Epitope-Based Peptide Vaccine Design against Fructose Bisphosphate Aldolase of Candida glabrata: An Immunoinformatics Approach

Background Candida glabrata is a human opportunistic pathogen that can cause life-threatening systemic infections. Although there are multiple effective vaccines against fungal infections and some of these vaccines are engaged in different stages of clinical trials, none of them have yet been approved by the FDA. Aim Using immunoinformatics approach to predict the most conserved and immunogenic B- and T-cell epitopes from the fructose bisphosphate aldolase (Fba1) protein of C. glabrata. Material and Method. 13 C. glabrata fructose bisphosphate aldolase protein sequences (361 amino acids) were retrieved from NCBI and presented in several tools on the IEDB server for prediction of the most promising epitopes. Homology modeling and molecular docking were performed. Result The promising B-cell epitopes were AYFKEH, VDKESLYTK, and HVDKESLYTK, while the promising peptides which have high affinity to MHC I binding were AVHEALAPI, KYFKRMAAM, QTSNGGAAY, RMAAMNQWL, and YFKEHGEPL. Two peptides, LFSSHMLDL and YIRSIAPAY, were noted to have the highest affinity to MHC class II that interact with 9 alleles. The molecular docking revealed that the epitopes QTSNGGAAY and LFSSHMLDL have the lowest binding energy to MHC molecules. Conclusion The epitope-based vaccines predicted by using immunoinformatics tools have remarkable advantages over the conventional vaccines in that they are more specific, less time consuming, safe, less allergic, and more antigenic. Further in vivo and in vitro experiments are needed to prove the effectiveness of the best candidate's epitopes (QTSNGGAAY and LFSSHMLDL). To the best of our knowledge, this is the first study that has predicted B- and T-cell epitopes from the Fba1 protein by using in silico tools in order to design an effective epitope-based vaccine against C. glabrata.


Introduction
Candidiasis is a fungal infection that has a high burden of morbidity and mortality in hospitalized and immunocompromised patients. It occurs in more than a quarter of a million patients every year with incidence rates for candidemia of 2-14 per 100000 [1][2][3][4]. In general, Candida species infection ranges from superficial mucosal candidiasis such as vulvovaginal candidiasis and oropharyngeal candidiasis to serious systemic infection such as candidemia or fungemia [5][6][7][8]. Pathogenicity is facilitated by a number of virulence factors, most importantly its ability to adhere to host surfaces including medical devices, biofilm formation, and secretion of hydrolytic enzymes. Also, Candida cells elaborate polysaccharides, proteases, phospholipases, and hemolysins that cause host cell damage which leads to the increase in the incidence and antifungal resistance of NCAC species, specifically C. glabrata, and the unfortunate high morbidity and mortality associated with these species [8,9].
Candida glabrata (C. glabrata) is a human opportunistic pathogen that can cause life-threatening systemic infections. C. glabrata is not polymorphic, grows as blastoconidia (yeast), and lacks pseudohyphal formation, so it is classified in the genus Torulopsis. C. glabrata cells (1-4 μm in size) forms glistening, smooth, and cream-colored colonies [10,11]. During the infection, C. glabrata pathogens invade the macrophages, which are considered part of the innate immune system which is the first line of defense against invading pathogens. C. glabrata is able to modify the macrophage's phagosomal compartment, avoiding full maturation and acidification, and thus prevents the forming of the phagolysosomal environment [12]. C. glabrata is able to invade the bloodstream and different organs in a mouse model that have intragastrointestinal infections [13].
C. glabrata has a haploid genome-published in 2004 by Dujon et al. [14]-that allows adaptation to a wide range of environments [9,15,16]. Also, its genome contains more tandem repeats of genes than the other Nakaseomyces [17] and covers 67 genes encoding putative adhesin (cell wall proteins), including the Epa family with 17 members [16,18], such as epithelial adhesin 1 (Epa1p) [19] and fructose bisphosphate aldolase protein (Fba1) which play an essential role in the pathogenicity of Candida species mainly in the adhesion of the pathogen to the host [20,21].
Fba1 is a yeast cell wall protein which presents in multiple species of Candida, e.g., C. glabrata, C. parapsilosis, C. tropicalis, and C. albicans fungal pathogens [22][23][24][25][26]. Fba1 is an important enzyme in the glycolytic pathway [27][28][29][30] and is also a multifunctional protein [31] that can facilitate the attachment (adhesion) to human cells or abiotic surfaces [32][33][34], protects Candida cells from the host's immune system [33], and promotes the detoxification of the ROS generated during the respiratory burst [21,33,34]. However, proteomics analysis revealed that Fba1 is the most abundant and stable enzyme in Candida. Moreover, it is considered one of the main immunodominant proteins [35,36] in Candida cells and has been tested in the murine model as a protected protein against Candida [37], especially C. albicans, and also introduced immunity to C. glabrata [34]; therefore, Fba1 is a potential antifungal target in yeast [38]. Multiple vaccines used Fba1 as an immunogenic protein against different pathogens such as the lethal and challenging S. pneumoniae, Salmonella spp., and M. bovis [39,40].
The incidence of fungal infection has been increasing in the last few years, due to several factors such as misuse of broad-spectrum antibiotics, cytotoxic chemotherapy, immunocompromised patients, and transplantations [15,41]. Invasive fungal infections are a major cause of global morbidity and mortality, accounting for about 1.4 million deaths per year [42]. Systemic fungal infections cost the healthcare industry approximately $2.6 billion per year in the USA alone [43]. However, Candida species pose a base problem in hospitals, according to Healthcare-Associated Infections (HAI) [19,[44][45][46]. Although there are multiple effective vaccines against fungal infections and some of these vaccines are engaged in different stages of clinical trials, none of them have yet been approved by the FDA [47]. Therefore, there is an urgent and crucial need to design vaccines against the Candida species that might improve the quality of life for immunosuppressed patients [48].
The aim of this study is to predict the most conserved and immunogenic B-and T-cell epitopes from the Fba1 protein of C. glabrata by using in silico tools with the immunoinformatics approach presented in the IEDB server [49,50]. This approach has multiple benefits in comparison to other approaches by being affordable, safe, time-saving, and clinically applicable using different computational software techniques [51][52][53]. To the best of our knowledge, this is the first study that has predicted the best candidates of multiple epitopes for Fba1 protein against C. glabrata.

