Mature Epitope Density - A strategy for target selection based on immunoinformatics and exported prokaryotic proteins
Current immunological bioinformatic approaches focus on the prediction of allele-specific epitopes capable of triggering immunogenic activity. The prediction of major histocompatibility complex (MHC) class I epitopes is well studied, and various software solutions exist for this purpose. However, currently available tools do not account for the concentration of epitope products in the mature protein product and its relation to the reliability of target selection.
We developed a computational strategy based on measuring the epitope's concentration in the mature protein, called Mature Epitope Density (MED). Our method, though simple, is capable of identifying promising vaccine targets. Our online software implementation provides a computationally light and reliable analysis of bacterial exoproteins and their potential for vaccines or diagnosis projects against pathogenic organisms. We evaluated our computational approach by using the Mycobacterium tuberculosis (Mtb) H37Rv exoproteome as a gold standard model. A literature search was carried out on 60 out of 553 Mtb's predicted exoproteins, looking for previous experimental evidence concerning their possible antigenicity. Half of the 60 proteins were classified as highest scored by the MED statistic, while the other half were classified as lowest scored. Among the lowest scored proteins, ~13% were confirmed as not related to antigenicity or not contributing to the bacterial pathogenicity, and 70% of the highest scored proteins were confirmed as related. There was no experimental evidence of antigenic or pathogenic contributions for three of the highest MED-scored Mtb proteins. Hence, these three proteins could represent novel putative vaccine and drug targets for Mtb. A web version of MED is publicly available online at http://med.mmci.uni-saarland.de/.
The software presented here offers a practical and accurate method to identify potential vaccine and diagnosis candidates against pathogenic bacteria by "reading" results from well-established reverse vaccinology software in a novel way, considering the epitope's concentration in the mature portion of the protein.
KeywordsMajor Histocompatibility Complex Major Histocompatibility Complex Class Binary Classifier Epitope Prediction Major Histocompatibility Complex Allele
Tuberculosis (TB) has been one of the major causes of morbidity and mortality worldwide for centuries, and control of the spread of Mycobacterium tuberculosis (Mtb) infection remains a public health priority . More than 9 million new cases of TB in humans arise every year, resulting in nearly 2 million deaths worldwide . Bacille Calmette-Guérin (BCG), the current vaccine for the treatment of TB, has its limitations; although it is protective against severe childhood TB, it does not satisfactorily prevent the pulmonary disease in adults . Effective prophylactic and therapeutic immunization is a key strategy for global epidemic control . Novel TB vaccine candidates include BCG or recombinant BCG (rBCG) strains, which are used in heterologous prime-boost strategies as a prime vaccination . Booster vaccinations can include viral vectors that express immunodominant Mtb antigens or fusion proteins of these antigens, combined with adjuvanticity to ensure immunogenicity . Many Mtb antigens, including Ag85A, Ag85B, TB10.4 and ESAT-6, have been tested as vaccine candidates; however, these have not been shown to be successful at treating TB . Consequently, discovering new antigens continues to be a crucial factor for the successful development of vaccines against TB .
Exported proteins are currently the main target for Reverse Vaccinology (RV) due to their essential role in host-pathogen interactions . Examples of this interaction include the following: (i) adherence to host cells; (ii) invasion of the cell to which there was compliance; (iii) damage to host tissues; (iv) resistance from the defense machinery of the cells to environmental stress; and (v) mechanisms for subversion of the host's immune response [9, 10]. In general, RV reveals a great number of proteins that could constitute potential targets of vaccine candidates that then have to be confirmed via cost-intensive and time-consuming wet-lab experiments. However, incorporating immunoinformatic filters, which identify target proteins with high potential in the RV process, could reduce these drawbacks . Immunoinformatics focuses mainly on small peptides ranging from 8 to 11 residues, called linear epitopes, particularly on those that strongly bind to MHC class I molecules. Just one epitope per protein can be enough to create an immune response in the host [12, 13, 14]. Bioinformatic techniques to search for epitopes are well understood and available, but can sometimes lead to high false positive rates . Despite this drawback, epitope predictors are capable of identifying weak or even strong epitope motifs that have been experimentally neglected .
Epitope density has been described in research as a function of "hot spots" or regions with enriched MHC class II binding epitopes . This work reported 544, 609 and 757 15mers peptides binding to three, two and just one of the molecules HLA-DR1, -DR2, and -DR4, respectively. An analysis of two of the 61 proteins examined in that study showed that Ag85B and MPT63 contain, respectively, 30 and 23 peptides with highest binding to MHC molecules; however, experimental data was only available for 10 peptides derived from MPT63.
