Iranian Journal of Medical Sciences

Document Type: Original Article(s)

Authors

1 Department of Medical Biotechnology, School of Advanced Medical Sciences and Technologies, Shiraz University of Medical Sciences, Shiraz, Iran

2 Pharmaceutical Sciences Research Center, Shiraz University of Medical Sciences, Shiraz, Iran

3 Student Research Committee, Shiraz University of Medical Sciences, Shiraz, Iran

4 Department of Nanobiotechnology, School of Biological Sciences, Tarbiat Modares University, Tehran, Iran

5 Department of Medical Biotechnology, School of Advanced Technologies in Medicine, Mazandaran University of Medical Sciences, Sari, Iran

6 Department of Biochemistry, School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran

7 Department of Pharmaceutical Biotechnology, School of Pharmacy, Shiraz University of Medical Sciences, Shiraz, Iran

Abstract

Background: The most prevalent cancer in women over the world is breast cancer. Immunotherapy is a promising method to effectively treat cancer patients. Among various immunotherapy methods, tumor antigens stimulate the immune system to eradicate cancer cells. Preferentially expressed antigen in melanoma (PRAME) is mainly overexpressed in breast cancer cells, and has no expression in normal tissues. FliCΔD2D3, as truncated flagellin (FliC), is an effective toll-like receptor 5 (TLR5) agonist with lower inflammatory responses. The objective of the present study was to utilize bioinformatics methods to design a chimeric protein against breast cancer.
Methods: The physicochemical properties, solubility, and secondary structures of PRAME+FliCΔD2D3 were predicted using the tools ProtParam, Protein-sol, and GOR IV, respectively. The 3D structure of the chimeric protein was built using I-TASSER and refined with GalaxyRefine, RAMPAGE, and PROCHECK. ANTIGENpro and VaxiJen were used to evaluate protein antigenicity, and allergenicity was checked using AlgPred and Allergen FP. Major histocompatibility complex
)MHC( and cytotoxic T-lymphocytes )CTL( binding peptides were predicted using HLApred and CTLpred. Finally, B-cell continuous and discontinuous epitopes were predicted using ABCpred and ElliPro, respectively.
Results: The stability and solubility of PRAME+FliCΔD2D3 were analyzed, and its secondary and tertiary structures were predicted. The results showed that the derived peptides could bind to MHCs and CTLs. The designed chimeric protein possessed both linear and conformational epitopes with a high binding affinity to B-cell epitopes.
Conclusion: PRAME+FliCΔD2D3 is a stable and soluble chimeric protein that can stimulate humoral and cellular immunity. The obtained results can be utilized for the development of an experimental vaccine against breast cancer.

Keywords

What’s Known

Standard therapies used to treat breast cancer are only effective in approximately half of the patients. Among various immunotherapy methods, vaccination has gained much attention in treating breast cancer.

What’s New

Preferentially expressed antigen in melanoma (PRAME) combined with FliCΔD2D3 is a stable and soluble chimeric protein. PRAME+FliCΔD2D3 stimulated humoral and cellular immunity against breast cancer.

Introduction

Breast cancer is considered as the most common women’s cancer worldwide, and is the second leading cause of death after lung cancer. 1 The World Health Organization (WHO) has estimated 2.1 million new cases of breast cancer each year, and has reported 627,000 related deaths in 2018. 2 , 3 Such a high number of deaths has been attributed to tumor progression and metastasis to other organs. Breast cancer is a heterogeneous disease, which requires targeted therapy consistent with the tumor’s characteristics. 4 Standard therapies used to treat breast cancer (e.g., surgery, radiation therapy, anti-estrogen therapy, and chemotherapy; individually or in combination) are only effective in approximately half of the patients. It has been shown that these therapies pose adverse side-effects, since they do not target specific tumor cells. 5 - 7 Immunotherapy is a targeted approach and a promising method to treat cancer patients. Among various immunotherapy methods, vaccination has gained much attention in treating breast cancer. Recently, it has been shown that the use of antigen-specific vaccines are effective in eradicating breast tumor cells. 8 , 9

