenic strains. Our study provides new results proving the usefulness of lens-free imaging on pathogenic species for rapid detection and identification tasks in low-resource settings.

Two-component regulatory systems (TCS) are among the most widespread mechanisms that bacteria use to sense and respond to environmental changes. In the human pathogen Streptococcus pneumoniae, a total of 13 TCS have been identified and many of them have been linked to pathogenicity. Notably, TCS01 strongly contributes to pneumococcal virulence in several infection models. However, it remains one of the least studied TCS in pneumococci and its functional role is still unclear. In this study, we demonstrate that TCS01 cooperates with a BceAB-type ABC transporter to sense and induce resistance to structurally-unrelated antimicrobial peptides of bacterial origin that all target undecaprenyl-pyrophosphate or lipid II, which are essential precursors of cell wall biosynthesis. Even though tcs01 and bceAB genes do not locate in the same gene cluster, disruption of either of them equally sensitized the bacterium to the same set of antimicrobial peptides. We show that the key function of TCS01 is to upregulate the expression of the transporter, while the latter appears the main actor in resistance. Electrophoretic mobility shift assays further demonstrated that the response regulator of TCS01 binds to the promoter region of the bceAB genes, implying a direct control of these genes. The BceAB transporter was overexpressed and purified from E. coli. After reconstitution in liposomes, it displayed substantial ATPase and GTPase activities that were stimulated by antimicrobial peptides to which it confers resistance to, revealing new functional features of a BceAB-type transporter. Altogether, this inducible defense mechanism likely contributes to the survival of the opportunistic microorganism in the human host, in which competition among commensal microorganisms is a key determinant for effective host colonization and invasive path.

ABSTRACT The capsule is the dominant Streptococcus pneumoniae virulence factor, yet how variation in capsule thickness is regulated is poorly understood. Here, we describe an unexpected relationship between mutation of adcAII, which encodes a zinc uptake lipoprotein, and capsule thickness. Partial deletion of adcAII in three of five capsular serotypes frequently resulted in a mucoid phenotype that biochemical analysis and electron microscopy of the D39 adcAII mutants confirmed was caused by markedly increased capsule thickness. Compared to D39, the hyperencapsulated ΔadcAII mutant strain was more resistant to complement-mediated neutrophil killing and was hypervirulent in mouse models of invasive infection. Transcriptome analysis of D39 and the ΔadcAII mutant identified major differences in transcription of the Sp_0505-0508 locus, which encodes an SpnD39III (ST5556II) type I restriction-modification system and allelic variation of which correlates with capsule thickness. A PCR assay demonstrated close linkage of the SpnD39IIIC and F alleles with the hyperencapsulated ΔadcAII strains. However, transformation of ΔadcAII with fixed SpnD39III alleles associated with normal capsule thickness did not revert the hyperencapsulated phenotype. Half of hyperencapsulated ΔadcAII strains contained the same single nucleotide polymorphism in the capsule locus gene cps2E, which is required for the initiation of capsule synthesis. These results provide further evidence for the importance of the SpnD39III (ST5556II) type I restriction-modification system for modulating capsule thickness and identified an unexpected linkage between capsule thickness and mutation of ΔadcAII. Further investigation will be needed to characterize how mutation of adcAII affects SpnD39III (ST5556II) allele dominance and results in the hyperencapsulated phenotype. IMPORTANCE The Streptococcus pneumoniae capsule affects multiple interactions with the host including contributing to colonization and immune evasion. During infection, the capsule thickness varies, but the mechanisms regulating this are poorly understood. We have identified an unsuspected relationship between mutation of adcAII, a gene that encodes a zinc uptake lipoprotein, and capsule thickness. Mutation of adcAII resulted in a striking hyperencapsulated phenotype, increased resistance to complement-mediated neutrophil killing, and increased S. pneumoniae virulence in mouse models of infection. Transcriptome and PCR analysis linked the hyperencapsulated phenotype of the ΔadcAII strain to specific alleles of the SpnD39III (ST5556II) type I restriction-modification system, a system which has previously been shown to affect capsule thickness. Our data provide further evidence for the importance of the SpnD39III (ST5556II) type I restriction-modification system for modulating capsule thickness and identify an unexpected link between capsule thickness and ΔadcAII, further investigation of which could further characterize mechanisms of capsule regulation.

