Skip to main content
European Commission logo print header

Combating Antibiotics Resistant Pneumococci by Novel Strategies Based on in vivo and in vitro Host – Pathogen Interactions

Final Report Summary - CAREPNEUMO (Combating Antibiotics Resistant Pneumococci by Novel Strategies Based on in vivo and in vitro Host – Pathogen Interactions)

Executive Summary:
The objective of CAREPNEUMO is to apply a multi-disciplinary approach that includes epidemiology, host-pathogen interactions, infection models and intervention strategies to combat antibiotic resistant S. pneumoniae. The consortium brought together 12 research organizations and 1 SME with expertise in the above-mentioned areas. For epidemiological analysis a total of more than 3000 isolates from different clinical manifestations were collected from Argentina, Germany, Greece, India, Poland and Portugal. These isolates were sero-typed and tested for antibiotics resistance and vaccine pressure serotype switching. In different geographical regions, different prevalence of serotypes was observed. The results showed increased emergence of non-vaccine serotypes after introduction of 7-valent vaccine and increasing antibiotic resistance. Also, serotype 6D was for the first time isolated from invasive pneumococcal disease in Europe. Interestingly, a very high carriage rate of 66% was observed in Indian children. In host-pathogen interaction section, the consortium sequenced 25 different pneumococcal clinical isolates and identified genetic islands that contribute to pneumococcal virulence. In addition, adherence and invasion mechanisms, structure of choline binding proteins and mechanisms underlying multidrug tolerance were elucidated. The results demonstrated for the first time that adherence of pneumococci to endothelial cells stimulated the release of active protein components and cytokines. The interaction of pneumococcal phosphoglyceric acid with plasminogen was identified a new tissue invasion mechanism. To determine the role of surface exposed lipoproteins a number of mutants were constructed and lipoproteins purified for further analysis, using experimental cell culture system and structural biology approach. A murine model for pneumococcal infections was established and is now being used to determine the virulence potential of different clinical isolates collected in epidemiology work packages. It was found that the behaviour of strains in this mouse model reflected the severity of infection in humans. In intervention section, the consortium identified new pneumococcal targets for development of a new generation of antimicrobials against resistant pneumococci. Atropine, ipratopium and polycholine dendrimers were found to be effective inhibitors of the attachment of choline-binding proteins, the important pneumococcal virulence factors, to the teichoic acids in the cell wall. The results indicate that these choline analogues, either free or attached to dendrimers, represent molecules with a potential to become novel antibiotics against Pneumococcus. On therapy side the focus was on meningitis caused by pneumococci resistant to antibiotics. It could be shown that an increase of phagocytic ability of reactive microglia through the stimulation of the TLR system leads to improved host defence of the brain against invasive pneumococci. We could also show that antibodies against surface proteins of pneumococci could reduce the neuronal injury in meningitis. The prevention strategies of CAREPNEUMO include the development of a novel polysaccharide-glycolipid conjugate vaccine and a protein based universal vaccine. The conjugated vaccine was based on serotype 4 conjugated with alpha GalCer which gave high titers of IgG and IgM against S. pneumoniae, indicating that alpha GalCer is a potent adjuvant. In mouse model of infection, the immunization with vaccine candidates showed full protection against the challenge. Reverse Vaccinology, using patient sera, led to the identification of a number of peptides with features of a promising vaccine candidate. Mice immunized with these peptides showed increased survival rates after challenge with strains of different serotypes. The set of biological data produced with the five generic fusion proteins can be used to rank the efficacy of these proteins as: p722>p721>p728>p727>p729. The results obtained clearly indicate a potential socio-economic impact and relevant societal implications. The prevalence of different serotypes in different regions and the emergence of non-vaccine serotypes indicate the need of a new region-specific or universal vaccine. The results obtained on the mechanisms of host-pathogen interaction, structural analysis and novel target identification would contribute towards the development of new combat strategies for resistant pneumococci. The establishment of a murine model of pneumococcal infection will be useful in validation of mechanisms and testing of two different vaccine strategies. All the milestones have been achieved, and the expected final results of CAREPNEUMO would for a strong basis for development of new and universal preventive and therapeutic strategies to control infections due to antibiotics resistant pneumococci.

Project Context and Objectives:
The objectives of this proposal were defined in order to obtain new knowledge on the molecular epidemiology of pneumococcal diseases and antibiotic resistance in different parts of the world, apply state-of-the-art technology to study host-pathogen interactions and apply innovative technologies for improving existing intervention strategies. Moreover, it would contribute towards the implementation of measures for prevention, control and treatment of pneumococcal diseases, especially those caused by multi-drug resistant strains. To achieve the purpose of this proposal, we defined the following objectives:

Objective 1: Epidemiology of drug resistance and vaccine pressure replacement of S. pneumoniae

The major tasks of this objective were to undertake monitoring of prevalent S. pneumoniae serotypes and their resistance profiles in various countries. This included the epidemiology of the serotype replacement in areas where pneumococcal vaccine is in use. Since the prevalence of serotypes as well as the magnitude and quality of drug resistance is different in different parts of the world, this objective addressed the epidemiology of S. pneumoniae in six countries, which include an Asian and a South American country. These studies determined the frequency of emergence of antibiotic resistance, particularly in those serotypes that have potential to spread rapidly in the community. The second part of this objective was to determine the changes in pneumococcal disease after introduction of the 7-valent pneumococcal conjugate vaccine in different regions.

The first part of this objective was achieved by collection of more than 3000 clinical isolates from different countries. All these isolates were serotyped, characterized and used to determine the antibiotic resistance. MLST analysis of the isolates showed a different pattern of clonal distribution in different countries. In Germany and Greece, the serotype 19A showed changed distribution. In Greece, clonal spread among both invasive and non-invasive isolates belonging to all four emerging serotypes 6A, 23A, 23B and 19A was observed. In Poland, the MLST of invasive isolates showed that drug-resistant epidemic clones constituted a significant part of this group. Overall, the antibiotic resistance is increasing, also in Otitis media patients in Argentina. Another cause of concern is the high percentage of Pneumococcus carriage in some countries like India.

The second part of the objective was achieved by clinical surveys of pneumococcal isolates in countries where 7-valent vaccine has been introduced. In Germany, the incidence of non-vaccine serotypes among children has significantly increased. Also in Greece the pressure of 7-valent vaccine introduced in 2004 and fully implemented in clinical practice during 2005-2006 resulted in a significant change of circulating serotypes. The main multi-resistant serotype 19F was replaced by clonal spread of serotypes 19A, 6A, 23A and 23B. The highlights of the results of objective 1 are the different serotype distribution of clinical isolates in different regions, increasing antibiotic resistance and pneumococcal carriage, emergence of serotype 6D in invasive disease in Europe and the serotype switching in countries where vaccine was introduced. This objective not only provided useful information about epidemiology, but also formed a base to achieve other objectives of CAREPNEUMO.

Objective 2: Analysis of host-pathogen interactions and identification of potential therapeutic targets and vaccine candidates

A prerequisite for developing new control strategies is the detailed understanding of the interaction between pneumococcus and the host cells. The analysis of host-pathogen interaction included the prevalent virulent strains, which were continuously identified in the work packages dealing with the epidemiological studies. The molecular mechanisms of host-pathogen interactions were elucidated by applying genomics and cell biological approaches. Identification and characterization of the factors involved were an important part of objective 2. The biological functions of these factors were determined by using pathogenicity tests involving cell cultures as well as animal models of pneumococcal diseases.
Host-pathogen interactions were studied at different levels to achieve this objective. Applying several cell culture based methods, we could demonstrate for the first time that adherence of S. pneumoniae to endothelial cells stimulated the exocytosis of these vesicles and thereby the secretion of active protein components and cytokines. ELISA-based biochemical studies allowed a quantification of vesicle-derived proteins secreted in response to pneumococcal adherence. Furthermore we successfully characterized the role of the pneumococcal cytotoxin pneumolysin and the effect of capsular polysaccharides in stimulation of vesicle exocytosis by pneumococci by using different pneumococcal mutants. A further very important result was obtained after generation of a transwell-based basal infection model of endothelial cells. The release of major vesicle components was induced after infection with pneumococci from both the apical and the basal cell surfaces of polarized endothelium, which implies a stimulation of vesicle exocytosis from both directions: From inside the blood system during a septicaemia and also following invasive pneumococcal transmigration from lung tissue into the bloodstream. This objective also revealed an important role of lipoproteins in host-pneumococcal interactions because of their involvement in induced signal transduction cascades and toll-like receptor dependent innate immune reactions. Further mechanistic insight was obtained by structural biological analysis of surface proteins that play a crucial role in host interactions. A part of this objective was achieved by whole-genome sequencing of 25 pneumococcal clinical isolates. This work has pointed to the role of a number of genes in varying aspects of pneumococcal disease. Polymorphisms within the pneumolysin, hyaluronidase, neuraminidase, choline-binding protein A, serine-threonine kinase and pyruvate oxidase genes have been determined. A wider range of clinical isolates has been collected in which a number of these polymorphisms have been tested to determine their presence and function in the pneumococcal population. In this objective we also developed two different mouse models. The pneumococcal infection model was used to evaluate the potential contribution of dendritic cells to systemic dissemination from the lower respiratory tract. The results suggested that pneumococci use dendritic cells for systemic dissemination, and interfering with functional activity of dendritic cells could be used as therapeutic intervention. The second model was used to optimize therapy of meningitis by modulating the growth of pneumococci. It could be shown that an increase of phagocytic ability of reactive microglia leads to improved host defense of the brain against invasive pneumococci.

Objective 3: Development of improved vaccine and intervention strategies

This objective dealt with the development of new therapeutic and vaccine tools against pneumococcal infections based on the knowledge generated in objectives 1 and 2. For designing new therapeutics, the molecular basis of pathogen recognition through specific components of pneumococcal cell wall by host proteins was identified by using a structural biological approach. In the area of vaccine development, this objective was to deal with a novel polysaccharide glycolipid conjugate vaccine as well as the development of a protein-based universal vaccine.

This objective was achieved in two parts. The first dealt with the new pneumococcal targets and development of new generations of antimicrobials, and the second with the development of novel vaccine strategies. The focus for novel antimicrobials was the choline binding proteins which are important virulence factors. The results indicated that atropine, ipratopium and polycholine dendrimers were effective inhibitors of the attachment of choline-binding proteins to the teichoic acids in the cell wall and therefore represent molecules with a potential to become novel antibiotics against pneumococcal infections. On the vaccine side, our focus was on the development of novel polysaccharide glycolipids conjugate vaccine and a protein-based universal vaccine. The conjugated vaccine was based on serotype 4 conjugated with alpha GalCer which gave high titers of IgG and IgM against S. pneumoniae, indicating that alpha GalCer is a potent adjuvant. In mouse model of infection, the immunization with vaccine candidates showed full protection against the challenge. Reverse Vaccinology, using patient sera, led to the identification of a number of peptides with features of a promising vaccine candidate. Mice immunized with these peptides showed increased survival rates after challenge with strains of different serotypes. The set of biological data produced with the five generic fusion proteins can be used to rank the efficacy of these proteins as: p722>p721>p728>p727>p729.

The combined results of the three objectives of CAREPNEUMO clearly indicate a potential socio-economic impact and relevant societal implications. The observed prevalence of different serotypes in different regions and the emergence of non-vaccine serotypes indicate the need of new region-specific or universal vaccines. The results obtained on the mechanisms of host-pathogen interaction, structural analysis and novel target identification would contribute towards the design of novel strategies to combat antibiotic resistant pneumococci. The development of murine models will be useful in validation of mechanisms and testing of two different vaccine strategies. CAREPNEUMO has delivered a strong basis for development of new and universal preventive and therapeutic strategies to control infections caused by antibiotic resistant pneumococci.

Project Results:
WP No. 1: Dr. Mark van der Linden; German National Reference Center for Streptococci, University Hospital RWTH Aachen, Germany: Changes in pneumococcal disease after the introduction of conjugate vaccine in Germany

The aim of this project was to gain insight in the impact of pneumococcal conjugate vaccination among German children under 2 years of age on invasive pneumococcal disease (IPD) among children. In particular, the aims were: Collection of isolates from IPD in children (age 0-15 years), analysis of changes in serotype distribution, analysis of possible clonal changes using MLST.

In August 2006 a general recommendation for vaccination of young children (≤2 years of age) with the 7-valent pneumococcal conjugate vaccine (7vPCV) was issued in Germany. The introduction of 7vPCV in Germany may result in serotype replacement and serotype switching. In this study we planned to analyse the results of the introduction of 7vPCV in Germany, concerning total number of infections, serotype composition and clonal relatedness of the isolates.