Asking whether specific defined domains have high epitope densities, one study found that signal peptides and trans-membrane domains have exceptionally high epitope densities . This work computed the high epitope density of signal peptides using in silico methods which corroborate with the high percentage of identified signal peptide epitopes in the IEDB (immune epitope database). The enhanced immunogenicity of signal peptides was experimentally confirmed using peptides derived from Mtb proteins. High antigen-specific response rates and population coverage to signal peptide sequences were found when compared with non-signal peptide antigens derived from the same proteins. The MED (Mature Epitope Density) concept is similar to epitope density . To demonstrate the potential of MED to uncover bacterial targets for RV, we collected a set of experimental evidence from the literature that demonstrates a relationship between high MED scores and promising targets in M. tuberculosis (Mtb) strain H37Rv.
Two results should be noted in the bottom panel. Firstly, the average MED scores were very similar among the three control groups, showing that MED is not necessarily a binary statistic classifier for targets but also a continuous statistic measure capable of defining the preferable targets; however, when significant differences between MED scores are shown, it can be used just like a binary classifier. This procedure was assessed in the evidence dataset shown in the next section. Secondly, the average MED score for proteins predicted as membrane-integral were shown to be twice as great as in the other sub-cellular compartments. This result agrees with other work in which signal peptides and trans-membrane domains were found to have exceptionally high CD8+ T cell epitope densities .
MED highest-scored proteins.
Unique publication identifier
10.1099/jmm.0.47565-0, 10.1046/j.1365-2958.1999.01593.x, 10.1016/j.vaccine.2004.08.046
10.1186/1471-2180-10-132, 10.1021/pr0500049, 10.1016/j.tube.2008.01.003
MED lowest-scored proteins.
Unique publication identifier
MED score limitations
MED score sensitivity
Novel probable putative Mtbantigens
The Mtb H37rv proteins Rv0235c, Rv0492A and Rv1004c were predicted to have some of the highest MED scores: 17.78, 20.31 and 18.58 nM/mer, respectively. The former two were predicted to be potentially exposed on the bacterial surface, and the latter was predicted to be secreted. Respectively, there are 78, 43 and 228 predicted epitopes against 138, 73 and 386 epitope chances for these proteins. This is the first published indication of their roles in bacterial antigenicity; MED scoring results suggest these proteins as useful putative targets for future investigations.
The available methods for MHC epitope prediction take into account allele frequency in the selection of potential candidates [18, 19]. Some alleles are extremely rare; others are specific to some population or widespread . The tools applied here to search for epitopes are not novel, but the way the results are read from standard software tools can be considered a novelty. We proposed to interpret not only epitope prediction from some specific MHC alleles, but from all available alleles. This proposition has a rationale: the idea of assessing the immunogenic potential of a protein, independent of alleles, helps to avoid excluding a protein from a list of in silico candidates just because the suitable allele for a specific population was not selected. For example, there are pathogenic organisms that cause different diseases in different hosts, including humans, caprines, ovines, equines, bovines and buffaloes [21, 22, 23, 24, 25, 26, 27, 28, 29]. In such cases, it is not reasonable to exclude a single allele from the current limited number available in software tools.
Even within the Dminus group, the average MED scores were similar to those from the Dplus and Mtbplus groups. Because of this, we focused on predicted exported proteins to create a priority list of targets for the Mtb genome, which is a reasonable strategy because one of the main differences between the Dminus and the Dplus groups are the number of predicted cytoplasmic versus exported proteins: 111 and 10 for Dminus versus 35 and 72 for Dplus, respectively. It is more likely that exported proteins interact with the host cells than membrane and cytoplasmic proteins [6, 9, 10, 30]. However, it is important not to neglect proteins that could be exported via non-classical mechanisms. This conclusion can also be drawn out by analyzing the middle panel of Figure 2, where the majority of Mtbplus proteins are classified as cytoplasmic. Medpipe allows the prediction of cytoplasmic targets, but this is the major part of any bacterial genome; medpipe still does not allow differentiating between cytoplasmic proteins without classical exportation motifs and those exported via non-classical pathways.
In addition, it is quite difficult to compare MED scores with previous trained software for antigenic features as such programs tend to be binary classifiers [31, 32, 33]. For instance, two control datasets used here were split into training sets (75 proteins) and test sets (25 proteins). Such division does not make sense for MED score because it does not depend on training steps; instead, the MED technique searches for immunological features based on a probable immunological memory concerning epitopes from known pathogens. In this regard, the results obtained with the evidence dataset is more informative because they represent experimental evidence of predictive strengthens or weaknesses of the method.