Preferentially expressed antigen in melanoma (PRAME) is a human tumor antigen found in melanomas. PRAME is also known as melanoma antigen, which is preferentially expressed in tumors (MAPE), OPA-interacting protein 4 (OIP4), and cancer-testis antigen 130 (CT130). The human PRAME gene is located on chromosome 22q11.22 and encodes a protein of 509-amino acids. 10 PRAME induces cell proliferation in melanoma cells by repressing retinoic acid (RA) signaling. 11 It belongs to the cancer-testis antigen (CTA) family based on its chromosomal position, expression, and immunogenicity. Typically, PRAME has no expression in normal tissues. However, low level of expression has been observed in some organs such as testis, ovaries, adrenal glands, and endometrium. 12 Aberrant expression of PRAME has been demonstrated in various cancers such as neuroblastoma, 13 ovarian adenocarcinomas, 14 head and neck cancer, 15 acute and chronic leukemia, 10 and breast cancer. 16 Previous studies have reported the overexpression of PRAME in breast cancer, making it a suitable prognostic and predictive biomarker. It is viewed as a potential candidate for immunotherapeutic strategies. 4 , 16

The goal of vaccination is to provide a strong immune response for long-term protection against antigens. To achieve this goal, a vaccine requires an adjuvant. Adjuvants are used to enhance the immune response against the vaccine and are classified in various groups. Toll-like receptors (TLR) ligands are known as a popular group of adjuvants that improve the immunogenicity of vaccines. 17 , 18 Bacterial flagellins (also termed FliC) is considered as an effective adjuvant candidate. Salmonella typhimurium FliC, a principal structural protein of flagella, is identified as a TLR5 ligand. FliC stimulates several innate immune cells to release specific cytokines and chemokines, which in turn stimulate an adaptive immune response. FliC is a 494-amino acid protein consisting of two separate domains, namely D0/D1 and D2/D3. The domain D0/D1 (with 170 and 90 amino acids) plays a role in TLR5 agonist activity, and the domain D2/D3 (with 170 and 400 amino acids) is essential for flagellin antigenicity. 19 - 21 Various studies have shown the high antigenicity and inflammatory effects of FliC. As a result, truncated flagellin (FliCΔD2D3) has been designed by removing the hypervariable D2/D3 domain. In comparison to FliC, FliCΔD2D3 exhibits remarkably lower inflammatory responses, TLR5 agonist potency, more efficient immunoreactivity, and least allergenicity. 22 , 23

The design and development of a new vaccine are still time-consuming and requires extensive experimental investigations. Among various approaches, bioinformatics methods have been utilized to design more effective vaccines. 24 The present study aimed to employ bioinformatics methods to design a chimeric protein as a novel recombinant construct against breast cancer, composed of PRAME as the antigenic fragment and FliCΔD2D3 as the adjuvant part.

Materials and Methods

The present study was conducted in 2019 at the School of Advanced Medical Sciences and Technologies, Shiraz University of Medical Sciences, Shiraz, Iran. A complete amino acid sequence of PRAME was obtained from the UniProt Knowledgebase (http://www.uniprot.org, UniProt ID: P78395). FliCΔD2D3 (D0/D1+D1/D0) sequence was in accordance with a previous study. 22 The construct included PRAME at N-terminus connected to FliCΔD2D3.

Physicochemical Properties and Solubility

Different physicochemical characteristics, such as theoretical isoelectric point (pI), the total number of positive and negative residues, molecular weight (MW), instability index, aliphatic index (AI), and grand average hydropathy (GRAVY) of the chimeric protein were computed using the ProtParam web server (https://web.expasy.org/protparam). Protein solubility was evaluated using the Protein-sol web server (http://scratch.proteomics.ics.uci.edu/).

Secondary Structure Prediction

The secondary structure of PRAME, FliCΔD2D3 (D0/D1+D1/D0), and PRAME+FliCΔD2D3 (D0/D1+D1/D0) proteins was predicted using the GOR IV web server with an accuracy of 73.5% (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_gor4.html).