To this end isolates of invasive pneumococcal disease were collected in children (≤15 years of age) and thoroughly characterize these isolates concerning species and serotype. Calculation of incidence rates, using capture-recapture and comparison with data from before the introduction of 7vPCV were used to reveal its effect on disease burden. Serotyping of all isolates allowed for comparison of the serotype composition before and after vaccine introduction and assessment of possible serotype replacement.
Finally, multilocus sequence typing of isolates collected 2,5 years before and 2,5 years after the introduction of the vaccine revealed the clonal relatedness of isolates and changes in distribution of clones in the two groups and enabled the detection of possible serotype switching.

Collection and characterization of isolates from IPD in German children has been successfully completed for the time period 2004-2008. All isolates have been characterized and serotyped. The results show the first effects of the vaccination program. The number of isolates with PCV7-vaccine serotypes decreased markedly. The amount of isolates with non-PCV7-vaccine serotypes has increased. But we have also witnessed an increased percentage of reported isolates that were actually sent to the Reference Center for analysis. Calculations of serotype specific incidence rates showed a decrease in vaccine serotypes for children under 2 years of age, but a significant increase in non-vaccine serotype incidence could not be determined. Therefore, the increase in isolates from non-vaccine serotypes sent to the Reference Center seems to be in part due to increased awareness.

MLST analyses of 720 isolates has been completed. Analysis shows that for most serotypes the clonal distribution has remained unchanged. Only for Serotype 19A the clonal distribution has changed. This change continued in the years 2008 to 2011, and has been followed closely. An expansion of the existing clones could be observed, accompanied by a loss of some clones and the appearance of others. Multi drug Resistant (MDR) clones appeared in Germany, but their numbers have remained very low. No cases of serotype switching have been found.

Incidence of IPD caused by PCV7-vaccine serotypes among children under two years has decreased from 15.6/100.000 before vaccination to 3.6/100.00 in 2008, and to 1.2/100.000 in 2010. Incidence decreases were also reported among 2-4 year olds (4.8 to 0.2/100.00)and 5 -15 year olds (0.4 to 0.2/100.000). The incidence of non-PCV7-vaccine serotypes among children under two years has increased from 8.2 to 11.7/100.000. Among 2-4 year old the increase was from 1.7 to 4.9/100.000 and for for 5-15 year olds from 1.0 to 1.7/100.000.

Work package No. 2 (Prof. Em. Nicholas J. Legakis, National and Kapodistrian University of Athens, Athens, Greece): Monitoring of S. pneumoniae isolates from invasive and non-invasive diseases in children after introduction of the vaccine in Greece

S. pneumoniae strains causing disease in children have always varied, both phenotypically and genotypically, with respect to geographic location, as well as over time. Since 2005, the selective pressure of the heptavalent conjugate vaccine has been an additional cause of variation. Some serotypes have been observed to decline after vaccination – to be replaced by emergent ones – but, nevertheless, some persisted. The present WP aimed to monitor serotype, antimicrobial drug resistance profile and clonal changes and fluctuations, under the selective pressure of the heptavalent conjugate vaccine, introduced in Greece in 2005.

A total of 951 isolates from both invasive and non-invasive pneumococcal disease in children were collected, together with appropriate epidemiological data, to supplement the existing collection, going back to 1995.

Susceptibility testing indicated an increase of penicillin intermediate susceptibility and erythromycin resistance after 2007, as opposed to stable penicillin (full) resistance rates. The overall predominant MLS phenotype was M (61.5% among the EryR isolates) – but only until 2009, when it was replaced by CR.

Serotyping indicated that serotypes 19F, 6B, 23F, 14, and 9V (all included in the current 7valent vaccine) decreased, whereas types 6A, 23A, 23B and 19A showed an increase.
MLST typing indicated clonal spread among both invasive and non-invasive isolates belonging to all four "new/emerging" serotypes (6A, 23A, 23B and 19A). In contrast, among invasive strains belonging to the decreasing but still persisting serotypes (19F, 6B, 23F, 14, and 9V), clonal spread was detected mainly among serotype 19F isolates (as reported previously in Greece).

These data indicate that the pressure of the 7valent vaccine introduced in 2004 and fully implemented in clinical practice during 2005-2006 resulted in a significant change of the circulating serotypes, especially after 2007. The main multi-resistant serotype, 19F, slowed or even decreased its clonal spread that had been reported previously in Greece, and was eventually replaced by clonal spread of serotypes 19A, 6A, 23A and 23B.

The change in serotype distribution resulted also in a significant change of the main antimicrobial resistance phenotypes detected. CR-MLSB replaced M as the main MLS resistance phenotype after 2008. In addition, although penicillin intermediate susceptibility increased, full resistance did not. Serotypes 19A and 6A, which replaced 19F and 6B, showed lower penicillin resistance but higher penicillin intermediate susceptibility rates.

Workpackage No. 3 (Prof. Dr. Waleria Hryniewicz, National Medicines Institute, Warsaw, Poland): Population structure, resistance mechanisms and virulence factors of S. pneumoniae in Poland

This workpackage was aimed at the in-depth characterization of the current epidemiological situation concerning Streptococcus pneumoniae in Poland. Isolates from particular types of pneumococcal diseases, especially the invasive disease, and isolates resistant to anti¬microbial drugs were of a special interest of this study.
T
he first step was establishing the representative collection of pneumococcal isolates for the subsequent investigations. The isolates from normally sterile sites (mostly blood and cerebrospinal fluid), lower respiratory tract and nasopharyngeal carriage were obtained from several laboratories from the whole area of Poland and analysed by susceptibility testing and serotyping. High non-susceptibility levels to penicillin, ampicillin, cephalosporins, macrolides and teracyclines were observed among collected isolates. Invasive isolates showed a high variability of serotypes, however, certain serotypes prevailed and the vaccine coverage was high (mass vaccination with conjugated antipneumococcal vaccine has not been introduced in Poland as yet). On the other hand, isolates showing resistance to beta-lactams and/or macrolides represented only a limited number of serotypes (including mostly 14, 23F, 9V, 19F, 6B).

These phenotypic studies were followed by molecular typing analyses and characterization of resistance and virulence determinants. Multilocus Sequence Typing (MLST) analysis allowed identification of major drug-resistant clones and clonal complexes among isolates from respiratory tract diseases, both from a likely importation (Spain9V-ST156, Spain6B-ST90, Spain23F-ST81, England14-ST9, Taiwan19F-ST320) and of the local origin (Poland23F-ST173 and Poland6B-ST315) as responsible for the current increase in the resistance levels. Further investigation of these clones revealed the acquisition of additional resistance determinants, such as e.g. modified pbp genes (resulting in the acquisition of ampicillin- and cephalosporin-resistance by penicillin-resistant strains) and the ermB gene by mef-positive strains (again leading to a broader resistance spectrum that included not only 14- and 15-membered macrolides but also 16-membered macrolides and lincosamides, as well as resulting in higher resistance levels to 14- and 15-membered macrolides).

MLST of invasive isolates showed that drug-resistant epidemic clones recently constitute a significant part of this group. Major drug-susceptible clones, observed also in other European countries (Denmark12F-CC218, Netherlands15B/C-CC199, Netherlands18C-CC113, Nether-lands3-CC180, Netherlands7F-CC191, Netherlands8-CC62, Sweden1-CC306, Sweden4-CC205) were found as well. In both groups several examples of serotype exchange (‘switch’) and divergence of a major surface antigen, pspA, within particular clones was observed. Thus, the pathogen shows efficient adaptation to the pressure caused both by the use of antimicrobials and by the human immune system.

Pneumococcus is a naturally transformable bacterium and this property is of a special relevance for the aforementioned adaptation to unfavourable conditions. We have shown that the competence locus comCDE is well conserved among clinical strains and that the transformation of several of these can be easily achieved in the laboratory conditions.

WP No. 4 (Prof. Dr. Mario Ramirez, Instituto Medicina Molecular, Portugal): Population dynamcis of S. pneumoniae under vaccine pressure in Portugal

The aim of this work package was to understand the dynamics, variability and response of the pneumococcal population to the use of the seven-valent conjugate vaccine available at the time. This knowledge will help to direct strategies aimed at reducing the impact of pneumococcal diseases. Specifically, we aimed at: Characterize the specific clones and serotypes decreasing upon vaccination implementation; Identify any emerging clones as important causes of invasive pneumococcal disease; Determine the diversity and distribution of the genes encoding pili-like structures, being considered has potential vaccine targets; Explore if there were conditionings to the horizontal transfer of DNA within the pneumococcal population.

1. We set out to characterize the pneumococcal population causing invasive disease in Portugal at a time of high seven-valent conjugate vaccine uptake.

2. We established that the initial decline shortly following vaccine use of the serotypes included in the vaccine as causes of pediatric invasive disease, has continued well into the late period. However, and in spite of close to having 75% of the children vaccinated, significant disease still occurred by vaccine serotypes. Most notably, these still accounted for a significant fraction of antibiotic resistant isolates, highlighting the importance of antibiotic use as a major driver of pneumococcal population dynamics. On the other hand, serotype 19A, a multiresistant serotype, continued to increase has a cause of pediatric invasive disease. Importantly we found that serotype 19A was associated with younger children, those targeted by vaccination, indicating that large benefits could be expected from a vaccine including this serotype. On the other hand, serotype 1 was associated with older children and empyema indicating that these serotypes may differ in their virulence factors causing them to be responsible for different diseases in different age groups, an issue that deserves further study. Overall, the data obtained supported the use of the seven-valent conjugate vaccine but highlighted the need for higher valency vaccines and continued surveillance (Aguiar et al. 2010).

3. The study of invasive pneumococcal disease among adults showed that the herd effect has been sustained in this population with a marked decline of the serotypes included in the seven-valent vaccine. The dominant serotypes in adults matched the most prevalent in children. In spite of its availability for more than a decade, we found no evidence of an effect of vaccination with the simple polysaccharide 23-valent vaccine. Similarly to children we found that some serotypes were not evenly distributed with age. In the case of adults, serotype 1 was associated with younger adults while serotype 3 was most frequent among the eldest group considered (≥65 yrs). Again, this points to heterogeneity within pneumococci that will be the object of future studies. Taken together, these data showed that, in spite of the marked changes in the major serotypes responsible for invasive infections the serotypes included in the new 13-valent conjugate vaccine still accounted for the majority of infections.

4. The analysis of the pilus islands of invasive isolates revealed yet another aspect of the changes induced by vaccination. The previously most frequent pilus island (PI-1) was reduced in frequency by more than 60%. In contrast, PI-2 increased in the population. We confirmed a strong association of the pilus island loci with particular lineages. The observed changes reflect the substitution of serotypes and consequently of the lineages that express them. The data also showed that although pili can be lost and acquired, these events seem to be infrequent and stable lineages with a defined genetic capability exist.

5. Critical to our understanding of pneumococcal evolution in face of vaccine pressure is to know the extent and limitations of horizontal gene exchange. To this end we studied the distribution and genetic characteristics of the two sub-populations carrying distinct alleles of the competence stimulating peptide. We found that these two sub-populations were genetically isolated. This finding has important consequences, since we noted a strong association of each allele with particular serotypes. The fact the population is not panmictic can create conditions in which one of the sub-populations may be in a better position to adapt to vaccine pressure than the other, conditioning the outcomes of vaccination.

Wp No. 5 (Dr Horacio Lopardo, Hospital Garrahan, Buenos Aires, Argentina): Antimicrobial resistance of Streptococcus pneumoniae from otitis media of pediatric patients

Since 02/05/09 to 31/08/10, 324 Argentinean children aged 1-120 months with a first episode of acute otitis media (AOM) were included. Demographic features and risk factors have been recorded and analyzed. There were no statistically significant differences between the clinical presentation and evolution of patients with AOM due to S. pneumoniae when compared with AOM due to other organisms. Tympanocentesis and culture of middle ear fluid was performed. Collection and characterization of pneumococci (Spn) have been successfully completed. In parallel, during the same period, 89 Spn were consecutively isolated from normally sterile fluids of children with invasive infections (INV). All isolates have been characterized and serotyped.

