Bell Eapen MD, PhD.

Bringing Digital health & Gen AI research to life!

Rational Drug Design

Rational drug design is at present based on docking studies. Docking predicts the binding orientation of the candidate drug molecule to its target protein that in turn predicts its affinity and activity. Docking algorithms are constantly being improved to make the predictions more reliable. However the pharmaceutical relevance of docking studies are constrained by the inability to predict the extent to which the drug molecules actually reach the target protein. Besides predicted affinity and activity may not be demonstrable in vitro. Hence most promising molecules on the drawing board fail to impress during clinical trials.

English: Drug overdose
English: Drug overdose (Photo credit: Wikipedia)

But this is about to change! Researchers at Karolinska Institutet in Sweden have developed the first method for directly measuring the extent to which drugs reach their targets in the cell. They have developed a new tool called CETSA (Cellular Thermal Shift Assay), which utilise the concept that target proteins usually get stabilised when drug molecules bind.

The tool may also be useful in individualised treatment of certain conditions, especially malignancies based on the protein profile. Hope this technology does take off one day  and introduce a paradigm shift in drug designing protocols.


1. Karolinska Institutet (2013, July 5). Technological breakthrough paves the way for better drugs. ScienceDaily. Retrieved July 16, 2013, from http://www.sciencedaily.comĀ­ /releases/2013/07/130705101541.htm

Delayed-Type hypersensitivity to latex: Computational prediction of MHC class II epitopes on latex allergens

This article was rejected by several journals. Hence I feel the concept is fundamentally wrong. I have posted the full article here. The referee comment from the last journal is also included.

  • The object of the study is laudable. However, the data presented here is too preliminary for publication.
  • Class II MHC molecules bind short peptides of 12 to 20 residues long. The concept of MHC binding to whole antigen is grossly not acceptable as given in Table 1.
  • Moreover, studies of this nature require experimental verification for publication.
  • It is strongly advised to look at MHCBN, MHCPEP and SYFPETHI for the type of peptides MHC molecules bind.
  • The design of the study should be narrowed down to a single antigen with overlapping peptides to multiple class 2 alleles for a reasonable argument.

Key words
Natural Rubber Latex, MHC Class II.

Delayed type hypersensitivity to natural rubber latex is rare compared to IgE mediated immediate reactions. Binding of allergens to MHC Class II is a crucial step in the presentation of antigens to CD4+ T Cells responsible for delayed reactions. Computational prediction of MHC class II epitopes on thirteen known latex allergens using SMM-align method revealed strong binding with several alleles. This shows that latex allergens are capable of initiating delayed type hypersensitivity in susceptible individuals.

Natural rubber latex (NRL) derived from the rubber tree (Hevea brasiliensis) is an important allergen causing mostly IgE mediated immediate reactions like urticaria, angioedema and asthma. [1] The existence of a true NRL induced delayed type hypersensitivity (type IV) is still debated. Delayed type hypersensitivity reactions are attributed to additives like thiurams and carbamates. [2] The coexistence of the type I and type IV patterns and entities like protein contact dermatitis [3] further complicate the issue. However incidence of true type IV hypersensitivities ranging from 1.2 to 6% has been reported following the introduction of an NRL containing patch test series. [4]

During the sensitization phase of contact hypersensitivity, epidermal Langerhans cells internalise the allergen, migrate to the regional lymph node and present the processed allergen bound to MHC Class II molecules to CD4 T cells. MHC Class II binding site consists of a groove open at both ends and several pockets binding peptides with 13 to 25 residues. [5] MHC molecules exhibit high degree of genetic variation which enables them to bind variety of peptides.

As IgE from latex-sensitive patients bind to several proteins, there is still no consensus on which latex proteins are antigenic. However information on thirteen officially accepted latex allergens designated serially from Hev b 1 to 13 is available from the online repository and are summarized in Table 1. [6,[7]

We tried to computationally predict the MHC class II epitopes on these antigens using SMM-align method. SMM-align is a novel stabilization matrix alignment method based on scoring matrices that evaluate the contribution to binding of different residues in a peptide. [8] T cells are likely to recognize MHC binding peptides and initiate a cellular response. [9]

The protein sequences of all thirteen allergens were downloaded from and were added to a single file. This file was used for MHC Class II epitope prediction with the publicly available SMM-align server using default parameters. The server returns IC50 prediction scores (concentration of competing ligand which displaces 50% of the specific ligand) covering fourteen HLA DR alleles for each nanomer within the query peptide. Allergens with at least one nanomer with IC50 value less than fifty were considered as binders.

Strong binding to one or more alleles was shown by all allergens except Hev b 5. Hev b 2 was the most promiscuous allergen binding several alleles. DRB1*0101 was the most commonly bound allele. Results are summarized in Table 1.