ABSTRACT Ovococci form a morphological group that includes several human pathogens (enterococci and streptococci). Their shape results from two modes of cell wall insertion, one allowing division and one allowing elongation. Both cell wall synthesis modes rely on a single cytoskeletal protein, FtsZ. Despite the central role of FtsZ in ovococci, a detailed view of the in vivo nanostructure of ovococcal Z-rings has been lacking thus far, limiting our understanding of their assembly and architecture. We have developed the use of photoactivated localization microscopy (PALM) in the ovococcus human pathogen Streptococcus pneumoniae by engineering spDendra2, a photoconvertible fluorescent protein optimized for this bacterium. Labeling of endogenously expressed FtsZ with spDendra2 revealed the remodeling of the Z-ring's morphology during the division cycle at the nanoscale level. We show that changes in the ring's axial thickness and in the clustering propensity of FtsZ correlate with the advancement of the cell cycle. In addition, we observe double-ring substructures suggestive of short-lived intermediates that may form upon initiation of septal cell wall synthesis. These data are integrated into a model describing the architecture and the remodeling of the Z-ring during the cell cycle of ovococci. IMPORTANCE The Gram-positive human pathogen S. pneumoniae is responsible for 1.6 million deaths per year worldwide and is increasingly resistant to various antibiotics. FtsZ is a cytoskeletal protein polymerizing at midcell into a ring-like structure called the Z-ring. FtsZ is a promising new antimicrobial target, as its inhibition leads to cell death. A precise view of the Z-ring architecture in vivo is essential to understand the mode of action of inhibitory drugs (see T. den Blaauwen, J. M. Andreu, and O. Monasterio, Bioorg Chem 55:27–38, 2014, doi:10.1016/j.bioorg.2014.03.007, for a review on FtsZ inhibitors). This is notably true in ovococcoid bacteria like S. pneumoniae, in which FtsZ is the only known cytoskeletal protein. We have used superresolution microscopy to obtain molecular details of the pneumococcus Z-ring that have so far been inaccessible with conventional microscopy. This study provides a nanoscale description of the Z-ring architecture and remodeling during the division of ovococci.

Release of conserved cytoplasmic proteins is widely spread among Gram-positive and Gram-negative bacteria. Because these proteins display additional functions when located at the bacterial surface, they have been qualified as moonlighting proteins. The GAPDH is a glycolytic enzyme which plays an important role in the virulence processes of pathogenic microorganisms like bacterial invasion and host immune system modulation. However, GAPDH, like other moonlighting proteins, cannot be secreted through active secretion systems since they do not contain an N-terminal predicted signal peptide. In this work, we investigated the mechanism of GAPDH export and surface retention in Streptococcus pneumoniae, a major human pathogen. We addressed the role of the major autolysin LytA in the delivery process of GAPDH to the cell surface. Pneumococcal lysis is abolished in the ΔlytA mutant strain or when 1% choline chloride is added in the culture media. We showed that these conditions induce a marked reduction in the amount of surface-associated GAPDH. These data suggest that the presence of GAPDH at the surface of pneumococcal cells depends on the LytA-mediated lysis of a fraction of the cell population. Moreover, we demonstrated that pneumococcal GAPDH binds to the bacterial cell wall independently of the presence of the teichoic acids component, supporting peptidoglycan as a ligand to surface GAPDH. Finally, we showed that peptidoglycan-associated GAPDH recruits C1q from human serum but does not activate the complement pathway.

Understanding the relationship between protein sequence and molecular recognition selectivity remains a major challenge. The antibody fragment scFv1F4 recognizes with sub nM affinity a decapeptide (sequence 6TAMFQDPQER15) derived from the N-terminal end of human papilloma virus E6 oncoprotein. Using this decapeptide as antigen, we had previously shown that only the wild type amino-acid or conservative replacements were allowed at positions 9 to 12 and 15 of the peptide, indicating a strong binding selectivity. Nevertheless phenylalanine (F) was equally well tolerated as the wild type glutamine (Q) at position 13, while all other amino acids led to weaker scFv binding. The interfaces of complexes involving either Q or F are expected to diverge, due to the different physico-chemistry of these residues. This would imply that high-affinity binding can be achieved through distinct interfacial geometries. In order to investigate this point, we disrupted the scFv-peptide interface by modifying one or several peptide positions. We then analyzed the effect on binding of amino acid changes at the remaining positions, an altered susceptibility being indicative of an altered role in complex formation. The 23 starting variants analyzed contained replacements whose effects on scFv1F4 binding ranged from minor to drastic. A permutation analysis (effect of replacing each peptide position by all other amino acids except cysteine) was carried out on the 23 variants using the PEPperCHIP® Platform technology. A comparison of their permutation patterns with that of the wild type peptide indicated that starting replacements at position 11, 12 or 13 modified the tolerance to amino-acid changes at the other two positions. The interdependence between the three positions was confirmed by SPR (Biacore® technology). Our data demonstrate that binding selectivity does not preclude the existence of alternative high-affinity recognition modes.