Penicillin, cefotaxime, amoxicillin, erythromycin, trimethoprim-sulfamethoxazole, tetracycline, rifampicin, chloramphenicol, vancomycin, levofloxacin, ofloxacin, and clindamycin resistance of AOM and INV isolates was determined by different methods. The prevalence of ermB (ribosomal methylase) and mefA (efflux pump) genes in macrolide-resistant Spn isolates from AOM and have been determined by the D-test and by PCR. Their genetic relatedness was determined by PFGE. MLST analyses of macrolide-resistant pneumococci was performed.

WP No. 6 (Dr. Anuradha Chakraborti, PGIMER, Chandigarh, India): Study of the prevalence of Streptococcus pneumoniae serotypes in North Indian Communities

The prevalence of S.pneumoniae from the community near Chandigarh and its surrounding areas and the antibiotic susceptibility pattern of the identified isolates has been done successfully along with the optimization of multiplex PCR assay to identify prevalent serotypes in our community. Further the adherence and invasive mechanisms with the selected virulent strains of S.pneumoniae in cell lines was standardized and done successfully.

Collection of nasopharyngeal cultures from healthy children from different regions of Chandigarh, the prevalence of S. pneumonia and their antibiotic susceptibility pattern has been successfully completed. A cross-sectional study was initiated in January 2010 in Union territory Chandigarh. A research team visited the selected sites and parents of children were sensitized regarding the significance of the study and its relevance in the current scenario and their consent for participation in the project was recorded. Institute ethical clearance was also obtained. 575 healthy children aged less than five years: 300 from urban region (Sector-38 and 41); 124 from rural area (Maloya village) and 151 from slums (Indira colony) of Chandigarh were enrolled. Nasopharyngeal specimens were collected and cultured as per standard microbiological methods and S. pneumoniae were identified. Our study revealed a significant nasopharyngeal carriage (49.6%) of S. pneumoniae among North Indian children with carriage rate considerably high (66.2%) in children residing in slums as compared to children in rural (46.8%) and urban (42.3%) areas.

S. pneumoniae were screened for antibiotic susceptibility against penicillin, erythromycin, co-trimoxazole, chloramphenicol, tetracycline, amoxicillin, vancomycin, ciprofloxacin, cefotaxime and levofloxacin on Mueller–Hinton agar medium supplemented with 5% sheep blood by Kirby Bauer disc diffusion method according to Clinical and Laboratory Standards Institute (CLSI) guidelines. All S. pneumoniae isolates were found to be sensitive for A, Ce and V, however, some isolates showed variable resistance with maximum resistance towards Co (97.2%) followed by T (28.4%), E (6.4%) and C (2.8%). A few isolates also showed intermediate susceptibility against T (11.2%), E (6%), Cf (3.2%), Co (2%), whereas only less than 1% isolates showed intermediate susceptibility against C and Le. Only 5.38% S. pneumoniae isolates tested emerged as multidrug resistant.

Using the multiplex PCR method with the known primers serotypes of only 50% of the isolates could be determined and because of shortage of time, as the project got delayed by almost 11 months, further optimization by real time PCR could not be completed. Genomic DNA extraction of S. pneumoniae was performed by QIAGEN DNA Mini protocol (Qiagen, Valencia, CA). A primer pair (primers cpsA-f and cpsA-r) was included as an internal control targeting the cpsA locus found in all pneumococci. Worldwide literature showing the prevalent types was intensively searched. The publications citing Asian data particularly the types circulating in the northern region of India were listed and based on that 39 primers were chosen, ordered and checked for the prevalent types circulating in our community. The primers were categorized into 7 groups based on their annealing temperature. Multiplex PCR thermo-cycling conditions for all the primers were followed as mentioned in the literature. The PCR amplification products were checked on 2% agarose gel. Out of 285 S. pneumoniae isolated, 209 were serotyped by the specific primers and about 104 isolates could be serotyped. A total of 27 serotypes were found: 6A/B/C, 5, 7C/B, 10A, 10F/C, 11A/D, 14, 15A/F, 15B/C, 19A, 22F/A, 23A etc. serotypes were circulating in our community, with 6A/B/C being the most prevalent. Six serotypes 6A/B/C, 11A/D, 15B/C, 10F/C, 34, 10A constituted 59.6% of our isolates. Maximum serotype variability (19 serotypes) was found among the isolates collected from the urban-slum region and minimum (11 serotypes) were identified from rural region. Serotypes 6A/B/C, 15A/F and 10A were common amongst isolates collected from urban, rural and slum region however, their frequency varied.

DNA sequencing of 29 representative S. pneumoniae strains (14 serotypes) was done to confirm their serotypes. The PCR amplification was done using serotype specific primers and Pwo polymerase. The amplicon was purified using QiAquick Kit (QIAGEN) and DNA sequencing was carried out using ABI Prism377 DNA sequencer (Applied Biosystems). DNA sequences were analysed with MicroSeq® ID analysis software.The sequences were analyzed using BLAST program (http://ncbi.nlm.nih.gov) of the National Center for Biotechnology Information (NCBI) and reconfirmed by submitting to Centers for Disease Control and Prevention, CDC. Results revealed that for 12 serotypes: 9N/L, 6C, 15B/C, 3, 35F/A/37, 15A/F, 10F/C, 22F/A, 14, 34, 23A, 11A/D, sequencing pattern was similar except for 6A/B/C where 6 isolates were sequenced as 6A, 1 as 6B and 2 as 6C and 7B/C where 1 was sequenced as 7C & 1 as 7B/C.

We successfully studied adherence and invasion of the most prevalent strains of S. pneumoniae from the population with A549 pulmonary epithelial cell line. Adherence to the A549 cell line was checked by light microscopy (Giemsa staining) and C.F.U count. S. pneumoniae strains showed variable adherence ranging from 0.03 – 24.2% respectively. We also optimized the invasion assay in A549 cells by Gentamicin protection assay. S. pneumoniae strains isolated from nasopharyngeal culturing were found to be poorly invasive.

WP No. 7 (Dr. Timothy Mitchell, University of Glasgow, Scotland): Genomic analysis and virulence of S. pneumoniae

The aim of this project was to investigate the genetic variation between S. pneumoniae isolates at both gene content and gene polymorphism levels: investigating the level of variation between pneumococci of clinical origin, identifying the genetic regions or loci of variation, elucidating the contribution of genetic variation to pneumococcal disease in experimental animals .

Streptococcus pneumoniae is a serious pathogen which is naturally transformable and known to exhibit considerable genetic variation. The pneumococcus is responsible for a spectrum of invasive and non-invasive disease, with a range of symptoms from the more life-threatening meningitis, pneumonia and sepsis to otitis media and sinusitis and is also asymptomatically carried in the nasopharynx. The genetic variation of this pathogen allows the colonization or invasion of different niches within the host.

During this project a core panel of twenty-five varied pneumococcal clinical isolates was chosen and first compared by microarray analysis for gene content, and subsequently genome sequenced. A number of additional phenotypically relevant strains were appended to this list.

The sequence data generated was used to assemble complete or as near complete genomes as possible. Draft genomes were compared in sub-groups of closely related strains, for example of the same serotype, to reduce any background of spurious differences, or genetic ‘noise’. Notably these included a group of serotype 1 strains with differing backgrounds of invasive disease potential; serotype 3 strains of similar clinical relevance, including a carriage strain; a pair of serotype 6A isolates with opposite phenotypes of carriage versus virulence; and a number of ‘paired’ isolates from two different body sites in a single patient. In addition to the clinical data from patient when isolated, strains were tested in vivo, in a murine model of sepsis and pneumonia. This has allowed a more detailed and flexible analysis of their virulence potential and pattern.

The gene/genome content across the groups of isolates has been compared bioinformatically for either presence or absence of genetic loci and also polymorphisms within loci. Some in vitro biological assays and also transcriptomic analysis has also been employed to investigate differences in gene expression between these strains. We have elucidated differences in a number of genes already identified as virulence factors, plus a number of genes and genetic loci not previously associated with pathogenesis requiring further study.

A number of isogenic mutant strains have been constructed in the laboratory within the Reference Genome Sequence strain TIGR4. These range from reconstruction of single nucleotide polymorphisms (SNPs) within single genes, to gene deletions or knockouts and allele replacements. The contribution of these genes to pneumococcal pathogenesis has been investigated by testing these mutants in the murine model.

This work has pointed to the role of a number of genes in varying aspects of pneumococcal disease. Polymorphisms within the pneumolysin, hyaluronidase, neuraminidase, choline-binding protein A, serine-threonine kinase and pyruvate oxidase genes have been determined. A wider range of clinical isolates has been collected in which a number of these polymorphisms have been tested to determine their presence and function in the pneumococcal population.

Additional loci have also been identified by bioinformatics analysis and await further investigation

WP No. 8 (Dr. Simone Bergmann, Helmholtz Centre for Infection Research, Braunschweig, Germany: Interaction of S. pneumoniae with endothelial cells

Invasive pneumococcal infections due to Streptococcus pneumoniae lead to inflammatory infection of the lung and brain, and to septic dissemination within the blood system. At the beginning of this EU-funded project, only limited information was available with respect to the response of the endothelial cell surface towards pneumococcal infection. Endothelial cells comprise a major part of the vessel walls of the lung vascular system and regulate main functional activities of the blood system. After the establishment of reliable and standardized endothelial in vitro cell culture infection models, the characterization of surface expression markers very quickly demonstrated the need to use primary human endothelial cells in these studies in order to generate results adequately mimicking the situation in vivo. As intensive response to inflammatory stimuli, primary endothelial cells secrete factors and proteins such as von Willebrand factor (vWF) from specialized vesicles. These factors comprise important key regulators of blood coagulation, immune defence and angiogenesis and are known to affect significantly the course of invasive bacterial infection. Our transcriptome-based microarray analyses, performed in the early stage of the project funding, revealed the induction of these endothelial-specific vesicle components after infection of primary lung endothelial cells with pneumococci. Applying several cell culture based methods, we could demonstrate for the first time that adherence of S. pneumoniae to these endothelial cells stimulated the exocytosis of these vesicles and thereby the secretion of active protein components and cytokines. ELISA-based biochemical studies allowed a quantification of vesicle-derived proteins secreted in response to pneumococcal adherence. Furthermore we successfully characterized the role of the pneumococcal cytotoxin pneumolysin and the effect of capsular polysaccharides in stimulation of vesicle exocytosis by pneumococci to detail employing different pneumococcal mutants. A further very important result was obtained after generation of a transwell-based basal infection model of endothelial cells. The release of major vesicle components was induced after infection with pneumococci from both the apical and the basal cell surfaces of polarized endothelium, which implies a stimulation of vesicle exocytosis from both directions: From inside the blood system during a septicaemia and also following invasive pneumococcal transmigration from lung tissue into the bloodstream. The results obtained from cell culture infection analyses generated a detailed picture of this new pathomechanism and were published recently in Cellular Microbiology. The data were presented on several national and international conferences and workshops. An intensive scientific exchange within the Carepneumo consortium resulted in very helpful discussion, which significantly improved the detailed investigations of this new pneumococcal pathomechanism. This project gained great benefit from the opportunity to use relevant isolates obtained from Dr. Mark van der Linden from the NRZ in Aachen. The scientific exchange with Prof. Dr. Tim J. Mitchell and Dr. Andrea Mitchell from the University in Glasgow enabled the detailed investigation of pneumolysin in stimulation of vesicle exocytosis. In cooperation with Prof. Dr. Jesús Sanz from the Universidad Miguel Hernández in Spain, the inhibitory effect of choline dendrimers on pneumococcal attachment to endothelial cells was analysed to detail. The manuscript for publication of the corresponding data is in preparation. A further intensive collaboration with Prof. Dr. Juan Hermoso from the CSIC in Madrid resulted in dissolving of the crystal structure of pneumococcal phospho glycerate kinase, which represents a new surface displayed binding factor and supported the characterization of this protein as angiostatin binding protein. In addition to an already in “Acta Crystallographica” published manuscript, a further publication focussing the biochemical and functional aspects of this protein is currently in preparation. In addition to the mentioned cooperation, the scientific exchange with all consortium partners resulted in continuing projects demonstrating the sustainability in scientific progress of this project funding.

Workpackage No. 9 (Dr. Juan A. Hermoso, Consejo Superior de Investiga¬ciones Cientificas, Spain): Structural Biology of host-pathogen interactions and pneumococcal virulence factors

This work package was focused into three main objectives: I) To understand the molecular basis of virulence and the host pathogen interactions mediated by some crucial pneumococcal surface proteins; II) To characterize the mechanisms of specific binding and hydrolysis of the pneumococcal cell wall by phage encoded endolysins in order to design more effective enzybiotics; III) To identify the molecular basis of pathogen recognition through specific components of pneumococcal cell wall by host proteins.