Materials and Methods
In this study, we have used a variety of bioinformatics databases and tools for the prediction of the most promising peptides, through three phases shown in Figure 1.   Figure 2: Bepipred linear epitope prediction: the red line is the threshold; above (the yellow part) is proposed to be part of the B-cell epitope.    The modeled 3D structure was submitted to the ElliPro (http://tools.iedb.org/ellipro/) prediction tool to filter out the antigenic residues. The minimum score and maximum

Result
3.1. B-Cell Epitope Prediction. The reference sequence of fructose bisphosphate aldolase from C. glabrata was analyzed using a Bepipred linear epitope prediction test; the average binder's score of the protein to B-cell was 0.199 and minimum was -0.009 and 2.424 for a maximum score; all values equal or greater than the default threshold 0.350 which were potentially linear epitopes are shown in Figure 2. 3.1.1. Prediction of Surface Accessibility. In Emini's surface accessibility prediction test, for a potent B-cell epitope, the average surface accessibility area of the Fba1 protein was scored as 1.000, with a maximum of 7.725 and a minimum of 0.113; all values equal or greater than the default threshold 1.000 were potentially in the surface shown in Figure 3.

Prediction of Epitope Antigenicity. For the Kolaskar and
Tongaonkar antigenicity prediction test, the average of antigenicity was 1.025, with a maximum of 1.223 and a minimum of 0.853; all values equal to or greater than the default threshold 1.025 are potential antigenic determinants (see Figure 4). The results of all proposed conserved predicted B-cell epitopes are shown in Table 1. The list of the most promising B-cell epitopes with their surface scores and antigenicity is shown in Table 2.
3.1.3. Discontinuous B-Cell Epitope Prediction. The modeled 3D structure of the Fba1 protein was submitted to the ElliPro prediction tool to filter out the antigenic residues. The minimum score and maximum distance (Angstrom) were calibrated in the default mode with a score of 0.5 and 6, respectively (see Table 3 for more illustrations).  Table 4.

Population
Coverage. The most interesting findings in this test is the population coverage analysis result for the most common binders to MHC I and MHC II alleles each and combined among the world, exhibiting an exceptional coverage with percentages 92.54%, 99.58%, and 98.5%, respectively.
3.3.1. Population Coverage for Isolated MHC I. Five epitopes are given to interact with the most frequent MHC class I alleles: AVHEALAPI, KYFKRMAAM, QTSNGGAAY, RMAAMNQWL, and YFKEHGEPL, representing a considerable coverage against the whole world population. The maximum population coverage percentage over these epitopes is 92.54% (see Figure 5).  Figure 6).

Population Coverage for MHC I and MHC II Alleles
Combined. Regarding the combined MHC I and MHC II alleles, five epitopes were supposed to interact with the most predominant MHC class I and MHC class II alleles (IAPAY-GIPV, AAFGNVHGV, VVAALEAAR, YIRSTIAPAY, and YQAGMVVLS), representing a significant global coverage by the IEDB population coverage tool which revealed coverage with percentage of 98.50% as shown in Figure 7.