3D Model Building, Refinement, and Docking

The I-TASSER web server (https://zhanglab.ccmb.med.umich.edu/I-TASSER/) was used to build a 3D model of the chimeric protein. The server is a combined platform for computerized protein structure and function prediction. The optimal 3D model from I-TASSER was used for refinement using the GalaxyRefine web server (http://galaxy.seoklab.org/cgi-bin/submit.cgi?type=REFINE). Validation of the 3D model of the chimeric structure was performed using the PROCHECK and RAMPAGE web servers. The PROCHECK (http://servicesn.mbi.ucla.edu/PROCHECK/) was used to check the stereochemical quality of the protein structure, and its results were verified using the RAMPAGE web server (http://mordred.bioc.cam.ac.uk/~rapper/rampage.php). The interaction between the 3D model of the chimeric protein and TLR5 was predicted using the ZDOCK web server with >70% accuracy.

Antigenicity and Allergenicity Evaluation

The ANTIGENpro and VaxiJen web servers were used to evaluate the antigenicity of PRAME, FliCΔD2D3, and the chimeric protein. ANTIGENpro (http://scratch.proteomics.ics.uci.edu/) is a sequence-based alignment-free and pathogen-independent predictor, which uses protein antigenicity microarray data for predicting protein antigenicity. VaxiJen (http://www.ddg-pharmfac.net/vaxijen/VaxiJen/VaxiJen.html) was used for the anticipation of conserved regions of antigens, and candidate vaccines according to physiochemical properties of proteins, neglecting alignment of sequences. The allergenicity of the chimeric protein was analyzed using the AlgPred and Allergen FP v.1.0 web servers. Allergen FP v.1.0 (http://ddg-pharmfac.net/Allergen_FP) is a descriptor-based fingerprint bioinformatics tool. AlgPred (http://crdd.osdd.net/raghava/algpred/) predicts allergens based on similarity to known epitopes.

Prediction of MHC Binding Peptides

Major histocompatibility complex (MHC) class I and class II binding peptides were identified using the HLApred web server (http://crdd.osdd.net/raghava/hlapred/). The server predicts HLA binder peptides that can be potential vaccine candidates.

Prediction of B-Cell Epitopes

The ABCpred web server (http://crdd.osdd.net/raghava/abcpred/) was used to predict linear (continuous) epitopes with 65.93% accuracy. This server has been established through a machine-based technique, which considered a recurrent neural network via fixed-length patterns. ElliPro (http://tools.iedb.org/ellipro/) is an online bioinformatics web server that has been used for the prediction of conformational (discontinuous) antigens, based on the 3D structure of antigenic proteins. This structure-based server uses three algorithms, namely approximation of a protein surface patch by an ellipsoid, computation of the residue protrusion index (PI), and cluster neighboring residues based on their PI values.

Results

Physicochemical Analysis of PRAME+FliCΔD2D3 Protein

ProtParam analysis of PRAME+FliCΔD2D3 protein showed that the molecular weight and theoretical isoelectric point (pI) of the chimeric protein were approximately 87 kDa and 5.89 units, respectively. We found 79 negatively charged aspartate and glutamate amino acids, as well as 70 positively charged arginine and lysine in the protein sequence (i.e., the net charge of the protein was -9). Based on the computed instability index of 42.44, the chimeric protein was considered an unstable protein. The AI and GRAVY indices were 101.63 and -0.107, respectively. The solubility of the protein was determined using the SOLpro web server. The result showed that this protein could be soluble with a probability of 0.585946 (cut-off: 0.5).

Secondary and Tertiary Structure

The results of GOR IV for the prediction of the secondary structure of the chimeric protein indicated that the PRAME consisted of 46.95% alpha-helix, 36.94% random coil, and 16.11% extended strand. Whereas the chimeric protein consisted of 50% alpha-helix, 37.34% random coil, and 12.66% extended strand. Therefore, the protein was mainly composed of regular structures. The 3D models of PRAME+FliCΔD2D3 protein were built using the I-TASSER web server. A model with the most positive c-score was chosen (data not shown). The putative tertiary structure is shown in figure 1.

Figure 1. Tertiary structure of PRAME-FliCΔD2D3 is illustrated. The structure was modeled and built using I-TASSER homology-based modeling.