The synthesis of peptidoglycan, the major component of the bacterial cell wall, is essential to cell survival, yet its mechanism remains poorly understood. In the present work, we have isolated several membrane protein complexes consisting of the late division proteins of Streptococcus pneumoniae: DivIB, DivIC, FtsL, PBP2x and FtsW, or subsets thereof. We have co-expressed membrane proteins from S. pneumoniae in Escherichia coli. By combining two successive affinity chromatography steps, we obtained membrane protein complexes with a very good purity. These complexes are functional, as indicated by the retained activity of PBP2x to bind a fluorescent derivative of penicillin and to hydrolyze the substrate analogue S2d. Moreover, we have evidenced the stabilizing role of protein-protein interactions within each complex. This work paves the way for a complete reconstitution of peptidoglycan synthesis in vitro, which will be critical to the elucidation of its intricate regulation mechanisms.

Zinc (Zn(2+)) homeostasis is critical for pathogen host colonization and invasion. Polyhistidine triad (Pht) proteins, located at the surface of various streptococci, have been proposed to be involved in Zn(2+) homeostasis. The phtD gene, coding for a Zn(2+)-binding protein, is organized in an operon with adcAII coding for the extracellular part of a Zn(2+) transporter. In the present work, we investigate the relationship between PhtD and AdcAII using biochemical and structural biology approaches. Immuno-precipitation experiments on purified membranes of Streptococcus pneumoniae (S. pneumoniae) demonstrate that native PhtD and AdcAII interact in vivo confirming our previous in vitro observations. NMR was used to demonstrate Zn(2+) transfer from the Zn(2+)-bound form of a 137 amino acid N-terminal domain of PhtD (t-PhtD) to AdcAII. The high resolution NMR structure of t-PhtD shows that Zn(2+) is bound in a tetrahedral site by histidines 83, 86, and 88 as well as by glutamate 63. Comparison of the NMR parameters measured for apo- and Zn(2+)-t-PhtD shows that the loss of Zn(2+) leads to a diminished helical propensity at the C-terminus and increases the local dynamics and overall molecular volume. Structural comparison with the crystal structure of a 55-long fragment of PhtA suggests that Pht proteins are built from short repetitive units formed by three β-strands containing the conserved HxxHxH motif. Taken together, these results support a role for S. pneumoniae PhtD as a Zn(2+) scavenger for later release to the surface transporter AdcAII, leading to Zn(2+) uptake.

Lipoproteins are an important class of surface associated proteins that have diverse roles and frequently are involved in the virulence of bacterial pathogens. As prolipoproteins are attached to the cell membrane by a single enzyme, prolipoprotein diacylglyceryl transferase (Lgt), deletion of the corresponding gene potentially allows the characterisation of the overall importance of lipoproteins for specific bacterial functions. We have used a Δlgt mutant strain of Streptococcus pneumoniae to investigate the effects of loss of lipoprotein attachment on cation acquisition, growth in media containing specific carbon sources, and virulence in different infection models. Immunoblots of triton X-114 extracts, flow cytometry and immuno-fluorescence microscopy confirmed the Δlgt mutant had markedly reduced lipoprotein expression on the cell surface. The Δlgt mutant had reduced growth in cation depleted medium, increased sensitivity to oxidative stress, reduced zinc uptake, and reduced intracellular levels of several cations. Doubling time of the Δlgt mutant was also increased slightly when grown in medium with glucose, raffinose and maltotriose as sole carbon sources. These multiple defects in cation and sugar ABC transporter function for the Δlgt mutant were associated with only slightly delayed growth in complete medium. However the Δlgt mutant had significantly reduced growth in blood or bronchoalveolar lavage fluid and a marked impairment in virulence in mouse models of nasopharyngeal colonisation, sepsis and pneumonia. These data suggest that for S. pneumoniae loss of surface localisation of lipoproteins has widespread effects on ABC transporter functions that collectively prevent the Δlgt mutant from establishing invasive infection.

BackgroundMembrane proteins are the targets of 50% of drugs, although they only represent 1% of total cellular proteins. The first major bottleneck on the route to their functional and structural characterisation is their overexpression; and simply choosing the right system can involve many months of trial and error. This work is intended as a guide to where to start when faced with heterologous expression of a membrane protein.Methodology/principal findingsThe expression of 20 membrane proteins, both peripheral and integral, in three prokaryotic (E. coli, L. lactis, R. sphaeroides) and three eukaryotic (A. thaliana, N. benthamiana, Sf9 insect cells) hosts was tested. The proteins tested were of various origins (bacteria, plants and mammals), functions (transporters, receptors, enzymes) and topologies (between 0 and 13 transmembrane segments). The Gateway system was used to clone all 20 genes into appropriate vectors for the hosts to be tested. Culture conditions were optimised for each host, and specific strategies were tested, such as the use of Mistic fusions in E. coli. 17 of the 20 proteins were produced at adequate yields for functional and, in some cases, structural studies. We have formulated general recommendations to assist with choosing an appropriate system based on our observations of protein behaviour in the different hosts.Conclusions/significanceMost of the methods presented here can be quite easily implemented in other laboratories. The results highlight certain factors that should be considered when selecting an expression host. The decision aide provided should help both newcomers and old-hands to select the best system for their favourite membrane protein.