NRL is an allergen causing mostly IgE mediated immediate type hypersensitivity. Though there have been significant advances in molecular biology of latex allergens, their ability to initiate delayed-type hypersensitivity has been subject to debate. Computational prediction of MHC class II binding regions on latex allergens adds credence to reports of type IV contact hypersensitivity to NRL. [4]

Prediction of MHC binding peptides is a commonly employed step in the identification of T cell epitopes. [9] MHC epitope related data is available from several databases like SYFPEITHI, [10] MHCBN [11] and IEDB [12]. Several algorithms like ARB [13] and SMM-align [8] are available for prediction of MHC class II binding peptides. Algorithms for scanning promiscuous peptides that can bind multiple MHC class II molecules are also available. [14] Promiscuous peptides are important for vaccine development and immunotherapy. [5] In our study Hev b 2 bound to six out of fourteen tested HLA DR alleles.

Most of the characterized latex allergens contain segments which can strongly bind MHC class II alleles and are capable of initiating delayed type hypersensitivity without a hapten in susceptible individuals. Certain promiscuous allergens like Hev b 2 bind to several MHC alleles making them ideal candidates for immunotherapy. [15] However computational prediction of MHC binding has limited accuracy and clinical validation is essential.


1 Reunala T et al.,. Latex allergy and skin. Curr Opin Allergy Clin Immunol 4: 397-401 (2004) [PMID: 15349039]
2 Miri S et al.,. Prevalence of type I allergy to natural rubber latex and type IV allergy to latex and rubber additives in operating room staff with glove-related symptoms. Allergy Asthma Proc 28: 557-63 (2007) [PMID: 18034975]
3 Doutre MS. Occupational contact urticaria and protein contact dermatitis. Eur J Dermatol 15: 419-24 (2005) [PMID: 16280292]
4 Sommer S et al.,. Type IV hypersensitivity reactions to natural rubber latex: results of a multicentre study. Br J Dermatol 146: 114-7 (2002) [PMID: 11841376]
5 Wang P et al.,. A systematic assessment of MHC class II peptide binding predictions and evaluation of a consensus approach. PLoS Comput Biol 4: e1000048 (2008) [PMID: 18389056]
6 Breiteneder H et al.,. Molecular and immunological characteristics of latex allergens. Int Arch Allergy Immunol 116: 83-92 (1998) [PMID: 9652300]
7 Wagner S et al.,. Hevea brasiliensis latex allergens: current panel and clinical relevance. Int Arch Allergy Immunol 136: 90-7 (2005) [PMID: 15627782]
8 Nielsen M et al.,. Prediction of MHC class II binding affinity using SMM-align, a novel stabilization matrix alignment method. BMC Bioinformatics 8: 238 (2007) [PMID: 17608956]
9 Sette A et al.,. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J Immunol 153: 5586-92 (1994) [PMID: 7527444]
10 Schuler MM et al.,. SYFPEITHI: database for searching and T-cell epitope prediction. Methods Mol Biol 409: 75-93 (2007) [PMID: 18449993]
11 Bhasin M et al.,. MHCBN: a comprehensive database of MHC binding and non-binding peptides. Bioinformatics 19: 665-6 (2003) [PMID: 12651731]
12 Peters B et al.,. The immune epitope database and analysis resource: from vision to blueprint. PLoS Biol 3: e91 (2005) [PMID: 15760272]
13 Bui HH et al.,. Automated generation and evaluation of specific MHC binding predictive tools: ARB matrix applications. Immunogenetics 57: 304-14 (2005) [PMID: 15868141]
14 Zhang GL et al.,. MULTIPRED: a computational system for prediction of promiscuous HLA binding peptides. Nucleic Acids Res 33: W172-9 (2005) [PMID: 15980449]
15 Tabar AI et al.,. Specific immunotherapy with standardized latex extract versus placebo in latex-allergic patients. Int Arch Allergy Immunol 141: 369-76 (2006) [PMID: 16943675]

Protein-Protein Docking Problem.

I am looking for a solution to the following problem. Any insight will be greatly appreciated.

A membrane receptor (A) has an extracellular domain (AE), transmembrane domain (AT) and intracellular domain (AI).

A bacteria (B) binds to (AE) leading to dimerization of (A) at (AT) and subsequent downstream signaling through (AI).

(A) has no known natural ligands.

(A) has one known inhibitor (I) binding to (AE)

The structure of (AE) bound to (I) is available from PDB.

How do we identify which protein in (B) binds to (AE).

The obvious solution is to dock all proteins with known structures in (B) to all known pockets in AE. Any better solutions?