3D Structure Refinement, Validation, and Docking

An optimal I-TASSER model was used as an input to the GalaxyRefine web server. This server suggested five refined protein models (data not shown) of which the model with the highest Rama favored score was chosen to carry out more analysis. Based on the Ramachandran plot, the results of PROCHECK and RAMPAGE web servers showed that residues in the favored and most favored regions were increased after model refinement (figure 2). The results of PROCHECK showed that residues in most favored regions before and after 3D model refinement were 80.6% and 88.5%, respectively. In addition, the results of RAMPAGE showed that residues in the favored region before and after 3D model refinement were 83.0% and 94.4%, respectively. PRAME+FliCΔD2D3 tertiary refined model and the TLR5 interaction showed that the chimeric vaccine could bind to TLR5 receptor (figure 3).

Figure 2. Ramachandran plot for the 3D structure before (left) and after (right) refinement using the PROCHECK web server is illustrated.

Figure 3. Docking of PRAME-FliCΔD2D3 and toll-like receptor (TLR5) shows suitable interaction between them.

Antigenicity and Allergenicity

The results of ANTIGENpro showed that the antigenicity of the chimeric protein was increased compared to that of the PRAME alone (from 0.161 to 0.485, threshold: 0.4). In addition, using the VaxiJen web server, the predicted antigenicity of PRAME and the chimeric protein was 0.527 and 0.611 (threshold: 0.5), respectively. This implied that FliCΔD2D3 might increase the antigenicity of PRAME. The AlgPred and Allergen FP v.1.0 web servers predicted the non-allergenic potential of the chimeric protein.

MHC-I and MHC-II Binding Peptides

The HLApred web server predicted possible MHC-I binding peptides in the PRAME+FliCΔD2D3 chimeric protein. The position, sequence, and score of the 10 best-ranked peptides with the potential to bind to MHC-I are listed in table 1. Moreover, the predicted binding peptides, which had an affinity to the DRB1_1101, DRB1_0301, and DRB1_1001 alleles of MHC-II are listed in table 2.

Allele Rank Position Sequence Score Prediction
HLA-A*0201 (threshold: 7.470) 1 371 ALLERASAT 12.900 Binder
2 194 YLIEKVKRK 12.770 Binder
3 190 ELFSYLIEK 12.390 Binder
4 28 ELAGQSLLK 12.290 Binder
5 237 DLEVTCTWK 12.230 Binder
6 429 HLIGLSNLT 11.030 Binder
7 118 KLQVLDLRK 10.780 Binder
8 401 SLSHCSQLT 9.940 Binder
9 469 ELLCELGRP 8.970 Binder
10 68 TLKAMVQAW 8.710 Binder
Table 1. Major histocompatibility complex I binding peptides of preferentially expressed antigen in melanoma
Allele Rank Position Sequence Score Prediction
HLA-DRB1*0301 (threshold: 2.960) 1 223 IKMILKMVQ 5.200 Binder
2 230 VQLDSIEDL 4.960 Binder
3 383 LVFDECGIT 4.800 Binder
4 301 LYVDSLFFL 4.360 Binder
HLA-DRB1*0401 (threshold: 1.480) 1 194 YLIEKVKRK 4.300 Binder
2 738 YATEVSNMS 3.800 Binder
3 321 MNPLETLSI 3.200 Binder
4 430 LIGLSNLTH 3.180 Binder
HLA-DRB1*0701 (threshold: 4.100) 1 513 VINTNSLSL 8.200 Binder
2 7 WGSIQSRYI 6.500 Binder
3 669 LKQINSQTL 6.300 Binder
4 238 LEVTCTWKL 5.720 Binder
HLA-DRB1*1501 (3% threshold=3.250) 1 259 MINLRRLLL 5.500 Binder
2 262 LRRLLLSHI 4.600 Binder
3 513 VINTNSLSL 4.300 Binder
4 321 MNPLETLSI 4.160 Binder
PRAME: Preferentially expressed antigen in melanoma
Table 2. Major histocompatibility complex II binding peptides of PRAME-FliCΔD2D3

B-cells Epitopes

The ABCpred and ElliPro web servers predicted B-cell continuous (linear) and discontinuous (conformational) epitopes, respectively. Residue, position, and score of the predicted epitopes are listed in tables 3 and 4.