Homology
Modeling. The 3-dimentional structure of the fructose bisphosphate aldolase protein from C. glabrata and the most promising peptides binding to MHC class II by using the Chimera tool powered by UCSF are shown in Figure 8.   (Met). The half-life of the fructose bisphosphate aldolase protein estimate is 30 hours (mammalian reticulocytes, in vitro) and more than 20 hours (yeast, in vivo). The aliphatic index and the grand average of hydropathicity (GRAVY) value of vaccine were determined as 80.55 and −0.264, respectively. Instability of the fructose bisphosphate aldolase protein is computed to be 29.93, meaning the protein is stable [74]. The amino acids that compose the protein fructose bisphosphate aldolase with their molecular weights are shown in Table 6 and Figure 9.
3.6. Molecular Docking. The best epitopes that displayed the lowest binding energies visualized by using UCSF chimera 1.13.1 software are shown in Table 7 and Figures 10-25.

Discussion
In the present study, we predicted the most conserved and immunogenic B-and T-cell epitopes from Fba1 protein of C. Glabrata by using the immunoinformatics approach in order to develop an effective epitope-based vaccine against this fungal pathogen which has emerged in recent years as a     Journal of Immunology Research serious health problem especially among immunosuppressed and hospitalized patients [7]. A previous study conducted by de Klerk et al. [79] showed that the Fba1 protein has the ability to provoke immune responses in human against M. mycetomatis [79]. Also, several recent publications have used the Fba1 protein as a strong antigenic target for predicting Band T-cell epitopes in order to design promising vaccines against fungal and bacterial pathogens such as M. mycetomatis, P. aeruginosa, L. monocytogenes, and S. mansoni by using in silico tools [80][81][82][83]. Hence, there are more studies to explore the fructose bisphosphate aldolase protein immunogenic role and the possibility to find common conserved epitopes for different organisms.
The principle of using a cocktail of B-and T-cell epitopes in the epitope-based vaccine to trigger humoral as well as cellular mediated immune response is very promising to clear infection instead of humoral or cellular immunity alone, and it was applied before to enhance protection against different kinds of infectious diseases [84,85]. In this study, the analysis of the Fba1 protein revealed 11 effective epitopes for B-cells (AYFKEH, VDKESLYTK, and HVDKESLYTK) and T-cells (AVHEALAPI, KYFKRMAAM, QTSNGGAAY, RMAAMNQWL, YFKEH-GEPL, IRGSIAAAH, LFSSHMLDL, and VVAALEAAR).
However, the molecular docking, which evaluates the binding affinity to MHC molecules [51,52], showed that the peptides QTSNGGAAY and LFSSHMLDL are the best      Journal of Immunology Research            corresponding to their interaction with MHC II alleles; both had the affinity to interact with 9 MHC II alleles (see Table 5). Moreover, the predicted epitopes which have the high affinity to interact with MHC I, MHC II, and combined MHC I with MHC II international alleles were analyzed by population coverage resource in the IEDB [68]. The population coverage of the five most promising epitopes (AVHEA-LAPI, KYFKRMAAM, QTSNGGAAY, RMAAMNQWL, and YFKEHGEPL) for MHC I alleles was 92.54%, while for the three epitopes (IRGSIAAAH, LFSSHMLDL, and VVAALEAAR) that showed high affinity to MHC II alleles, it was 99.58% throughout the world according to the IEDB database as shown in Figures 5 and 6. It should be noted that the population coverage of the five most promising epitopes that exhibited binding affinity to both MHC I and MHC II alleles (IAPAYGIPV, AAFGNVHGV, VVAALEAAR, YIRSTIAPAY, and YQAGMVVLS) was 98.50% globally (see Figure 7). However, the molecular docking revealed that the epitopes QTSNGGAAY and LFSSHMLDL have high binding energy to MHC molecules HLA-A * 02:06 and HLA-DRB1 * 01:01, respectively, which indicate favored affinity and stability in the epitope-molecule complex shown in Table 7 and Figures 10-25. This study was limited by being strictly computational, and more in vitro and in vivo studies to prove the effectiveness of the proposed peptides are highly recommended.

In Conclusion
The epitope-based vaccines predicted by using immunoinformatics tools have remarkable advantages over the conventional vaccines in that they are more specific, less time consuming, safe, less allergic, and more antigenic. Further in vivo and in vitro experiments are needed to prove the effectiveness of the best candidate's epitopes QTSNGGAAY and LFSSHMLDL. To the best of our knowledge, this is the first study that has predicted B-and T-cell epitopes from the Fba1 protein by using in silico tools in order to design an effective epitope-based vaccine against C. glabrata.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.   16 Journal of Immunology Research

Conflicts of Interest
The authors declare that there are no conflicts of interest.