An extensive literature search for proteins from the well-studied Mtb organism gave experimental indication to validate our hypothesis that promising proteins for reverse vaccinology can be revealed based on the overall set of predicted epitopes. When searching for literature evidence, regarding the proteins within the evidence dataset, experimental results of other proteins were also found but not included in this work. This approach was chosen because it is not possible to determine a mean value for MED scores in order to use it as a binary classifier because the number of epitopes predicted per protein can vary significantly. This limitation was less difficult to work with when considering only 60 proteins: the 30 higher and the 30 lowest MED scored proteins out of 553 Mtb's predicted exported proteins (Figure 3).
The newest NetMHC software (version 3.2) offers the ability to predict epitopes for 57 MHC alleles (http://www.cbs.dtu.dk/services/NetMHC/), but there is not yet a stand-alone version available to download. The NetMHC version (3.0) used here is the previous version and offers the possibility to predict epitopes for 55 MHC alleles . However, the changes in version 3.2, compared to version 3.0, include a small increment in the number of MHC alleles and the possibility to predict epitopes of lengths ranging from 8 to 14mers. The authors of version 3.2 advise that predictions of peptides longer than 11mers have not been extensively validated. They also advise caution regarding predictions involving 8mers, as some alleles might not bind 8mers to any significant extent (http://www.cbs.dtu.dk/services/NetMHC/). Moreover, most MHCs prefer peptides of 9mers and the alleles' set from the version 3.0 are still present in version 3.2 . Therefore, epitope predictions based on version 3.0 are still valid to answer relevant biological queries.
Are these pathogenic proteins?
The method presented here was initially conceived to predict antigenic proteins. Our approach is based on the fact that both antigenic and pathogenic proteins can be useful for vaccines and diagnosis and such targets can be revealed by the overall set of predicted epitopes and their concentrations in mature proteins. As related in the methods section, the in silico predicted exoproteins were ordered by decreasing MED score values. Following this sorting, the literature was searched for evidence proving or denying the contribution for the bacterial pathogenicity of each protein. The majority of the true positives presented here (Table 1) showed pathogenic instead of antigenic evidence (16 out 21 true positives), as detailed in the additional file 1. One protein (Rv3018c) has evidence for both antigenicity and pathogenicity simultaneously. In the same way, this criterion was also applied to the true negatives (Table 2), where seven out of 14 contrary cases fit into the pathogenic class instead of the antigenic one. Could these apparently unexpected results have a rationale? Could pathogenomics explain these findings? Pathogenomics is defined as the analysis, at genomic level, of the processes involved in bacterial pathogenesis caused by the interaction of pathogenic microbes and their hosts . The identification of mutants showing altered pathology may be a useful framework to understand tuberculosis, but it is not clear how these phenotypes relate to the human disease . Here, we presented evidence that Mtb pathogenic proteins have some of the highest MED scores within the Mtb genome.
The search for new vaccine targets against prokaryotic microorganisms has been aided by extensive use of software motif recognition in sequences; nevertheless, considerable experimental effort is necessary to filter out the most promising candidates. The method presented here and the software available online can help to minimize experimental efforts by indicating promising prokaryotic proteins for target selection. The proposed method was called MED score and exhibits a strong relation to proteins proved to be important in the M. tuberculosis pathogenesis.
The complete genome of Mtb H37Rv was obtained from the GenBank database under the NCBI identifier NC_000962. All coding sequences were selected and exported as amino acids in FASTA format using the annotation software ARTEMIS from the Sanger Institute.
Equation 1 divides the number of linear predicted epitopes from each amino acid sequence by the number, for instance, of possible 9mers peptides overlapping windows. To ensure qualitative differentiation for this ratio calculation, the epitopes' MHC binding affinity average is also multiplied, after being normalized according to the maximum MHC strong binding affinity (50 nM). This calculation returns the Mature Epitope Density (MED), a number measured in nanomolar per mer (nM/mer) units. All amino acid sequences are ordered by descending MED score and presented as the final result. The prediction schema was implemented using a Linux shell script. The web server is hosted on Ubuntu OS, release 11.10 and the whole processing takes approximately 90 minutes for Mtb H37Rv amino acid sequences using a standard personal desktop computer.