Rank Sequence Start position Score
1 HCGDRTFYDPEPILCP 489 0.92
2 WGSIQSRYISMSVWTS 7 0.90
3 FPEPEAAQPMTKKRKV 146 0.87
4 PPLFMAAFDGRHSQTL 54 0.86
4 RPRRWKLQVLDLRKNS 113 0.86
5 DSIEDLEVTCTWKLPT 233 0.83
5 MQDIKMILKMVQLDSI 220 0.83
5 MSVWTSPRRLVELAGQ 17 0.83
5 FWTVWSGNRASLYSFP 132 0.83
6 DGRHSQTLKAMVQAWP 62 0.81
6 NCRLSEGDVMHLSQSP 331 0.81
7 LRHVMNPLETLSITNC 317 0.80
8 EVTCTWKLPTLAKFSP 239 0.79
9 ASSYISPEKEEQYIAQ 272 0.78
9 RASLYSFPEPEAAQPM 140 0.78
10 DECGITDDQLLALLPS 386 0.77
10 LKMVQLDSIEDLEVTC 227 0.77
Table 3. B-cell continuous epitopes of Preferentially expressed antigen in melanoma (threshold: 0.5)
No. Residues Number of residues Score
1 A:H129, A:Q130, A:F132, A:W133, A:T134, A:V135, A:W136, A:S137, A:G138, A:N139, A:R140, A:A141, A:S142, A:L143, A:Y144, A:S145, A:F146, A:P147, A:E148, A:P149, A:E150, A:A151, A:A152, A:Q153, A:P154, A:M155, A:T156, A:K157, A:K158, A:R159, A:K160, A:V161, A:D162, A:G163, A:L206, A:R207, A:L208, A:C209, A:C210, A:K211, A:K212, A:L213, A:K214, A:I215, A:F216, A:A217, A:M218, A:P219, A:M220, A:Q221, A:D222, A:I223, A:K224, A:M225, A:L227, A:V230, A:A376, A:S377, A:A378, A:T379, A:L380, A:Q381, A:D382, A:L383, A:V384, A:F385, A:E387, A:C388, A:G389, A:I390, A:T391, A:D392, A:D393, A:Q394, A:L395, A:L396, A:V439, A:Y441, A:P442, A:V443, A:P444, A:L445, A:E446, A:S447, A:Y448, A:E449, A:D450, A:I451, A:H452, A:G453, A:T454, A:L455, A:H456, A:L457, A:E458, A:R459 96 0.731
2 A:M1, A:E2, A:R3, A:R4, A:R5, A:L6, A:W7, A:G8, A:S9, A:I10, A:Q11, A:S12, A:R13 13 0.603
Table 4. B-cell discontinuous epitopes of Preferentially expressed antigen in melanoma

Discussion

In the present study, a chimeric protein was designed as a breast cancer vaccine candidate. PRAME is a suitable candidate for immunotherapy, since its expression increases in breast cancer cells, and it is a suitable prognostic and predictive biomarker. 4 , 16 As an adjuvant FliCΔD2D3, a truncated FliC, exhibits less inflammatory properties and more efficient TLR5 potency than FliC. Therefore, we used FliCΔD2D3, as the adjuvant molecule capable of stimulating both the innate and adaptive immune responses. 22 This chimeric protein was then analyzed and evaluated using various bioinformatics web servers. Based on the physicochemical parameters and structural features of PRAME+FliCΔD2D3, the results showed an instability index of 42.44, indicating that the protein to be unstable (values above 40.0 are considered unstable). However, previous studies reported that value in the range of 40 to 45 (as in our case) could be considered stable. 25 Based on the calculated net charge (-9), our chimeric protein could be soluble. Note that the protein net charge is a significant physical property, which directly affects its solubility, aggregation, and crystallization. Since the body recognizes hydrophobic particles as foreign substances, and the reticulo-endothelial system removes them from the bloodstream, proteins with a net negative charge can mask the hydrophobicity and prevent the removal of the particles. 26 The net charge of proteins can also impact thermal stability, and the free energy interplay between proteins and ligands. 27 Depending on the desired level of stability and charge, PRAME+FliCΔD2D3 could be soluble, easily circulate in the bloodstream, be processed by antigen-presenting cells (APCs), and would not be easily removed from the body. The computed AI was 101.63. It is indicated that AI is related to amount of aliphatic side chains in a protein (A, V, I, and L), and is a positive factor for increased thermostability of a protein. Lower thermostability of PRAME+FliCΔD2D3 indicated a more flexible structure. The GRAVY index of PRAME+FliCΔD2D3 was as low as -0.107, indicating better interaction with water. 28

