Decision-making critical amino acids: role in designing peptide vaccines for eliciting Th1 and Th2 immune response
- First Online:
- Cite this article as:
- Mushtaq, K., Chodisetti, S.B., Rai, P.K. et al. Amino Acids (2014) 46: 1265. doi:10.1007/s00726-014-1692-4
- 275 Downloads
CD4 T cells play a cardinal role in orchestrating immune system. Differentiation of CD4 T cells to Th1 and Th2 effector subsets depends on multiple factors such as relative intensity of interactions between T cell receptor with peptide-major histocompatibility complex, cytokine milieu, antigen dose, and costimulatory molecules. Literature supports the critical role of peptide’s binding affinity to Human Leukocyte Antigens (HLAs) and in the differentiation of naïve CD4 T cells to Th1 and Th2 subsets. However, there exists no definite report addressing very precisely the correlation between physicochemical properties (hydrophobicity, hydrophilicity), pattern, position of amino acids in peptide and their role in skewing immune response towards Th1 and Th2 cells. This may play a significant role in designing peptide vaccines. Hence in the present study, we have evaluated the relationship between amino acid pattern and their influence in differentiation of Th1 and Th2 cells. We have used a data set of 320 peptides, whose role has been already established experimentally in the generation of either Th1 or Th2 immune response. Further, characterization was done based on binding affinity, promiscuity, amino acid pattern and binding conformation of peptides. We have observed that distinct amino acids in peptides elicit either Th1 or Th2 immunity. Consequently, this study signifies that alteration in the sequence and type of selected amino acids in the HLA class II binding peptides can modulate the differentiation of Th1 and Th2 cells. Therefore, this study may have an important implication in providing a platform for designing peptide-based vaccine candidates that can trigger desired Th1 or Th2 response.
KeywordsBinding affinity Multiple sequence alignment Promiscuous peptides Th1 and Th2 immune response
Human leukocyte antigen
Multiple sequence alignment
- pTh1 and pTh2
Peptides eliciting Th1 and Th2 immune response, respectively
CD4 T cells play a central role in eliciting and balancing the immune response against invading pathogens. These cells identify antigenic peptides through T cell receptor (TCR) in the context of major histocompatibility complex (MHC) expressed on the surface of antigen-presenting cells (APCs). Interaction of peptide–MHC complex (p-MHC) with TCR and triggering through co-stimulatory molecules lead to the optimum activation of naïve CD4 T cells (Zhu and Paul 2008). CD4 T cells can be subdivided into Th1, Th2, Th17 and Treg subsets (Dubey et al. 1995). Differentiation of naïve CD4 T cells into a particular subset depends on multiple factors like the strength of p-MHC-TCR interaction, delivery and nature of co-stimulatory signals and cytokines milieu (Murray 1998; Constant and Bottomly 1997). Th1 cells mainly produce IFN-γ and TNF-α, and are responsible for cell-mediated immunity. These cells are vital in the elimination of intracellular pathogens like Leishmania major, Mycobacterium tuberculosis, Human immunodeficiency virus, etc. (Reiner and Locksley 1995; North and Jung 2004). On the other hand, Th2 cell chiefly secretes IL-4 and IL-5 (Mosmann and Coffman 1989). Th2 cells are responsible for allergic reactions and provide immunity against extra-cellular pathogens like helminthes such as Nippostrongylus brasiliensis, Schistosoma mansoni, etc. (Pulendran and Artis 2012).