In order to achieve the objectives of this WP, mainly the X-ray crystallography coupled with biochemical techniques and bioinformatics was used.

Surface-exposed proteins of pathogenic bacteria are considered as potential virulence factors through their direct contribution to host-pathogen interactions. The specific interaction of bacterial proteins with host proteins often subverts the physiologic function of host-derived proteins, and therefore the bacterial proteins are considered as key players in the infectious process. The direct binding of host proteins is exploited by the pathogens for colonization, host tissue invasion, or immune evasion. Despite the relevance of bacterial surface proteins most of them are, however, still uncharacterized. Three main groups of surface proteins have been identified in S. pneumoniae: (i) lipoproteins, (ii) LPXTG consensus sequence-carrying proteins that are covalently linked via sortase to the cell wall peptidoglycan, and (iii) choline-binding proteins (CBPs). In addition to the main three groups of surface proteins, the cell envelope of pneumococci is decorated with another cluster of proteins that lack classic leader peptide and membrane anchoring motifs. These proteins are termed nonclassic surface proteins (NCSP). Among the four families, structural information was available for only one pneumococcal CBP and for some LPXTG proteins. In this WP the focus was on structural determination of CBPs, Lipoproteins and NCSP proteins (Task 9.1). Besides, structural characterization of pneumococcal bacteriophage encoded lysins (enzybiotics) (Task 9.2) and structural determination of host proteins sensing components of bacterial cell wall (Task 9.3) was also intended.

Task 9.1: During the project the Choline-binding protein F (CbpF) has been solved in complex with choline. CbpF displays a novel modular structure composed only by perfect and modified choline-binding repeats. Our results demonstrated that CbpF inhibit the activity of autolysin LytC both in vitro and in vivo and therefore provided the first example of a regulatory system to tune the activity of an autolysin. Autolysin LytC is involved in the virulence mechanism of fratricide. Pneumococci that are competent for natural genetic transformation kill and lyse non-competent sister cells that are present in the same environment. This phenomenon has been termed fratricide and requires the expression of proteinaceous toxins (LytA, LytC and CbpD) that will lyse non-competent siblings. As lysis coincides with competence, DNA released from target cells is taken up by competent attacker cells, resulting in increased efficiency of gene transfer and therefore also in antibiotic resistance. We have reported the functional and structural characterization of the main killing factor of pneumococcal fratricide LytC. Due to its unusual hook-shaped conformation, LytC is only able to hydrolyze non-crosslinked peptidoglycan chains. Therefore activation of LytC during fratricide should require prior peptide-stem hydrolysis by CbpD. These results explained the activation of LytC by CbpD in fratricide and provided the first structural insights into the critical and central function that LytC plays in pneumococcal virulence (explained in the movie: http://bit.ly/gSu7eV).

In WP9 a great effort has been done in the structural and functional characterization of pneumococcal lipoproteins. This work has been done in collaboration with WP10. Six different lipoproteins have been crystallized (SP_0198, SP_0667, SP_0659, SP_1000, SP_0629 and SP_0899) and their three-dimensional structures have been solved or are currently ongoing. Our results provide, for the first time, relevant information about implication of lipoproteins in pathogenesis and virulence. For example, structural and functional characterization of SP_0659 and SP_1000 revealed that both are critical components of the pneumococcal reactive oxygen species (ROS) defense system. Besides, they are essential for initial nasopharinx colonization and also for virulence making all of them attractive targets to selectively block the initial states of pneumococcal invasion.
In collaboration with WP8, the relationships between NCSP proteins and host proteins have been analysed. Three-dimensional structure of pneumococcal phospho glycerate kinase (PGK) has been solbed and biophysical studies have been performed. Our results confirm that this cytoplasmic protein is indeed also located on pneumococcal surface where it can bind plasminogen with nanomolar affinities. Protein overlay analysis, peptide spot array and crystal structure determination provided insights about the molecular basis of human plasminogen recognition by PGK.

Task 9.2: Endolysins have been shown to be highly efficient in killing pneumococci in vitro and can eradicate this organism from the upper respiratory tract or from the bloodstream of mice acting as new antimicrobial agents (i.e. enzybiotics). In this project, structure of endolysin Cpl-7 has been solved and its novel cell wall-binding motif unveilled.

Task 9.3: The intricate system of serum complement proteins provides resistance to infection. A pivotal step in the complement pathway is the assembly of a C3 convertase. The spleen and C3 provide resistance against blood-borne S. pneumoniae infection. We have solved the structure of SIGN-R1, a lectin that captures pneumococcal polysaccharides in the spleen and activates the complement pathway by an unusual C3 activation pathway. We found that sialylated-glycoproteins, such as C1q and Ig, bind SIGN-R1 through its classical lectin polysaccharide recognition site, while microbial pathogens can be also recognized through a, here described, novel substrate binding site in SIGN-R1. These two different recognition sites allow SIGN-R1 to simultaneously bind both immune glycoproteins and microbial components, thereby accommodating SIGN-R1’s unusual C3 activation pathway.

WP No. 10 (Prof. Dr. Sven Hammerschmidt, Universität Greifswald, Germany): Role of surface–exposed lipoproteins in host-pathogen interactions of Streptococcus pneumoniae

The contact between pathogenic bacteria and host cells is mediated via an interaction of surface-exposed adherence molecules produced by the pathogen and cellular receptors of host cells. In addition to typical cellular receptors several microbial pathogens exploit components of the extracellular matrix such as the glycoproteins fibronectin and vitronectin for adherence to host cells. The cell surface of Streptococcus pneumoniae is decorated with the three classical clusters of proteins: choline-binding proteins (CBPs), LPXTG-anchored proteins, and lipoproteins. Lipoproteins have recently received considerable attention, however, their role in pathogen-host interactions and immune evasion is not known yet. Therefore, this project focuses on the systematic assessment of the role of pneumococcal lipoproteins in colonization, virulence, and interaction with host-derived proteins. Importantly, bacterial adherence molecules and their cellular receptors represent target molecules for antimicrobials or novel intervention strategies.

Some of the pneumococcal lipoproteins, including peptide permeases, have been shown to be essential for substrate transport and bacterial fitness. A direct contribution to virulence has yet only been shown for the lipoproteins PpmA and SlrA, which belongs to the family of peptidyl-prolyl isomerases. The role of other surface-exposed lipoproteins for pneumococcal colonization, virulence and transmission has been systematically investigated in this study to gain insights into their role in pneumococcal pathogenesis and to decipher their suitability as potential vaccine candidates. The surface-exposed lipoproteins have a specific signal peptide, which includes a characteristic consensus lipobox sequence (commonly leu-3-Ser/Ala-2-Ala/Gly-1-Cys+1). Lipoproteins are anchored to the outer surface via a diacylglyceryl moiety, which is bound to the invariant cysteine of the lipobox. This modification is a prerequisite for the cleavage of the signal peptide by a lipoprotein-specific signal peptidase referred to as signal peptidase II (LSP). The enzyme that transfers the diacylglyceryl moiety to the prolipoprotein is the phosphatidylglycerol-prolipoprotein diacylglyceryltransferase (Lgt). Both enzymes are produced by all pneumococci. It is estimated that at least 10 lipoproteins are affected in pneumococcal Lgt knockout strains.

In this project the focus was on the interaction of surface-exposed lipoproteins on host cells and host-derived plasma and extracellular matrix proteins as well. An initial systematic characterization of pneumococcal lipoproteins using a proteomic approach was performed in order to identify interesting candidate lipoproteins. This was followed by an individual assessment of their role in host-pathogen interaction including cell culture and in vivo mice infection experiments. Pneumococcal lipoprotein mutants were generated in different genetic backgrounds and lipoproteins were purified after expression cloning in order to assess their role in the pathogen interaction with the host. Cell culture experiments showed the impact of the lipoproteins on adherence while phagocytosis with macrophages experiments revealed their role in immune evasion. In order to identify host proteins targeted by pneumococcal lipoproteins, the impact on interactions with host-derived plasma and extracellular matrix proteins was demonstrated using knockout strains and purified lipoproteins. Furthermore, in vivo infection experiments conducted with an acute pneumonia model or sepsis model showed the impact of the individual lipoproteins on colonization, dissemination into the lungs and transcytosis of epithelial barriers. The dissemination was monitored in real-time using the IVIS Spectrum Imaging system (Caliper Life Science) and by plating the bacterial load of organs. The interaction and intimate binding of the pathogen to host cells induces often a receptor-specific outside-inside signalling and host response. Lipoproteins are thought to be essential components of induced signal transduction cascades and toll-like receptors (TLR) dependent innate immune reactions. Therefore, the mutants and the purified lipoproteins were delivered to WP8 and WP11 and involvement of lipoproteins in TLR induced innate immunity and pro-inflammatory signalling cascades in lung cells was assessed in experimental cell culture systems. In order to determine the protein structure purified recombinant lipoproteins were provided to WP9 (Beneficiary 9) and WP15 (Beneficiary 12) for X-ray structure analysis.

WP no. 11 (Dr. Stefan Hippenstiel, Charité, Berlin, Germany): Impact of surface-exposed lipo-proteins in pneumococcal host cell activation

We examined the analysis of host cell activation by using the different mutants. Therefore, we infected cultured human lung bronchial epithelial cells (immortalized BEAS-2B cells) with the wild-type strains R6x, and mutant strains D39Δlgt, R6Δlgt, D39ΔcpsΔlgt, D39ΔcpsΔlgtΔlsp, and R800Δlgt in varying concentrations (105-106 CFU) as well as lower concentrations (data not shown) for 16 hours and measured IL-8 liberation (Fig. 1). IL-8 is a well-established, NF-κB/AP-1 driven reference cytokine, which is known to be induced in pneumococcal infection in vitro and in vivo. Furthermore, in pneumococci-infected lung epithelial cells the expression of IL-8 is connected to the p38 MAPK pathway. In addition, experiments with primary isolated human bronchial airway epithelial from bronchial brushing were conducted showing the same response.

Besides the important chemoattractant IL-8, other chemo- and cytokines as well as growth factors contributed to the inflammatory reaction in pneumococcal pneumonia. Therefore, we wondered whether Δlgt mutants were also able to induce the expression of important pro-inflammatory reference molecules. Experiments using immortalized bronchial BEAS-2B cells and primary isolated human bronchial epithelial cells showed that D39ΔcpsΔlgt bacteria induced significantly lower induction of cytokines than D39Δcps pneumococci.

However, MCP1 as well as PDGF-bb protein release is not influenced by pneumococci in these cells. In cultured human alveolar bronchial epithelial A549 cells we noted induction of IL-1β, TNFα and IL-6 by D39Δcps but not by D39ΔcpsΔlgt (data not shown). However, A549 used did not produce significant amounts of IL-8, MCP1 or PDGF-bb when exposed to pneumococci (data not shown). In line with the cytokine data, we also observed reduced release of the arachidonic acid-derived mediator prostaglandin E2 in D39ΔcpsΔlgt infected BEAS-2B cells compared to D39ΔcpsΔ exposed cells.

Next, we investigated the activation of central kinase pathways by the different mutants. However, by the time we observed no difference between the mutants with respect to the activation of p38 MAPK and ERK (Fig. 3).

The effect regarding kinase activation is somewhat unexpected. Nevertheless, we never observed a difference between the WT bacteria or the mutants used with respect to the phosphorylation of these kinases. In accordance with these results, gene silencing of TLR2, or TLR4 by siRNA reduced the phosphorylation of the kinases studied, but we observed no difference between the different bacterial strains (data not shown).

The activation of canonical NF-κB pathway depends on the degradation of IκBα allowing for the translocation of the transcription factor into the nucleus. In our study we found reduced expression of NF-κB dependent genes in D39ΔcpsΔlgt infected cells compared to infection with WT bacteria. Therefore, we tested how Δlgt deficiency in pneumococci affected the degradation of IκBα in human lung epithelial cells (Fig. 4).

Since D39ΔcpsΔlgt infected cells showed no degradation of IkBa within 4h, we tested longer infection times up to 16h. No degradation of IkBa was observed in the time frame tested (data not shown). However, after 16h both, D39ΔcpsΔlgt and D39Δcps started to induce cell damage indicated by increased LDH release in the supernatant (data not shown).