100 antigen and 100 non-antigen swissprot identifiers were obtained from a previous work . These protein identifiers were retrieved from the Uniprot database , culminating in 107 and 121 amino acid sequences used as positive (Dplus) and negative (Dminus) control groups, respectively. To enrich our tests, a set of 38 Mtb's proteins (Mtbplus) were similarly retrieved from the AntigenDB  and from Uniprot. The Mtbplus control group was obtained selecting the antigenic proteins from M. tuberculosis and filtering for those known as eliciting immune cellular responses.
Sixty proteins out of the 553 in silico predicted as exported were chosen for detailed investigation of experimental proof concerning their capacity to induce cellular responses. In this regard, based on MED, 30 out of 60 proteins were designated as the lowest scored, and the other 30 were designated as the highest scored. An extensive literature search was carried out to look for evidence concerning whether these proteins were related to antigenicity or contribute to the bacterial pathogenicity. Supporting evidence for 39 out of 60 proteins was found, depending on whether a protein induces a cellular response, has evidence of frame shifts, has evidence of differential expression, is part of a known pathogenic protein family or has a cloning experiment that has failed. The complete evidence dataset and corresponding published evidence can be found in the additional file 1.
We thank our colleagues at the Molecular and Cellular Genetics Laboratory (LGMC) of the Federal University of Minas Gerais (UFMG), Brazil, for sharing research skills and for their support. JB, JP and RR are grateful for the financial support from the Cluster of Excellence for Multimodal Computing and Interaction (MMCI) of the German Research Foundation.
Publication for this article has been funded by CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico - Brasil.
This article has been published as part of BMC Genomics Volume 14 Supplement 6, 2013: Proceedings of the International Conference of the Brazilian Association for Bioinformatics and Computational Biology (X-meeting 2012). The full contents of the supplement are available online at http://www.biomedcentral.com/bmcgenomics/supplements/14/S6.
- 5.Pitt JM, Blankley S, McShane H, O'Garra A: Vaccination against tuberculosis: How can we better BCG?. Microb Pathog. 2012Google Scholar
- 6.Santos AR, Ali A, Barbosa E, Silva A, Miyoshi A, Barh D, Azevedo V: The reverse vaccinology - A contextual overview. IIOABJ. 2011, 2: 8-15.Google Scholar
- 8.Seib KL, Zhao X, Rappuoli R: Developing vaccines in the era of genomics: a decade of reverse vaccinology. Clin Microbiol Infect. 2012, 109-116. 18 Suppl 5Google Scholar
- 12.Gustiananda M: Immunoinformatics analysis of H5N1 proteome for designing an epitope-derived vaccine and predicting the prevalence of pre-existing cellular-mediated immunity toward bird flu virus in Indonesian population. Immunome Res. 2011, 7: 1-11.Google Scholar
- 16.Gaseitsiwe S, Valentini D, Mahdavifar S, Reilly M, Ehrnst A, Maeurer M: Peptide microarray-based identification of Mycobacterium tuberculosis epitope binding to HLA-DRB1*0101, DRB1*1501, and DRB1*0401. Clin Vaccine Immunol. 2010, 17: 168-175. 10.1128/CVI.00208-09.PubMedPubMedCentralCrossRefGoogle Scholar
- 17.Kovjazin R, Volovitz I, Daon Y, Vider-Shalit T, Azran R, Tsaban L, Carmon L, Louzoun Y: Signal peptides and trans-membrane regions are broadly immunogenic and have high CD8+ T cell epitope densities: Implications for vaccine development. Mol Immunol. 2011, 48: 1009-1018. 10.1016/j.molimm.2011.01.006.PubMedPubMedCentralCrossRefGoogle Scholar
- 20.Gupta SK, Smita S, Sarangi AN, Srivastava M, Akhoon BA, Rahman Q, Gupta SK: In silico CD4+ T-cell epitope prediction and HLA distribution analysis for the potential proteins of Neisseria meningitidis Serogroup B--A clue for vaccine development. Vaccine. 2010, 28: 7092-7097. 10.1016/j.vaccine.2010.08.005.PubMedCrossRefGoogle Scholar
- 30.Sibbald MJJB, van Dij JML: Secretome Mapping in Gram-Positive Pathogens. Bacterial secreted protein: secretory mechanisms and role in pathogenesis. Edited by: Karl Wooldridge. 2009, Caister Academic Press, 193-225. VolumeGoogle Scholar
- 36.Hacker J: Pathogenomics: Insights into Tuberculosis and Related Mycobacterial Diseases. Pathogenomics: genome analysis of pathogenic microbes. Edited by: Dobrindt U, Gobel W. 2007, Wiley VCH Verlag GmbH & Co. KGaA Weinheim; 2006, 2: 616-Google Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.