Presence of IL-12 and IL-4 drives naïve CD4 T cells to Th1 and Th2 subsets, respectively (Swain et al. 1990). Apart from the cytokines, other factors may also selectively control the development of Th1 and Th2 cells. It has been reported that p–MHC interactions also specifically influence the differentiation of naïve CD4 T cell to either Th1 or Th2 cells (Kumar et al. 1995). Nevertheless, there is a strong correlation between peptide affinity to TCR and the type of immune response elicited. The affinity and specificity of a particular peptide to its TCR are mainly determined only by specific amino acid residues called anchoring residues. Peptides with strong affinity to TCR induce more Th1 response, whereas those with low affinity prompt Th2 response (Kumar et al. 1995; Iezzi et al. 1999; Boyton and Altmann 2002; Badou et al. 2001). Interestingly, some peptides have been reported to possess intrinsic property to induce Th1 or Th2 response (Whelan et al. 2000; Zhang et al. 2001). Literature suggests that proteins may have intrinsic motifs which may selectively induce Th1 or Th2 response (Guy et al. 2005; Nakajima-Adachi et al. 2012). The influence of altering the sequence of the peptide on elicitation of Th1 and Th2 immune response has been previously described (Kumar et al. 1995; Pfeiffer et al. 1995; Soloway et al. 1991; Liew et al. 1990; Milich et al. 1995). These studies indicated that the peptide–MHC interaction can control not only the magnitude, but also the direction of the functional immune response. Apart from this peptide-MHC interaction, strength of signaling through TCR ligand interaction can also regulate this lineage development (Constant and Bottomly 1997; Iezzi et al. 1999; Boyton and Altmann 2002; Blander et al. 2000). Recently, it has been even shown that strength of T cell stimulation is a vital factor that determines the ability of T cells to become IL-17 producers (Purvis et al. 2010). Moreover, single amino acid-substituted alternate peptide ligands (APL) or antagonists can induce immune deviation in T cell populations, indicating that cytokine profiles can be altered by stimulation with ligands of reduced affinity (Pfeiffer et al. 1995).
Peptides thus alone can determine the type of Th1 and Th2 response (Whelan et al. 2000; Zhang et al. 2001); however, so far no study has been conducted to find the type and position of amino acids in the peptide epitope, which can bias Th1 or Th2 response. Understanding the relationship between amino acids pattern and type of immunity elicited may provide a platform for designing a desired Th1 or Th2 epitope-based vaccine. Hence, in the present study, we employed in silico methods to identify amino acid and pattern in the peptides responsible for skewing immune response towards Th1 or Th2 lineage. We have selected set of peptides inducing either Th1 or Th2 response experimentally. We identified certain conserved anchoring residues responsible for inducing Th1 and Th2 immune response. Strikingly, there were considerable differences in the binding affinity and conformation of these peptides. This finding may have an important application in designing novel peptide vaccine candidates for generating selective Th1 or Th2 immunity.
Materials and methods
Peptide data set
A total data set of 320 peptides was retrieved from Immune epitope database (IEDB), website: http://www.iedb.org (Vita et al. 2010). These peptides were selected based on the ability to induce IFN-γ (for Th1 response) or IL-4 (for Th2 response). This data set included peptide from distinct origins like virus, bacteria, protozoan, allergens, and auto antigens and showed diverse HLA restriction (HLA DR, DP and DQ).
Selected HLA-DR alleles
Human leukocyte antigen class II (HLA-II) alleles predominant in human population, i.e., DRB1_0101, DRB1_0301, DRB1_0401, DRB1_0405, DRB1_0701, DRB1_1101, DRB1_1302, DRB1_1501, and DRB5_0101 were selected for the study.
Programs and databases
This database contains sequences related to T cell epitopes, immunogens, allergens, autoantigens and alloantigens. It also provides data of peptide binding to diverse HLA alleles (Vita et al. 2010).
ProPred is a prediction server that uses matrix-based pocket profiles to determine the binding region of peptide to HLA class II alleles (Singh and Raghava 2001). It also predicts the promiscuous binding regions in antigenic peptides.
This server uses matrix-based method for prediction of HLA class II binders. It predicts binding regions within a peptide and also provides information about HLA alleles anchoring residues (Donnes and Elofsson 2002).
NetMHC2.2 server predicts binding of peptides to various HLA class II alleles using artificial neural networks (ANNs). It also calculates binding affinity of peptides to a particular HLA class II allele (Nielsen et al. 2007).
ClustalW is a multiple sequence alignment program used for clustering protein sequences. It uses neighbor joining method as a default method (Larkin et al. 2007).