WP No. 12 (Prof. Dr. Roland Nau, University of Göttingen): Optimizing therapy of meningitis caused by pneumococci resistant to antibiotics

The aim of this project was to develop adjunctive therapies for the treatment of meningitis by clinical isolates of S. pneumoniae strains, which may be useful to combat pneumococci with a reduced sensitivity to penicillin and/or cephalosporins. In particular, we aimed at: increasing the infection resistance of the brain to pneumococci, augmenting the bactericidal activity of antibiotics, and reducing mortality and neuronal injury of pneumococcal meningitis in experimental animals.

Several approaches targeting the infection resistance of the host at different levels were pursued:

1. We attempted to modify the growth of pneumococci in order to enable the host`s immune system to better phagocytose and kill them and were unable to modify the growth of pneumococci by iprotropium bromide, an anticholinergic compound already used clinically in order to treat airways obstruction, bradykardia and hypersalivation at concentrations of acceptable toxicity in vitro. By the use of choline-bearing dendrimers synthesized by Prof. Jesus Sanz, Elche, we were able to alter the morphology of pneumococci during their replication. This increased the ability of microglial cells to phagocytose and kill them in vitro. Preliminary experiments in experimental mice, however, were negative (manuscript in preparation).

2. Different exogenous and endogenous agonists of the Toll-like receptor (TLR) system and of the innate immune receptor “nucleotide-binding oligomerization domain 2” (NOD2) were tested with respect to their ability to increase phagocytosis of pneumococci by microglial cells and to strengthen of the brain to pneumococci in experimental animals. In vitro, the following compounds were effective in stimulating phagocytosis and intracellular killing of bacteria by microglial cells: the “bacterial” TLR agonists Pam3CSK4 (a lipopeptide stimulating TLR1/2), endotoxin (agonist of TLR4), an oligonucleotide containing unmethyl-athed cytosine-guanosine (TLR9 agonist), polyinosine–polycytidylic acid [poly(I:C)], a synthetic double-stranded RNA analogue mimicking “viral” RNA (agonist of TLR3) and the endogenous compound fibronectin which is abundant in plasma and enters the cerebrospinal fluid and brain tissue during meningitis (agonist of TLR4 and of integrin receptors). Preliminary experiments demonstrated that this approach can also increase the resistance of the brain to bacterial infections in experimental animals.

3. Different antibodies (produced by Sven Hammerschmidt, Greifswald, or commercially available) directed against surface structures of pneumococci also were effetive to increase phagocytosis of pneumocci by microglial cells. In experimental rabbits, the monoclonal mouse anti-phosphorylcholine IgA antibody TEPC-15 reduced the most prominent form of neuronal injury in this model, i.e. neuronal apoptosis in the hippocampal formation (Gerber et al 2012). This antibody, however, failed to increase the killing of bacteria during ceftriaxone treatment, the standard antibiotic therapy of bacterial meningitis. We hypothesize that the antibody reduced neuronal injury by scavenging pneumococcal teichoic and lipoteichoic acids from the cerebrospinal fluid.

4. The endocannabinoid palmitoylethanolamide (PEA) also increased the phago¬cytosis of S. pneumoniae and Escherichia coli by murine microglial cells (Redlich et al 2012) in vitro. Whether this compound also increases the resistance of animals to bacterial infections, is addressed by an ongoing study.

5. Moreover, as part of this project, we reviewed data on the penetration of antibiotics into the cerebrospinal fluids in humans in the absence and presence of inflammation with the aim to identify reserve antibiotics for the antibiotic treatment of central nervous system infections caused by pneumococci with a reduced sensitivity to β-lactam antibiotics. The entry of an antiinfective into the central nervous compartments is determined by properties of the drug, the host, or both. The most important physicochemical drug property appears lipophilicity at pH 7. Molecular mass, binding to plasma proteins less strongly influenced CSF penetration, and the role of active transport was difficult to predict. In conclusion, physicochemical drug properties help to predict which antiinfective will readily enter the CNS, but because of the great heterogeneity of published human data, the prognosis of exact CSF concentrations based on physicochemical drug properties and pharmacokinetic data in the systemic circulation, however, in clinical practice is very difficult.

WP No. 13 (Dr. Eva Medina, Helmholtz Centre for Infections Research, Braunschweig, Germany): Mouse model of S. pneumoniae infection to elucidate disease mechanisms and to validate new therapeutic strategies

The goal of this work package is to gain a better understanding of the pathogenesis of pneumococcal infections using an experimental mouse model. Specifically, we have evaluated the interaction of S. pneumoniae with the different component of the host innate immune response as well as the relevance of these interactions for the infection outcome. The first objective consisted in the establishment of the experimental murine model of pneumococcal infection. Since the genetic background of the mouse strain can influence the course of pneumococcal infection, BALB/c and C57BL/6 mice were used for the characterization of the infection process after intranasal inoculation with S. pneumoniae serotype D39. A sublethal dose of S. pneumoniae was chosen in order to use the model for the identification of host defense mechanisms important for controlling the infection. This murine model was then used to compare the virulence of different clinical isolates of serotype 1: INV 104B (pneumonia), 344 03 3038 (epyema) and P1041 (meningitis) with the reference strain 342 NCTC 7465. For this purpose, C57BL/6 female mice were infected with 108 CFU of S. pneumonia by intranasal instillation and groups of 5 mice per time point were sacrificed at day 1, day 2 and day 3 after bacterial inoculation. Bacterial loads were determined in lungs, lymph nodes, liver and spleens by preparing organ homogenates and serial plating. Based on the weight loss of infected mice, control of bacterial growth in lungs and dissemination into the lymph nodes and systemic organs, the S. pneumoniae strain INV 104B (pneumonia) and P1041 (meningitis) were significantly more virulent than the reference strain 342 and the epyema strain P1041. Therefore, the behavior of the strains in the mouse model of streptococcal pneumonia reflected the severity of infection caused by these strains in humans.

This mouse model of S. pneumoniae infection was further used to evaluate the potential contribution of dendritic cells to bacterial systemic dissemination from the lower respiratory tract. For this purpose, CD11c-DTR transgenic mice, which harbor a human diphtheria toxin receptor (DTR) gene under the CD11c (Itgax) promoter, were used. Application of diphtheria toxin to CD11c-DTR transgenic mice allows the conditional ablation of dendritic cells in the intact organism. As a single dose of diphtheria toxin application did not induce a significant depletion of dendritic cells and repetitive doses are lethal for the animals, bone marrow chimeras consisting on normal BALB/c mice, which have been exposed to a single lethal dose of total body irradiation and reconstituted intravenously with 2–5 x 106 bone marrow cells from CD11c-DTR transgenic mice were generated. The chimeric mice were allowed to rest for 8 weeks before use. These chimeras can be treated repeatedly with DT without deleterious effect. Interestingly, depletion of DCs rendered CD11c DTR chimeras much more resistant to respiratory infection with S. pneumoniae as demonstrated by the enhanced bacterial clearance in the lungs and lower levels of bacterial dissemination into the draining lymph nodes and the systemic organs than non-depleted control mice. These results suggested that S. pneumoniae uses dendritic cells for systemic dissemination. Therefore interfering with the dendritic cells functional activity during S. pneumonia infection could be used as a therapeutic intervention to reduce bacterial dissemination and to improve the infection outcome.

The mouse model was also used to investigate the role of the CXCR3 receptor in the host response to S. pneumonia. CXCR3 is a G-protein-coupled receptor preferentially expressed by activated T cells, NK cells, dendritic cells, neutrophils, macrophages, epithelial and endothelial cells. CXCR3 is the unique receptor for the chemokines CXCL9, CXCL10 and CXCL11, which are responsible for the recruitment of immune cells at infection/inflammation sites. This chemokine receptor is highly expressed on effector T cells and plays an important role in T cell trafficking and function. The relevance of CXCR3 for the host defense against S. pneumonia infection was evaluated using mice deficient in the expression of CXCR3 (CXCR3-/-). The results showed that CXCR3-/- mice exhibited shorter survival times, greater bacterial loads in the lungs and higher levels of bacterial dissemination to systemic organs than CXCR3+/+ C57BL/6 animals. The aggravated infection in CXCR3-/- mice was accompanied by a pronounced increase of serum levels of the pro-inflammatory cytokines TNF-alpha and IL-6. Interestingly, B- and T cells-deficient RAG2-/- mice reconstituted with CXCR3+/+ splenocytes exhibited significantly lower amount to bacteria in the lungs and systemic organs than those reconstituted with CXCR3-/- splenocytes. These results clearly indicate that cells of the immune system mediate the beneficial effect of CXCR3 in host defence against S. pneumonia.

WP no. 14 (Dr. Manuel Espinosa, Consejo Superior de Investigaciones Científicas, Madrid, Spain): Mechanisms involved in pneumococcal multi-drug tolerance, increased drug sensitivity, and strain polymorphisms

In this workpackage the influence of checkpoint point elements, known as toxin and antitoxin systems (TAS), in antibiotic persistence (antibiotic refractoriness) and in their contribution to serotype replacement were studied, and the exploitation of TAS as an antimicrobial (AM) strategy via artificial activation of the toxin were analyzed.

Construction of tools. Pneumococcal vectors for cloning or tagging genes are scarce; three novel plasmids have been constructed. First, a promoter- and a promoter-terminator-probe vector (pAS and pAST) were constructed in which a promoterless gene encoding the green fluorescent protein was cloned after a multi-cloning site sequence (MCS). When promoter sequences are inserted, the fluorescence provided by the GFP protein can be easily recorded. Second, to regulate gene expression in S. pneumoniae the maltose inducible promoter (P) of the malMP operon (PM) followed by a MCS (plasmid pLS1ROM), was constructed. Cells growing in sucrose as energy source are induced by addition of maltose. The molecular basis of PM utilization is: expression from PM is negatively regulated by the repressor MalR, which is released from the DNA operator sequence upon growth of cells in maltose-containing media. To reduce the noise of PM utilization, MalR was provided in cis by placing the malR gene under control of a constitutive promoter. In addition, the selectable gene was the erythromycin-resistance gene. A third vector, provisionally termed pLS1pMrelE2 was designed as a positive selection vehicle for cloning pneumococcal genes that affect cell fitness. The relE toxin gene, under the control of the PM, can be cloned, maintained and amplified in the presence of sucrose. However, when the cultures were grown in maltose-containing medium, repression on gene relE was relieved, the toxin was synthesized, and cell proliferation halted. Thus, cloning any DNA segment within the relE gene will lead will lead to viable transformants in the presence of maltose, whereas those transformants in which the gene encoding the toxin remains intact (non-recombinants) will not grow as a healthy colony. The replicons used for the aforementioned vectors was that of plasmid pMV158; since it is able to establish in more than twenty bacterial species, these vectors would be useful for a variety of Firmicutes.

Drug-tolerance and increase of drug-sensitivity. A genetically homogeneous bacterial population contains physiologically distinct cells that are highly tolerant to different AMs (persisters). The persister subpopulation is genetically identical to AM-sensitive cells but has entered a dormant state in which they are tolerant to the killing activity of the AMs. In γ-Proteobacteria, especially in E. coli cells, it was shown that activation of a host-encoded toxin with mRNA endonucleases (mRNases) activity increases the level of persisters, and supports a simple model according to which activation of the mRNases inhibits global cellular translation, and thereby induces dormancy and increase AM tolerance (persistence). Sequence analyses performed on the available pneumococcal genomes have shown that the bacterium harbours at least eight TAS homologues: relBE1Spn, relBE2Spn, yefM-yoeBSpn, higAB, phd-doc, tasAB, hicAB and pezAT (ε-ζ. Of these, three TAS, namely, relBE2Spn, yefM-yoeBSpn, and pezAT have been studied and characterized as bona fide TAS, whereas relBE1Spn was shown to be not functional. The genuine and hypothetical TAS can be classified in two groups: those with mRNases activity and one with phosphotransferase activity (pezAT [ε-ζ]. The role of the toxins has been related to multi-drug tolerance (MDT) due to shut-off of protein translation leading to survival under stress conditions and to persistence (e.g. RelE toxin); however expression of ζ toxin lead to increased drug sensitivity (IDS). To unveil the molecular basis of IDS a bacterium of the same Class (Bacillus subtilis) that has only one host-encoded TAS that can be activated during exponential growth was used. The wild type or a short-living variant of ζ toxin (ζY83C) gene was ectopically integrated into the B. subtilis and then the expression cassette was moved into different genetic backgrounds. Unlike γ-Proteoteobacteria where MDT is associated with increased levels of (p)ppGpp, in Firmicutes AM tolerance is enhanced in the absence of RelA. Except β- and γ-Proteobacteria that possess two, RelA and SpoT, enzymes, other bacteria have either a single bifunctional RelA-SpoT enzyme, RelA or as Firmicutes that possesses two (S. pneumoniae, RelA y RelQ) or three (B. subtilis, RelA, RelQ and RelP) (p)ppGpp synthetases. Furthermore, many AMs triggers (p)ppGpp accumulation. Independently of growth phase and culture density, ζ toxin expression, at or near physiological concentrations, reversible induces dormancy, selects for a subpopulation of tolerant cells, but kills a fraction of AM persisters, leading to IDS. These responses are modulated by varying the (p)ppGpp pool: basal levels of (p)ppGpp (relA- cells) enhanced toxin and AM tolerance, but its absence (in relA- relP1- relQ-) potentiated IDS and AM eradication. Efforts aiming to dissect the scope of the (p)ppGpp regulon will help us to identify the pathway(s) responsible for such effect.