The secondary structure of polypeptides plays a central role in their eventual formation and function. The GOR IV web server was used to predict the secondary structure of PRAME+FliCΔD2D3. The algorithm of this web server is based on the GOR (Garnier-Osguthorpe-Robson) method, which utilizes a combination of mathematical probability and experimental data (e.g., data from nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallography). 29 The results from the GOR IV webserver showed that the addition of FliCΔD2D3, as an adjuvant, did not alter the secondary structure of the antigen. The percentage of alpha helices and random coils increased in PRAME+FliCΔD2D3, and its construction became more regular than that of PRAME alone (table 1). Among protein secondary structures, alpha helices are reported to be more stable and more resistant to conformational variations. 30 Various web servers were used to build, refine, and validate a 3D model of PRAME+FliCΔD2D3. Based on the RAMPAGE web server results, most residues were distributed in the favored and allowed regions, and the residues increased after refinement. These results were confirmed using the PROCHECK web server. The refined 3D model of protein docking of the chimeric vaccine showed that it could bind to TLR5 and stimulate APCs’ augmentation and T-helper cells (Th) responses.

Since bacterial flagellin was used as an adjuvant in our chimeric vaccine, it was important to evaluate its potential allergenicity and ensure that it does not induce allergic responses in the human body. Allergenic substances may adversely affect the host’s immune system and result in undesired allergic diseases. The results of the allergenicity assessment confirmed that PRAME+FliCΔD2D3 did not induce an allergic response.

To elicit an immune response against tumor cells, immune cells such as B- and T-lymphocytes must be presented to antigen epitopes. MHC-I and MHC-II binding peptides can stimulate cytotoxic T-lymphocytes (CTL) and Th, respectively. The MHC human genes are among the most polymorphic genes. MHC-I and MHC-II alleles were used to determine the potential presentation and binding affinity of PRAME+FliCΔD2D3 by APCs. These alleles were selected because of their global and local comparative haplotypes frequency and their incidence in allele frequency database. 31 - 33 The position, sequence, and score of the 10 best-ranked peptides with MHC-I binding affinity were identified. Peptides that had the best binding affinity to DRB1-1101, DRB1-0301, and DRB1-1001 alleles of MHC-II were also identified. These confirmed that PRAME+FliCΔD2D3 could be effectively presented to T-cells and stimulate immunity against cancer cells.

Circulating soluble antigens such as vaccines could be composed of B-cells epitopes, which stimulate the generation of specific antibodies and activate humoral immunity. B-cell epitopes (antigenic determinants) are recognized by B-cell receptors or secreted antibodies. In terms of spatial structure, B-cell epitopes are categorized as continuous (linear) and discontinuous (conformational) epitopes. For the latter, amino acid residues are in close interaction because of the 3D structure. In fact, B-cell epitopes are the instigators of antibody generation. 34 As the results of the presented study indicated, the PRAME+FliCΔD2D3 chimeric protein possessed both linear and conformational epitopes with a strong affinity to B-cells, indicating that it can stimulate humoral immunity.

The main limitation of the present study was that we only performed in silico analysis. It is recommended that future studies experimentally validate our bioinformatics data.

Conclusion

Given its structure, PRAME+FliCΔD2D3 is a suitable candidate for a new breast cancer vaccine and could effectively stimulate both cellular and humoral immunity. It is recommended to validate this vaccine candidate through in vitro and in vivo studies.