Classification and characterization of peptides inducing Th1 and Th2 response
It has been already reported that peptides bind to HLA class II molecules with a common amino acid pattern (Rothbard and Taylor 1988). To identify various peptide patterns, we have selectively chosen a set of peptides which are experimentally proven to be inducers of either IFN-γ or IL-4, which are the hallmark cytokines for Th1 and Th2 response, respectively. Henceforth, such peptides that induce Th1 and Th2 response will be referred to as ‘pTh1’ and ‘pTh2’, respectively, in the text. Within the selected peptides, binding regions were predicted using ProPred and SVMHC prediction servers which have relatively high accuracy (Gowthaman and Agrewala 2008; Wang et al. 2008). We have considered consensus binding by combining the results of these two servers and classified peptides as binders and non-binders. The binding affinity of these peptides to a particular HLA class II allele was calculated using NetMHC 2.2 server based on IC50 values. These were classified as strong binders (IC50 ≤ 50nM) and weak binders (50nM ≤ IC50 ≤ 500nM). The peptides having IC50 value of >500nM were considered as non-binders. Peptides exhibiting binding to 3 or more than 3 of selected alleles were considered as promiscuous peptides. Multiple sequence alignment of binding regions restricted to particular HLA allele was performed using ClustalW to identify patterns and conserved amino acid sequences. The results were then analyzed using Jalview considering the binding affinity of peptides to HLA alleles, type of anchoring residues and nature of amino acids present in binding region (Waterhouse et al. 2009).
Molecular docking of model peptides with MHC
Peptide sequences known to induce Th1 or Th2 response were selected from literature (Agrewala and Wilkinson 1998). The initial peptide structure was generated using “build and edit protein” program of Accelrys Discovery studio (Accelrys software Inc. 2007). The conformations were then submitted to energy minimization followed by molecular dynamics simulation for 20 nano seconds in an implicit solvent model using GROMACS (Hess et al. 2008). The refined peptide conformations were used as starting ligand molecule and docked with HLA-DR2 crystal structure using AutoDock (Huey et al. 2007). For this, we have chosen crystal structure of HLA-DR2 bounded with a peptide from human myelin basic protein (PDBID: 1BX2). The existing peptide in PDBID: 1BX2 was deleted and docking studies were performed with p91–110 and p21–40 peptides. Out of the ten binding poses predicted by AutoDock, we selected optimal binding pose based on the binding free energy. Results were analyzed using AutoDock Tools (Sanner 1999) and PyMol (DeLano 2002).
pTh1 are strong binders and binding affinity increases with addition of flanking residues
Predominance of hydrophobic amino acids in the peptide at the position 4 and 5 skews immune response towards Th1 and Th2, respectively
Consensus sequences derived from multiple sequence alignment of strong binders to a particular HLA allele
(a) Thl consensus sequence
(b) Th2 consensus sequence
No strong binders were detected
Molecular docking reveals conformational differences in pTh1 and pTh2-MHC complex
T helper cells mediate adaptive immunity against several infectious agents. Th1 and Th2 cells cross regulate the function of each other (de Montmollin et al. 2009). In general, immune response to whole antigen is attributed by the presence of both Th1 and Th2 epitopes (Basu et al. 2005). Hence, epitope-based vaccines have potential advantage of selecting immunodominant epitopes that can incline immune response towards Th1 or Th2 cells. Further, undesired autoreactive and immunosuppressive pathogenic epitopes can be eliminated. Thus, a profound understanding of underlying mechanisms of peptide and MHC allele interactions can provide a skeleton to develop synthetic peptides which can be attractive vaccine candidates.
Many factors can influence the differentiation of naïve CD4 cells into a Th1 or Th2 cell, and these factors need not be mutually exclusive. There may well be other dominant influences but the impact of epitope specificity on Th phenotype development should not be underestimated. The affinity of the p-MHC-TCR interaction is thought to be an essential factor for determining the differentiation of CD4 T cells into Th1 or Th2 phenotypes.
In the present study, we have tried to establish a correlation between amino acid pattern in a specified peptide and elicitation of either Th1 or Th2 response. Following major findings have emerged from the study: (1) majority of the pTh1 were strong binders, whereas pTh2 were weak binders; (2) hydrophobic amino acids were conserved at position 1 for both pTh1 and pTh2; (3) position 3 of promiscuous pTh1 was predominantly occupied by charged amino acids whereas by hydrophobic amino acids in pTh2; (4) 100 % hydrophobicity was observed at position 4 of pTh1, whereas at position 5 of pTh2; (5) at position 4, leucine was the most predominant hydrophobic amino acid for pTh1 but absent in pTh2; (6) nature of amino acids at positions 3, 4, 5 was very crucial for distinct Th1 and Th2 responses; (7) there were conformational differences in the binding nature of pTh1 and pTh2 to MHC molecules.