In vivo characterization of the role of pneumococcal TA systems. The role of pneumococcal TAS is still obscure. It is known that the majority of them (7 out of 8) are mRNases that cleave the mRNA before or during translation. By employment of inhibitors of translation and/or by the use of different lab conditions (competence, stationary phase, anoxia, starvation), the role of TAS was assessed. A broader picture emerges indicating that the regulation is more complex that initially though because a relBE- yefM-yoeB- strain appeared to be impaired in biofilm formation and genetic competence. The in vivo role of toxin activation in the serotype replacement deserved further analyses.

Polymorphisms associated to TA. A limited analysis of a dozen pneumococcal strains showed different degrees of re-arrangements in the chromosomal region encoding the relBE2Spn, but not in the yefM-yoeB Spn operon. The changes were point mutations or deletions, but all of them inactivated the relE2Spn gene (encoding toxin RelESpn). Thus, the role of relBE2Spn in the bacterial physiology was addressed. Differences in viability between the wild type and a null relE2Spn mutant strain were found only when protein synthesis was impaired. As a criterion for the permanence of this locus in a variety of clinical strains, an in-depth analysis was done by checking whether the relBE2Spn locus was conserved in around 100 pneumococcal strains, including clinical isolates and strains with known genomes. All strains, although having various types of polymorphisms at the vicinity of the TA region, contained a functional relBE2Spn locus and the type of its structure correlated with the multilocus sequence type. Functionality of this TAS was maintained even in cases where severe rearrangements around the relBE2Spn region were found. It was concluded that: i) the relBE2Spn TAS may provide additional advantages to the bacteria for colonization and/or infection; ii) Serotyping of pneumococcal isolates may not correspond to the bacterial genome organization, and iii) It would appear much more convenient to use the MLSTs hallmark rather than the serotype of clinical isolates.

Development of novel therapeutic against TAs. To battle AM persistence in S. pneumoniae among other avenues the new generation of AMs, which does not induce natural competence, the activation of TAS could be a possible target. The exploitation of TAS as an AM strategy via artificial activation of the toxin has considerable potential. As a proof of principle the ε antitoxin gene was fused to the Renilla luciferase (luc) gene (luc-ε) and the ζ toxin gene or its non-toxic or constitutively variant was fused to the gfp gene (ζ-gfp). In vivo the wild type fused render high bioluminescence resonance energy transfer (BRET) signals amenable for target–based high-throughput screening (HTS) assay, suggesting that Luc-ε2 antitoxin efficiently transfers the excited energy to the ζ-GFP acceptor and counteract the activity of the ζ-GFP toxin. It was shown that peptide can be used to modulate the TAS interaction, and the development of a toxin activator as a single entity agent is at reach.

WP No. 15 (Dr. Jésus M. Sanz, Universidad Miguel Hernández, Elche, Spain): Structural characterization of new pneumococcal targets and development of new generations of antimicrobials

Antibiotic resistance is a cause of major concern in public health. This is particularly serious when it comes to infections provoked by Streptococcus pneumoniae. Resistance of this pathogen to traditional antibiotics has skyrocketed in the last years, reaching levels up to 50% in some developed countries. Moreover, the general lack of development of new antibiotic families in the last 50 years, together with the relatively modest results on vaccination (mainly due to the insufficient serotype coverage that limits its success to certain developed countries) draws an upsetting picture for the next future. Therefore, to tackle this problem, innovative approaches are needed.

Our workpackage was focused into the identification and structural characterization of novel targets common to all pneumococcal serotypes against which new antimicrobials would be designed and tested. Our main efforts in this sense have been aimed to the so-called choline-binding proteins (CBP's), a family of around 15 polypeptides that are involved in many essential, physiological processes such as cell separation upon division, cell-wall rupture and subsequent release of virulence factors, and cellular adhesion to the host. To perform their biological functions, all CBPs need to be attached to choline molecules that decorate the external surface of the pneumococcal cell wall. It is then evident that the specific inhibition of the attachment of the CBPs to the bacterial surface would generate catastrophic consequences to the pathogen. Moreover, targeting 15 proteins simultaneously with the same inhibitor should greatly hamper the appearance of antimicrobial resistances. Finally, since the CBP's are common to all serotypes, any possible inhibitors would be of widespread therapeutic use.

Choline cannot be therapeutically used as a competitor ligand to "sequester" the CBP's since too high concentrations are needed to do so. Therefore, we searched for choline analogs with improved properties.

First, some commercial compounds similar to choline were tested by their ability to bind to CBP's using a fluorescence spectroscopic technique. Simultaneously, we synthetized a set ("library") of tens of compounds of similar chemical structures and little differences between them. We selected the best molecules and these were assayed for their effect on pneumococcal cultures.

Next step was trying to reduce the doses even more. To do so, we employed a kind of nanoparticles called dendrimers. A dendrimer can hold several copies of the same molecule. The resulting complex increases dramatically its affinity to its target. We synthetized dendrimers functionalized with choline or choline analogs, and found that their antibacterial effects were augmented between 1000-10000-fold, meaning that the necessary doses are concomitantly lower when compared to the free molecule. All these results point to our selected choline analogs, either free or attached to dendrimers, as molecules with a potential to become novel antibiotics against pneumococcus.

It should also be mentioned that we have successfully employed magnetic nanoparticles functionalized with diethylaminoethanol as specific, strong ligands of CBP's, to a point that these proteins could be purified from a bacterial extract upon addition of these nanoparticles and the simple application of a magnet. The magnetic nanoparticles may serve in the future as agents to remove the CBP's from the bacterial surface leading to antimicrobial effects.
On the other hand, we have determined the structure of a small part of one of the most important CBP's, the LytA autolysin. We have experimentally determined that this part of the protein is the minimal structure capable of recognizing choline. This information will be used in the future for the construction of peptidic nanoparticles that "emulate" the role of CBPs and that may interfere with the role of the latter.

Finally we have carried out a structural analysis of another kind of surface protein, the extracellular domain of the StkP Ser/Thr protein kinase that is involved in the regulation of a high number of genes in the cell. We have elaborated a three-dimensional model and found that small peptidoglycan fragments and beta-lactam antibiotics are able to specifically bind to this protein. This identifies this polypeptide as a suitable target for the development of new beta-lactam antibiotics that will be explored in the future.

WP No. 16 (Gennaro de Libero, Uni Basel): Prevention of pneumococcal disease with a novel polysaccharide-glycolipid conjugate pneumococcal vaccine

Our goals were: to generate a S. pneumoniae polysaccharide (from a replacing serotype causing disease) covalently linked to synthetic α-galactosylceramide (αGalCer); to test the capacity of this new glycoconjugate to induce B cell activation, switch and affinity maturation of specific antibodies; to test the protective activity of the glycoconjugates in mice during infection with S. pneumoniae of the vaccine serotype in comparison with the pure carbohydrate vaccine.

Results obtained from direct cooperation with other CAREPNEUMO partners. We have discussed with several partners of the Consortium the best strategy to test the glycoconjugate in vivo and to perform challenges with S. pneumoniae after immunization.

Achievement of milestones. In vivo activity of glycoconjugate (Milestone M363).

Data or results which contribute to deliverables. The CPS of type 4 S. pneumoniae coupled to synthetic αGalCer was synthesized by Dr. P. Stallforth and Dr. D. Rathwell, two collaborating chemists, and verified for activity using in vitro and in vivo assays.

The CPS4 coupled to αGalCer glycoconjugate was used to immunize wild-type C57BL/6 mice. Three immunizations were performed with intervals of 14 days. The mice showed high titers of anti-polysaccharide antibodies up to 3 months after the last immunization. This argues in favour of a long-lasting antibody response by B cells. Analysis of polysaccharide-specific antibodies showed the presence of IgG isotypes. This isotype is not frequent when B cells generate a T cell-independent response, whereas they compare in the presence of helper T cells which secrete helper cytokines and provide cognate help to specific B cells. Therefore, these findings strongly support the conclusion that our novel vaccination approach exploits the function of helper cells of NKT cells activated by the αGalCer moiety present in the new vaccine.

Mice were also challenged with live S. pneumoniae 6 months after the last immunization and showed full protection. These findings support the conclusion that the used immunization protocol may induce memory B cells, and the generation of circulating long-lasting antibodies capable of providing full protection.

In additional experiments we also generated monoclonal antibodies from immunized mice.
Several hybridomas expressing CPS4-specific antibodies were established. The hybridomas expressed all IgG subclasses, with the exception of IgG2b. Sequence analysis of the heavy chains of these antibody genes revealed the presence of somatic mutations also in IgM-positive hybridomas thus showing the occurrence of affinity maturation of polysaccharide-specific antibodies.

In another set of experiments, the purified monoclonal antibodies were used to confer passive transfer protection. They also revealed protective activity, this confirming their important role during infection. Finally, in vitro experiments revealed that the monoclonal antibodies have opsonising activity, which facilitated killing of S. pneumoniae.

The vaccine candidate was also used to immunize wild-type mice and test their protection. Vaccinated mice with the glycoconjugate were protected from infection, as shown by their survival upon bacterial challenge. On the contrary, mice vaccinated only with the polysaccharide succumbed upon infection. Protection was also tested in animals vaccinated 6 months earlier. Also this group of animals showed full protection, thus indicating the generation of memory B cells, capable of releasing protective antibodies at later time points. The latter findings showed that the immunization with the glycoconjugate vaccine provided long-term protection.

In a latter set of experiments vaccinated animals were challenged in a meningitis model. Although in these experiments the number of boosts and the infection model were not optimal, a clear resistance of vaccinated animals was observed.

Workpackage No. 17: (Dr. Micahel Tal, Protea Vaccine Technologies Ltd., Israel): Development of protein-based universal vaccine against S. pneumoniae

To identify proteins with vaccine potential, a pneumococcal cell wall protein fraction was screened by 2D-immunoblotting using sera obtained longitudinally from infants attending day-care centers and sera from healthy adult volunteers. Proteins that exhibited age-dependent immunogenicity were identified. Several such proteins were cloned, expressed in E. coli, and purified, and their vaccine potential was tested in the mouse S. pneumoniae challenge system. All the proteins from this group elicited at least partially protective immune responses against S. pneumoniae challenge. Moreover, these proteins and the antibodies elicited against them demonstrated the ability to interfere in bacterial adhesion in vitro and colonization in vivo. A protein that did not demonstrate age-dependent immunogenicity typically did not confer a protective immune response.

To reduce the physical protein load of the potential candidate vaccine and eliminate homology to human sequences, we identified specific fragments and peptides within these proteins that may provide immune protection in the mouse challenge model. Two approaches were used to identify such peptides: i) A peptide array designed from the regions with no homology to human sequences was screened with sera obtained from infants and healthy adults. ii) Five chimeric genes were designed and synthesized based on the peptides derived from the above mentioned proteins (known as "chimeric fusion protein").

Here we report results of the peptide array screen. To identify novel targets for a protein-based vaccine against Streptococcus pneumoniae, we used a high-throughput screening strategy. Human serum samples from 87 donors of different age groups were used to scan 386 peptides derived from 22 S. pneumoniae proteins, which were spotted and blotted on microarray slides. Peptide arrays were scanned for intensity and sorted by analysis through multiple screens and filters. We were able to select 13 peptides with age-dependent immunogenicity. Three of the 13 were from the carbamoyl phosphate synthase large subunit NP_35739.1 protein, and 5/13 were from the glutamate dehydrogenase NP_345769.1 protein

These 13 peptides epitopes, which have age-dependent immunogenicity, could be potentially used for a future common-group protein-based vaccine.