References

  1. Hamdan D, Nguyen TT, Leboeuf C, Meles S, Janin A, Bousquet G. Genomics applied to the treatment of breast cancer. Oncotarget. 2019; 10:4786-801. Publisher Full Text | DOI | PubMed
  2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019; 69:7-34.
  3. World Health Organization. Global status report on noncommunicable diseases 2014. Geneva: World Health Organization; 2014.
  4. Ravi M, Sneka MK, Joshipura A. The culture conditions and outputs from breast cancer cell line in vitro experiments. Exp Cell Res. 2019; 383:111548. DOI | PubMed
  5. Lanari C, Wargon V, Rojas P, Molinolo AA. Antiprogestins in breast cancer treatment: are we ready?. Endocr Relat Cancer. 2012; 19:R35-50. DOI | PubMed
  6. Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans V, et al. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005; 366:2087-106. DOI | PubMed
  7. Zhou J, Zhong Y. Breast cancer immunotherapy. Cell Mol Immunol. 2004; 1:247-55. PubMed
  8. Mohseni MM, Amani J, Gheybi E, Salmanian AH. In silico analysis of chimeric subunit vaccine containing HER-2-MUC1 against breast cancer. 2018; 17:224-33.
  9. Parvizpour S, Razmara J, Pourseif MM, Omidi Y. In silico design of a triple-negative breast cancer vaccine by targeting cancer testis antigens. Bioimpacts. 2019; 9:45-56. Publisher Full Text | DOI | PubMed
  10. Wadelin F, Fulton J, McEwan PA, Spriggs KA, Emsley J, Heery DM. Leucine-rich repeat protein PRAME: expression, potential functions and clinical implications for leukaemia. Mol Cancer. 2010; 9:226. Publisher Full Text | DOI | PubMed
  11. Epping MT, Wang L, Edel MJ, Carlee L, Hernandez M, Bernards R. The human tumor antigen PRAME is a dominant repressor of retinoic acid receptor signaling. Cell. 2005; 122:835-47. DOI | PubMed
  12. Ikeda H, Lethe B, Lehmann F, van Baren N, Baurain JF, de Smet C, et al. Characterization of an antigen that is recognized on a melanoma showing partial HLA loss by CTL expressing an NK inhibitory receptor. Immunity. 1997; 6:199-208. DOI | PubMed
  13. Oberthuer A, Hero B, Spitz R, Berthold F, Fischer M. The tumor-associated antigen PRAME is universally expressed in high-stage neuroblastoma and associated with poor outcome. Clin Cancer Res. 2004; 10:4307-13. DOI | PubMed
  14. Zhang W, Barger CJ, Eng KH, Klinkebiel D, Link PA, Omilian A, et al. PRAME expression and promoter hypomethylation in epithelial ovarian cancer. Oncotarget. 2016; 7:45352-69. Publisher Full Text | DOI | PubMed
  15. Szczepanski MJ, Whiteside TL. Elevated PRAME expression: what does this mean for treatment of head and neck squamous cell carcinoma?. Biomark Med. 2013; 7:575-8. Publisher Full Text | DOI | PubMed
  16. Doolan P, Clynes M, Kennedy S, Mehta JP, Crown J, O’Driscoll L. Prevalence and prognostic and predictive relevance of PRAME in breast cancer. Breast Cancer Res Treat. 2008; 109:359-65. DOI | PubMed
  17. Song L, Xiong D, Song H, Wu L, Zhang M, Kang X, et al. Mucosal and Systemic Immune Responses to Influenza H7N9 Antigen HA1-2 Co-Delivered Intranasally with Flagellin or Polyethyleneimine in Mice and Chickens. Front Immunol. 2017; 8:326. Publisher Full Text | DOI | PubMed
  18. Dubensky TW, Jr.Reed SG. Adjuvants for cancer vaccines. Semin Immunol. 2010; 22:155-61. DOI | PubMed