Although all the peptides have equal length of binding region (9 amino acid), strong affinity was associated with peptides spanning 20 amino acids in case of both pTh1 and pTh2. These results were further supported by several reports, which showed the importance of peptide flanking residues (PFRs) present outside the MHC anchor region in determining the binding affinity (Arnold et al. 2002; O’Brien et al. 2008). Although the peptide data set has been experimentally validated and in most cases by overlapping peptide libraries, exclusively identifying the true ligands that would have been isolated from the cells after peptide dissociation from its cognate MHC would be preferable. Still we employed the in silico analysis to predict the binder nonamer peptide out of the sequence and analyze its binding efficiency with sequential addition of flanking amino acid residues. This approach of computational prediction has been shown to be in harmony with experimental data (Zhang et al. 2012; Tongchusak et al. 2008; Pereira et al. 2011) and apart from some obvious advantages needs some refinement to overcome the limitations (Brusic et al. 2004; Lafuente and Reche 2009; Liao and Arthur 2011).
Further, we investigated amino acids responsible for promiscuity of peptides. Notably, we have observed that position 3 is critical for promiscuous binding of both pTh1 and pTh2. In case of pTh1, promiscuity was associated with occurrence of charged/polar amino acids, whereas non-promiscuity was attributed to hydrophobic amino acids present at position 3. In contrast, reverse order was observed in the case of pTh2. This observation signifies that the nature and sequence of amino acids in the peptide vaccine can determine the Th1 and Th2 immunity. It is well established that different antigens have tendency to specifically trigger either Th1 or Th2 response (Whelan et al. 2000; Zhang et al. 2001). Hydrophobic amino acids present at positions 1, 7 and 9 were found to be conserved in both Th1 and Th2 eliciting peptides, indicating importance of these positions in MHC–peptide binding, irrespective of type of immune response. However, positions 3, 4 and 5 were differentially conserved in pTh1 and pTh2. This observation was further validated with additional HLA alleles, indicating that these positions may have important bearing in designing a peptide vaccine to specifically elicit Th1 or Th2 response.
The present study also denotes that a specific amino acid pattern can favor immune response towards either Th1 or Th2 immunity. Further, it establishes a strong correlation between nature, position and binding affinity of the amino acids that ultimately dictates the desired immune response. Hence a particular position of an amino acid may be quite crucial in a peptide to differentially regulate the Th1/Th2 immunity. Furthermore, our study designates that manipulation of the position and the type of amino acids in a peptide vaccine can bias immune response towards Th1 and avoiding the Th2 lineage, or vice versa. It is important to mention here that Th1 and Th2 reciprocally regulate the generation and function of each other (de Montmollin et al. 2009). Therefore, the presence of one phenotype may impair the function of other, hereby rendering vaccine ineffective. We suggest that it is very important to carefully choose the physicochemical properties and position of amino acids, while designing a peptide vaccine. Further, in a peptide vaccine, amino acids and their patterns should be mapped to identify their role in Th1 or Th2 response. The nature of majority of antigens/peptides derived from pathogens or host tissues still largely remains unidentified for their vaccination potential for eliciting Th1 or Th2 immunity. In this regards, our current study contributes to provide information in successfully designing a vaccine employing in silico methods. This information will ultimately help in profitably synthesizing a peptide for mass vaccination in a cost-effective manner (Black et al. 2010).
In essence, the knowledge about position, properties and binding affinity of amino acids provided in the current study will be extremely beneficial in designing peptide-based vaccines, which may have potential in taming and tuning the immune system to evoke either Th1 or Th2 response to combat and eliminate intracellular and extracellular pathogens, respectively.
The authors are thankful to Dr. Balvinder Singh, CSIR-Institute of Microbial Technology, Chandigarh for valuable suggestions and Dr. Uthaman Gowthaman, Yale School of Medicine, New Haven, USA for peptide data collection. The authors also thank Council of Scientific and Industrial Research (CSIR), India for financial support. SBC, KM, PKR, SM, JAS are recipients of fellowship of CSIR, and MA of UGC.
Conflict of interest
The authors have no financial conflict of interest.