Five fusion chimeric genes were synthesized, expressed in E. coli and purified. The vaccine potential of these chimeric fusion proteins was tested. The mice were immunized with the purified protein emulsified with adjuvant. Vaccine potential was tested in two mice models: BALB/c and CBA/Nxid. Two types of end points were used, either the bacterial load in the nasopharynx and the lungs of immunized mice was determined 48 hours following a sublethal intranasal challenge with S. pneumoniae, or survival rates were determined for immunized mice challenged intranasally with a lethal S. pneumoniae dose. Moreover, two S. pneumoniae strains were used in this study, the serotype 3 strain WU2 and serotype 4 strain TIGR 4. The set of biological data produced with the five chimeric fusion proteins can be used to rank the efficacy of these proteins, as follows: P722> P721>P728>P727> P729.

Potential Impact:
WP01: Changes in pneumococcal disease after introduction of conjugate vaccine in Germany (RWTH Aachen, Germany)

Streptococcus pneumoniae, commonly known as the pneumococcus, is the leading pathogen causing severe community-acquired pneumonia (CAP) and responsible for >1.5 million death worldwide and kills more people in the United States and Europe than other vaccine-preventable disease. Since 2000, pneumococcal conjugate vaccines (PCVs) have been available, enabling the vaccination of children under 2 years of age.

In August 2006 a general recommendation for vaccination of young children (≤2 years of age) with the 7-valent pneumococcal conjugate vaccine (7vPCV) was issued in Germany. The introduction of 7vPCV in Germany has resulted in a marked decrease in vaccine serotypes. The amount of isolates with non-vaccine serotypes, however, has increased. But we have also witnessed an increased percentage of reported isolates that were actually sent to the Reference Center for analysis. Calculations of serotype specific incidence rates showed a decrease in vaccine serotypes for children under 2 years of age, but a significant increase in non-vaccine serotype incidence could only be determined starting from 2009. Therefore, the increase in isolates from non-vaccine serotypes sent to the Reference Center seems to be at least in part due to increased awareness.

MLST analysis showed that for most serotypes the clonal distribution has remained unchanged since the introduction of childhood vaccination. Only for Serotype 19A the clonal distribution has changed. This change continued in the years 2008 to 2011, and has been followed closely. An expansion of the existing clones could be observed, accompanied by a loss of some clones and the appearance of others. Multi drug Resistant (MDR) clones appeared in Germany, but their numbers have remained very low. No cases of serotype switching have been found.

WP02: Monitoring of S. pneumoniae isolates from invasive and non-invasive diseases in children after introduction of the vaccine in Greece (National and Kapodistrian University of Athens, Greece)

Epidemiological data on the serotypes and the development of antibiotic resistance in different geographical areas is an essential prerequisite for vaccine development against S. pneumoniae. In this respect, resistance rates, serotype distribution and clonal relatedness were investigated in a large collection of S. pneumoniae isolates, in an area where clonal spread of specific multi-resistant serotypes had been documented in the past.

Analysis of the susceptibility, serotyping and MLST typing data indicated that implementation of the 7-valent vaccine in Greece seems to have achieved its main targets: (a) decrease of the isolation rate of strains from invasive disease, (b) decrease in the spread of the multi-resistant serotypes 19F, 23F and 6B, (c) replacement of these serotypes by the less resistant 19A, 6A, 23A and 23B, (d) slow-down in the increase of penicillin resistance that had been detected in the previous years in Greece.

Nevertheless, analysis of the data also indicated that the MLS predominant phenotype changed from the less resistant M to the more resistant CR-MLSB, and that the isolates belonging to the emerging serotypes 19A, 6A, 23A and 23B were also clonally related (a limited number of MLST types accounted for the majority of isolates of each individual serotype), as had been those of the initially persistent serotype.

These results, following their publication, will be valuable in designing the coverage of the next generation polyvalent vaccine.

WP03: Population structure, resistance mechanisms and virulence factors of S. pneumoniae in Poland (National Medicines Institute, Poland)

In the course of this art of the project, a country-wide, representative collection of S. pneumoniae isolates was assembled and analyzed by phenotypic (serotyping, susceptibility testing) and molecular methods (genotyping using MLST, molecular analyses of resistance and virulence determinants as well as mobile genetic elements). Data thus obtained are of importance in several aspects:

• The knowledge of frequencies of resistance to particular drug classes is of big importance for formulation of appropriate therapeutic recommendations for empirical treatment of pneumococcal diseases.

• The knowledge of serotype distributions, especially among the risk-groups (such as children, elderly) provides a rational for the use of vaccines of particular formulations.

• The identification, with the use of molecular typing methods, of major clones beyond the current spread of antimicrobial resistance in the country and responsible for a significant proportion of pneumococcal invasive disease allows for tracing of these clones both in the country and at the above-country level.

• Studies on the evolution of resistance and virulence determinants, together with the data on clonality contribute to a better understanding of the population biology of pneumococcus as recombining species.

WP04: Population dynamics of S. pneumoniae under vaccine pressure in Portugal (Instituto Medicina Molecular, Portugal)

Streptococcus pneumoniae, commonly known as the pneumococcus, is the leading pathogen among those vaccine-preventable. This pathogen has been known for capacity to rapidly adapt to change. It capacity to transform, i.e. to capture and recombine exogenous DNA with its own chromosome, has been implicated in this species enormous plasticity. A prime example is the development of antibiotic resistance that has resulted from acquisition of mobile genetic elements and the remodeling of the penicillin-bindings proteins using DNA from other streptococcal species. The advent of conjugate polysaccharide-based vaccines offered a new tool with which to effectively prevent pneumococcal disease. However, in spite of their current success at eliminating vaccine serotypes and reducing antibiotic resistance, there are fears that these may be short-lived due to the limited number of targeted serotypes and capacity to adapt of pneumococci. Thus, a better understanding of the dynamics, diversity and response of the pneumococcal population to current interventions is needed in order to make the best use of the tools available today and to guide the design of novel vaccines and approaches that can allow us to sustain the benefits of reduced pneumococcal disease we are enjoying today.

In this project we have characterized the changes in the pneumococcal population causing invasive disease in a period of intense use of the pneumococcal seven-valent conjugate vaccine (PCV7). We showed that while the serotypes covered by the vaccine have diminished as causes of invasive infections, these have been counterbalanced by an increase in serotypes not covered by PCV7. Some of these serotypes are included in the last-generation conjugate vaccines and we could estimate the benefits of vaccination with the newer vaccines. We have highlighted differences in the association of particular serotypes with age that suggest the pneumococci are not identical in their disease causing capacity. The mechanisms behind these preferences will be investigated in the future. We have documented the changes and diversity of some protein targets that are being considered has potential vaccine components, indicating that these offer limited additional benefits over currently available vaccines. Finally we have identified conditions leading to genetic structuring within the pneumococcal population that can have important implications in the adaptation of the bacteria to vaccination.

WP05: Antimicrobial resistance of Streptococcus pneumoniae from otitis media of pediatric patients (Hospital de Pediatría “Prof. Dr. Juan P. Garrahan”, Argentine)

This project contributes to know the relative prevalence of pneumococcal serotypes from acute otitis media (AOM) and invasive infections (INV) of Argentinean children. Serotype 19A, present only in the 13-valent conjugate vaccine is more prevalent in AOM than in INV. The knowledge of prevalent serotypes is very important for choosing the best vaccine formulation to cover both INV and AOM in Argentinean children.

Knowledge of antimicrobial resistance trends, contributes to establish the best therapeutic schemes for treating both AOM and INV.

Knowledge of the presence of different erythromycin-resistant clones let us understand the way in which such resistance spreads in Argentina.

WP06: Study of the prevalence of Streptococcus pneumoniae serotypes in North Indian Communities (Post Graduate Institute of Medical Education and Research, India)

Nasopharyngeal colonization provides a convenient niche for Streptococcus pneumoniae to cause lower respiratory tract infections. The knowledge of current serotypes circulating in a community and their antimicrobial susceptibility is crucial for effective therapeutic strategies; however, there is lack of information in this regard from North India. Hence, for the first time a cross-sectional surveillance was done in children from Chandigarh, North Indian to evaluate the carriage, prevalent serotypes, their antimicrobial susceptibility and adherence and invasion of S. pneumonia in epithelial cell line. Nasopharyngeal swabs were collected from 575 healthy children aged <5 years residing in the urban, rural and slum areas of Chandigarh, North India. S. pneumoniae isolates were identified by gram staining, optochin sensitivity and bile solubility test and characterized by multiplex PCR based on the cps gene cluster. Further, S. pneumoniae were screened for antimicrobial susceptibility by Kirby Bauer disc diffusion method using CLSI guidelines. Adherence and invasion of the most prevalent strains of S. pneumoniae with A549 pulmonary epithelial cell line was also looked for. Significant nasopharyngeal carriage rate (49.6%) of S. pneumoniae was observed. Serotypes 6A/B/C, 11A/D, 15B/C, 10F/C, 34, 10A, 15A/F, 23F, 14, 7C/B, etc. were found prevalent in descending order. Six serotypes: 6A/B/C, 11A/D, 15B/C, 10F/C, 34, 10A accounted for 60% of S. pneumoniae isolates. Serotypes 6A/B/C, 15B/C, 10A, 10F/C and 7C/B were common among urban, rural and slum regions however, their frequency varied. Maximum prevalence and serotype variability was found from urban-slum region. All S. pneumoniae isolates were sensitive for amoxicillin, cefotaxime and vancomycin with resistance maximum towards co-trimoxazole (97.2%) followed by tetracycline (28.4%), erythromycin (6.4%) and chloramphenicol (2.8%). About 53% isolates showed complete susceptibility towards penicillin. Only 5.38% S. pneumoniae isolates tested emerged as multidrug resistant. Low adherence and negligible invasion was also seen for the prevalent serotypes in our region. This study highlights the considerable carriage rate of S. pneumoniae showing increasing antibiotic resistance as well as the presence of non-vaccine serotypes in the community which require continuous monitoring for the management of pneumococcal diseases.

WP07: Genomic analysis and virulence of S. pneumoniae (University of Glasgow, UK)

The pneumococcus or Streptococcus pneumoniae is known as a major pathogen of man but also detected in other mammalian species. S. pneumoniae is naturally transformable and displays considerable variation in both genome content and sequence between strains. This variation is linked to the ability of the pneumococcus to cause a number of different diseases in man, ranging from pneumonia, sepsis, and meningitis through otitis media (middle ear infection), conjunctivitis and sinusitis. The high degree of genetic variation contributes in two major ways to the continuing burden of pneumococcal disease in Europe and worldwide. Both the increase and diversity in antibiotic resistance observed in pneumococcal isolates, and its evasion of current polysaccharide-based vaccination strategies by switching its polysaccharide capsule type are achieved through its innate ability to vary its genome.

More detailed knowledge of pneumococcal essential ‘core’ genes and the ‘accessory’ genome of variable gene content and their relative contribution to the different types of disease are essential in understanding the virulence of this organism. The identification of the roles of key virulence genes in the different stages of pneumococcal disease will aid in the battle against pneumococcal disease in a number of relevant areas. Genetic testing of isolates will aid diagnostically by providing prognostic information as well as predicting therapeutic agents.

Additionally genome sequence information can be used to predict future candidates for new vaccine approaches to replace or improve those currently available. Wider identification of the gene pool available to the pneumococcus as it faces selective pressures from both the current vaccine and novel vaccine strategies in development may also aid in predicting the efficacy of new vaccines proposed. The data generated in this project has elucidated further the roles of key virulence factors, such as the protein pneumolysin. Even more genes and regions have been discovered and our data predicts their roles in the pathogenesis of pneumococcal disease, their precise contribution remains to be explored in future studies.
Data have been disseminated through invited seminars, presentations at scientific conferences and outreach activities. All genomic data are, or will be, available in public databases.