  19. Smith KD, Andersen-Nissen E, Hayashi F, Strobe K, Bergman MA, Barrett SL, et al. Toll-like receptor 5 recognizes a conserved site on flagellin required for protofilament formation and bacterial motility. Nat Immunol. 2003; 4:1247-53. DOI | PubMed
  20. Yoshioka K, Aizawa S, Yamaguchi S. Flagellar filament structure and cell motility of Salmonella typhimurium mutants lacking part of the outer domain of flagellin. J Bacteriol. 1995; 177:1090-3. Publisher Full Text | DOI | PubMed
  21. Mizel SB, Bates JT. Flagellin as an adjuvant: cellular mechanisms and potential. J Immunol. 2010; 185:5677-82. Publisher Full Text | DOI | PubMed
  22. Song L, Xiong D, Kang X, Jiao Y, Zhou X, Wu K, et al. The optimized fusion protein HA1-2-FliCDeltaD2D3 promotes mixed Th1/Th2 immune responses to influenza H7N9 with low induction of systemic proinflammatory cytokines in mice. Antiviral Res. 2019; 161:10-9. DOI | PubMed
  23. Xiao Y, Liu F, Yang J, Zhong M, Zhang E, Li Y, et al. Over-activation of TLR5 signaling by high-dose flagellin induces liver injury in mice. Cell Mol Immunol. 2015; 12:729-42. Publisher Full Text | DOI | PubMed
  24. Karkhah A, Saadi M, Nouri HR. In silico analyses of heat shock protein 60 and calreticulin to designing a novel vaccine shifting immune response toward T helper 2 in atherosclerosis. Comput Biol Chem. 2017; 67:244-54. DOI | PubMed
  25. Singh R. Bioinformatics: genomics and proteomics. Uttar Pradesh: Vikas Publishing House; 2015.
  26. Tabatabaei Mirakabad FS, Akbarzadeh A, Milani M, Zarghami N, Taheri-Anganeh M, Zeighamian V, et al. A Comparison between the cytotoxic effects of pure curcumin and curcumin-loaded PLGA-PEG nanoparticles on the MCF-7 human breast cancer cell line. Artif Cells Nanomed Biotechnol. 2016; 44:423-30. DOI | PubMed
  27. Gitlin I, Carbeck JD, Whitesides GM. Why are proteins charged? Networks of charge-charge interactions in proteins measured by charge ladders and capillary electrophoresis. Angew Chem Int Ed Engl. 2006; 45:3022-60. DOI | PubMed
  28. Amala S. In silico analysis and 3D modeling of ASAH1 protein in farber lipogranulomatosis. Advanced Biotech. 2010; 10:6-8.
  29. Kouza M, Faraggi E, Kolinski A, Kloczkowski A. The GOR Method of Protein Secondary Structure Prediction and Its Application as a Protein Aggregation Prediction Tool. Methods Mol Biol. 2017; 1484:7-24. DOI | PubMed
  30. Abrusan G, Marsh JA. Alpha Helices Are More Robust to Mutations than Beta Strands. PLoS Comput Biol. 2016; 12:e1005242. Publisher Full Text | DOI | PubMed
  31. Farjadian S, Naruse T, Kawata H, Ghaderi A, Bahram S, Inoko H. Molecular analysis of HLA allele frequencies and haplotypes in Baloch of Iran compared with related populations of Pakistan. Tissue Antigens. 2004; 64:581-7. DOI | PubMed
  32. Gonzalez-Galarza FF, Takeshita LY, Santos EJ, Kempson F, Maia MH, da Silva AL, et al. Allele frequency net 2015 update: new features for HLA epitopes, KIR and disease and HLA adverse drug reaction associations. Nucleic Acids Res. 2015; 43:D784-8. Publisher Full Text | DOI | PubMed
  33. Awad Elkareem M, Ahmed Osman S, Mohamed H. Prediction and conservancy analysis of multiepitope based peptide vaccine against merkel cell polyomavirus: an immunoinformatics approach. Immun Res. 2017; 13:2. DOI
  34. Potocnakova L, Bhide M, Pulzova LB. An Introduction to B-Cell Epitope Mapping and In Silico Epitope Prediction. J Immunol Res. 2016; 2016:6760830. Publisher Full Text | DOI | PubMed