WP08: Interaction of S. pneumoniae with endothelial cells (Helmholtz-Zentrum für Infektionsforschung GmbH, Germany)

The highest potential impact of this study was obtained with regard to scientific understanding, clinical diagnosis and medical therapy. Our study demonstrated that the most relevant pulmonary pathogen S. pneumoniae induces the release of pro-inflammatory and pro-coagulative components directly contributing to pathophysiological processes leading to fatal tissue injury during course of infection. The obtained data of pneumococcal induced vesicle exocytosis enrich the understanding of clinical symptoms and disease progression of invasive and systemic pneumococcal infections with respect to thrombosis, inflammatory collateral tissue destruction, and systemic organ failure. A further important gain in experience was obtained from the employment of choline dendrimers as novel inhibitory substances within the infection analyses. The observed inhibition of pneumococcal adherence to endothelial cells by choline dendrimers offers new starting points for the development of an effective interference therapy with the aim to inhibit the invasive dissemination of pneumococci. The invasive pneumococcal disease progression is predominantly observed in elderly and immune compromised patients. The prognosis of survival for these patients is significantly better, if an adequate medical therapy is started in early stage of infection, which requires a reliable and fast diagnostic. In addition to the descriptive elucidation of the endothelial response to pneumococcal pathogenicity factors, the establishment of ELISA-based biochemical quantification of human factors secreted in response to pneumococcal infection offer new diagnostic possibilities and generate the basis of further detailed investigation with respect to early and discriminative diagnosis of invasive bacterial infection processes. Therefore, the data generated in this project have the potential impact to save the life of people suffering from invasive pneumococcal infection.

WP09: Structural biology of host-pathogen interactions and pneumococcal virulence factors (Consejo Superior de Investigaciones Científicas, Spain)

Antibiotic consumption has led to a gradually increasing pneumococcal resistance against penicillin and cephalosporins, and to the emergence of multi-drug resistant strains. Prerequisites for developing new control strategies are (i) the detailed understanding of the interaction between pathogen and the host cells (ii) In-deep understanding of antibiotics resistance mechanisms and (iii) to find new therapeutic weapons against bacterial pathogens. The WP9 focused on the first issue and the main results were:

(a) The study of pneumococcal surface proteins (CbpF, LytC, LytB). These objectives have been mainly achieved: we have reported in high impact factor journals the structural characterization of CbpF

(b) The structural analysis of the recently discovered human proteins involved in recognition of pneumococcus (SIGN-R1). We have determined the 3D structure of SIGN-R1 and also of its complexes with different carbohydrates mimicking bacterial cell wall, we are currently preparing the manuscript for a high impact factor journal.

(c) The structural determination of novel endolysins (Cpl-7, Gp61). We have solved both structures Cpl-7 and Gp61. In the case of Cpl-7 we are now preparing the manuscript. In the case of GP61 we intend to explore more in detail the structural determinants of the translocation and hydrolysis mechanisms.

(d) We have opened new research topics such as functional characterization of lipoproteins and nonclassic surface proteins providing, for the first time, both structural and functional information indicating their interaction with the host cells. We are currently preparing two manuscripts in high impact factor journals and structural analysis is ongoing for the rest of proteins.

These results have unveiled new important pneumococcal virulence factors and will provide the basic knowledge needed to identify novel therapeutic targets as well as vaccine candidates.

Dissemination about results of the research performed in the WP9 is fully available in the web page of the WP responsible (http://www.xtal.iqfr.csic.es/grupo/xjuan). Popular conferences (see for ex. http://semanaciencia.uniovi2masd.com/vienes-18-de-noviembre-de-2011/conferencia-la-estructura-de-las-proteinas.php). Spreading in mass media is summarized in http://www.xtal.iqfr.csic.es/grupo/xjuan/CHBP_archivos/press-notes/press-notes.html

WP10: Role of surface-exposed lipoproteins in host-pathogen interactions of Strepto-coccus pneumoniae (Ernst Moritz Arndt Universität Greifswald, Germany)

Streptococcus pneumoniae, commonly known as the pneumococcus, is the leading pathogen causing severe community-acquired pneumonia (CAP) and responsible for >1.5 million death worldwide and kills more people in the United States and Europe than other vaccine-preventable disease. Traditionally, pneumococcal diseases are treated with antibiotics and prevented with polysaccharide-based vaccines. However, due to the dramatic increase in antibiotic resistance and limitations of the current available vaccines, the burden of diseases remains high. Thus, combating pneumococcal transmission and infections has emphasized the need for a new generation of protein-based vaccines. All currently available PCVs still have limitations as they do not cover all pneumococcal serotypes causing pneumonia and severe invasive diseases. Therefore, the development of protein-based vaccines with expanded coverage and immunogenicity will be essential to combat pneumococcal transmission and diseases, especially in developing countries. Pneumococcal virulence requires adherence, colonization, invasion, fitness and immune evasion during the host infection process. Therefore, one alternative avenue to fight against the pneumococcal pathogenicity is offered by employing anti-virulence strategies. Specifically, the disruption of pneumococcal adhesin-host protein interactions by anti-virulence drugs, represented by small organic compounds, and/or a more effective (multi-)adhesin-based vaccine with expanded coverage, may serve as an alternative strategy in the battle against the pneumococcal infections. Lipoproteins represent a major proportion of surface-exposed and highly conserved proteins contributing to the pathogenesis of pneumococci. The results of our projects identified biological activities of several extracellular lipoproteins and their role in maintenance of bacterial fitness. Due to their important role and their conservation among clinical isolates and serotypes the lipoproteins represent promising candidates for a protein-based vaccine, if they are immunogenic and confer protection against pneumococcal infections which has to be elucidated in the future.

WP11: Impact of surface-exposed lipo-proteins in pneumococcal host cell activation (Charité, Berlin, Germany, for Ernst Moritz Arndt University Greifswald)
(in close cooperation with WP 10)

WP12: Optimizing therapy of meningitis caused by pneumococci resistant to antibiotics (University Hospital Göttingen, Germany, for Helmholtz Centre for Infections Research, Braunschweig, Germany)

Multi-drug resistant Streptococcus pneumoniae are a major public health problem worldwide. The compounds identified to increase the infection resistance of the host organism against S. pneumoniae and other bacteria in this project do not belong to the classical antibacterials. They are of interest for the treatment of S. pneumoniae infections in humans. This is particularly true for the monoclonal mouse IgA anti-phosphorylcholine antibody TEPC-15 which increased phagocytosis of pneumococci by microglial cells and decreased neuronal injury in experimental meningitis. A humanized form of this antibody could serve as a biological drug for a clinical study on pneumococcal meningitis. The other compounds will first have to prove their in-vivo efficacy in experimental models, before studies in humans can be considered.

WP13: Mouse model of S. pneumoniae infection to elucidate disease mechanisms and to validate new therapeutic strategies (Helmholtz-Zentrum für Infektionsforschung GmbH, Germany)

• Identification of host immune mechanisms involved in the control of S. pneumonia infection will uncover novel components of the host immune system that can be targeted by novel therapeutic strategies to strengthen the host defense against this pathogen during the infection process.

• Identification of bacterial factors required for the establishment and progression of pneumococcal infection can represent promising candidates to design compounds targeting disease-causing virulence factors.

WP14: Mechanisms involved in pneumococcal multi-drug tolerance, increased drug sensitivity, and strain polymorphisms (Consejo Superior de Investigaciones Científicas, Spain)

Exploiting TAS to promote bacterial eradication. Bacterial persisters, the survivors from AM exposure, pose a significant challenge for our control of diseases. The ectopic expression of the MazF or RelE mRNase toxins causes an increase in the number of persisters. Here, survival following treatment with one AM did not correlate with survival against another. Conversely, the activation of ζ (PezTSpn) toxin enhances IDS. Here, persisters, which are heterogeneous in nature, implicate multiple pathways including energy production, the stringent response, activation of TASs, sugar and lipids metabolism, energy metabolism, membrane and cell wall biosynthesis pointing to redundant pathways for persister formation. Expression of ζ toxin partially reverses AM persisters to the AM sensitive kin potentiating the AM action. The ζ (PezTSpn) phosphotransferase induces dormancy but contributes to eradication of persisters, suggesting that the switch from growing to dormancy is not sufficient for MDT. A decreased response (relA-) to stress increases toxin and AM tolerance, but absence of response (ppGpp0) leads to toxin sensitization and AM eradication. A small compound or peptide disruptor might activate the ζ (PezTSpn) toxin that blocks cell proliferation and the boostering of the AM action contributes to bacterial eradication.

In vivo role of pneumococcal TASs. The role of pneumococcal TAS is still obscure. By employment of inhibitors of translation and/or by the use of different experimental conditions (competence development, stationary phase, biofilm formation, anoxia and starvation) the role of TAS was assessed. A broader picture emerges indicating that the regulation is more complex that initially though because a relB- yoeB- strain appeared to be impaired in biofilm formation and genetic competence. The in vivo role of a phosphotransferase toxin as an anti-persisters strategy and mRNases toxin activation as anti serotype replacement deserved a further analyses.

The TASs have been implicated in a multiple processes and contributes to either bacterial persistence MDT) or to bacterial eradication (IDS). Given the physiological complexity of each bacterial cell, it is plausible that persistence and anti-persistence may be the result of fluctuations in different processes, suggesting distinct mechanisms for evading the lethal effects of an AM as well as for eradicating persists.

WP15: Structural characterization of new pneumococcal targets and development of new generations of antimicrobials (University Miguel Hernández, Spain)

The main outcome of our Workpackage 15 is the identification of the choline-binding proteins from Streptococcus pneumoniae as targets of newly designed antibiotics, and the generation of inhibitors of these proteins that arrest pneumococcal growth in vitro and that may constitute the starting point for the development of new antimicrobial drugs. More importantly, we have derived a set of rules that allow the rational or semi-rational design of further generations of drugs of this kind, and a nanobiotechnological method to reduce the doses by the use of bioactive nanoparticles. This represents a novel approach for the development of antipneumococcal antibiotics circumventing the current problem of antibiotic resistance and that may have a beneficial influence in the Society, especially the infants and the elder, who are the main targets of pneumococcal diseases. Moreover, a new family of antimicrobials would alleviate the use of traditional antibiotics, thus contributing to the decrease in the appearance of resistant strains. Some of the molecules discovered in our studies display an antimicrobial effect either free or linked to nanoparticles, and have been sufficiently characterized to allow their entrance in the development pipeline of any pharmaceutical company.

WP16: Prevention of pneumococcal disease with a novel polysaccharide-glycolipid conjugate pneumococcal vaccine (University Hospital Basel, Switzerland)

Despite the recent development of glycoconjugate vaccines, pneumococcal diseases remain a major threat to public health worldwide with over one million casualties every year. Protection against the disease is provided by neutralization of virulence factors or opsonizing antibodies. We have provided the proof of concept of a simple, easy to produce, cheap and highly efficient anti-pneumococcal vaccine, that is universally applicable as not limited by polymorphic MHC molecules.

If this vaccine will confirm its efficacy also in human beings, it will allow great advance in our fight against pneumococcal infections. We have initiated a series of dissemination activities by presenting the scientific data in international meetings, by interacting with Pharma companies interested in our vaccine and by discussing with relevant clinical centers the possibility of human clinical studies. A patent application is being submitted to protect our findings.

We plan to continue our studies by investigating the use of the vaccine in human beings. For this purpose, a series of contacts have been initiated with different companies and we are also preparing a new international grant application to financially support the planned human studies

WP17: Development of protein-based universal vaccine against S. pneumoniae (Protea Vaccine Technologies Ltd., Israel)

Novel therapeutic strategies are necessary to counter the increasing prevalence of antibiotic-resistant pneumococci and the limitations of currently available vaccines against S. pneumoniae. A pertinent advantage of proposed protein-based vaccines is protection from carriage and diseases caused by serotypes not represented in the existing polysaccharide-conjugate-based vaccines. It is believed that in such a protein vaccine, one will have to use several proteins to prevent the development of escape mutants, which will result in high protein content. To reduce the protein load and to eliminate homology to human sequences, we have selected peptides from the previously identified protein set with age-dependent antigenicity (which correlates negatively with natural morbidity) that lack homology to human proteins. Moreover, in order to achieve common-group vaccine, we selected and included peptides of conserved sequences of S. pneumoniae. We then tested their antigenicity in humans and their age-dependent antigenicity. Out of the screen we identified 13 short peptides (~15 amino acids each), which can be fused, expressed and included in a future vaccine formulation.

In addition we selected short protein peptide sequences by the same methodology of conservation to S. pneumoniae sequences and lacking homology to human sequences to produce five chimeric proteins, three of which elicited protective immune responses in the different mouse models.

This new approach will enable the construction of future protein vaccine to S. pneumoniae that: i. will minimize the amount of protein needed for vaccination; ii. will reduce the risk of autoimmunity; iii. will include immunogenic epitopes from various proteins with various distinct biological functions and therefore will minimize the chance of the development of escape mutants; and iv. may protect against many different strains of S. pneumoniae, being derived from conserved sequences of S. pneumonia.
final1-223111-1041621-carepneumo-logo.jpg