Structure, interaction and design studies involving regulatory peptides and proteins

 
Principal Investigator :  Dinakar M Salunke

Project Associates / Assistants
Chitra Rani
Sanjeev Saxena
Anupriya Agarwal

Technical Officer
Sushma Nagpal

Ph D Students
Manisha Goel
Gaurav Sahani
Dhruv K Sethi
Lavanya Krishnan
Veenu Arora

Collaborators
Kanwal J Kaur
Satyajit Rath
KVS Rao, ICGEB, New Delhi
MJ Swamy, Univ Hyderabad, Hyderabad
J Nagaraju, CDFD, Hyderabad

The structural aspects of molecular recognition and its applications in analyzing the mechanisms associated with specific regulatory events and in rational molecular design is the theme of research. The objectives of the project are (i) understanding the protein architecture, (ii) analysis of the structural principles of molecular recognition and mimicry, (iii) structural biology of various regulatory events and (iv) rational molecular design studies based on the above.

As reported in the previous report, anti-mannopyranoside monoclonal antibodies that could also recognize the carbohydrate mimicking peptides have been generated. A total of about 50 clones were found to be positive for recognizing the mannopyranoside antigen. Both, the clones showing cross-reactivity to the 12mer peptide or recognizing only the carbohydrate antigen were present, though the latter were only about 10% of the total anti-mannopyranoside clone population. A panel of four IgG antibodies named 1H7, 2D10, 1H11 and 5F5 were selected for further studies. Of these antibodies, 2D10, 1H11 and 5F5 show crossreactivity to the peptide while the 1H7 antibody shows binding to the mannopyranoside ligand only. Among the antibodies, 1H7, 1H11 and 5F5, are of IgG1 isotype while the 2D10 displays IgG2b isotype.

The binding kinetics of these antibodies for both the ligands were studied by surface plasmon resonance based measurements at ambient temperature. Conjugates of the mannopyranoside and 12mer peptide with BSA were used as immobilized ligands and each one of the antibodies were studied for their binding to the mannopyranoside and the peptide ligands. The association and the dissociation rate constants were determined and used for deriving the dissociation constants of all antibodies for their respective ligands. Among the various carbohydrate-binding antibodies, 1H7 has the highest affinity (KD=0.2mM) for the mannopyranoside. 2D10 shows slightly higher affinity of KD=0.9uM for the carbohydrate as compared to that for the peptide ligand (KD=8.8mM). Both 1H11 and 5F5 show highly comparable affinity for mannopyranoside although 1H11 shows higher affinity for the peptide than for the mannopyranoside.

The four different antibodies not only show different ligand affinities for the peptide and carbohydrate ligands but also show differences in the kinetics of binding to the two ligands. We therefore wanted to further explore the diversity among these antibodies at the gene level. To determine the germline origin and relatedness among the four antibodies, the nucleotide sequencing for both the light and the heavy chains of these antibody clones were carried out. The nucleotide sequences thus obtained were used to search for the germline origins of the antibodies from the database using Ig-BLAST.

The nucleotide sequence of the light chain of the 1H7 antibody shows homology (97%) to ce9, the V region germline gene sequence and JK1, the J region germline sequence. 2D10 antibody is highly homologous (97%) to the 21-12 variable germline gene sequence and JK2. The light chain of 1H11 antibody shows identity to the 19-15 (99%), V region germline gene sequence and JK4. The 5F5 antibody shares homology (99%) with the bb1, V region germline gene sequence and JK2 sequence. The heavy chains of all the four antibodies originate from the various gene members of the J558 family. Heavy chain of 1H7 antibody is derived from the J558.17 showing sequence homology (97%) to this V region germline gene sequence and to D element, DSP2.2, and JH2 germline sequence. 2D10 heavy chain has high sequence identity (93%) to the VH gene J558.12 and also shows homology to D element gene DFL16.1 and JH2. 1H11 shows sequence homology (95%) to the J558.44 germline sequence and to DSP2.9 and JH3 gene elements. The heavy chain of 5F5 antibody shows homology to J558.8 (98%) germline member and to DQ52, the D-element sequence and JH4 sequence. The combination of different DH and JH elements to the various members of the J558 VH chain as seen in the case of four anti-mannopyranoside antibodies suggests that the four clones have independent B-cell origin. The variability seen in the light chain gene usage reinforces this view, that the four clones are not the product of somatic mutations from a given B-cell. Thus, a variety of gene arrangements can probably give rise to the mannopyranoside recognizing paratopes but these may differ in their recognition specificity for both, the carbohydrate as well as the 12mer peptide, epitopes.

The YPY motif of the crossreacting peptide has been implicated in mimicking a variety of other carbohydrates, we therefore wanted to analyze if the various antibodies recognizing the 12mer peptide would cross-react with any other carbohydrate ligands. Since these antibodies show variable affinities for the mannopyranoside ligands, we also wanted to see if they differed in their specificity profiles as well. For this, competition based assays were carried out where various sugars in solution were made to compete for binding to the antibody as against the immobilized mannopyranoside or the peptide ligand. It is evident that the three mimicry recognizing peptides do not show much differentiation between the three sugars, whereas 1H7, the only carbohydrate recognizing antibody, shows much higher affinity for the mannopyranoside as compared to glucopyranoside and lactose. It is interesting to note that polyclonal antibodies have earlier been shown to recognize glucopyranoside with lower affinity than mannopyranoside, while lactose shows minimal competition and behaves as a non-specific sugar. Similar degenerate specificity is shown for various sugars by the three peptide recognizing antibodies when sugars were competing against the immobilized 12mer peptide. In contrast to the polyclonal antibody cross-reactivity data where the sugars were competing against the immobilized mannopyranoside ligands, the polyclonal data in case of immobilized peptide behaves exactly similar to the three antibodies.

As part of another on-going activity concerning structural basis of carbohydrate mimicry, we had earlier determined the crystal structure of concanavalin A bound to a porphyrin analog, meso-tetrasulphonatophenylporphyrin (H2TPPS), and established the mimicry of mannopyranoside and porphyrin in terms of the interactions with the lectin. We have now determined and refined the crystal structure of H2TPPS with jacalin at 1.8Å resolution and in the process of refining the structure in complex with peanut lectin at 2.9Å resolution.

In the crystal structure of H2TPPS-jacalin complex, a pair of stacked porphyrins interact with two jacalin molecules, both of which belong to two independent jacalin tetramers. This dimer of porphyrins with a monomer of jacalin forms the asymmetric unit of the present crystal structure. Thus, a dimer of porphyrin molecules is sandwiched between two symmetry related jacalin monomers. This leads to extensive non-covalent cross-linking of the jacalin molecules in the crystal structure. The two porphyrins do not interact at the similar site with the jacalin molecules. One of them exhibits specific interactions at the carbohydrate binding site on jacalin, while the other interaction is largely resulting while facilitating crystal contacts. The binding of the porphyrin molecule does not seem to cause any major changes in the backbone conformation of the jacalin molecule.

The two porphyrin molecules are staggered with respect to each other such that the porphine cores of the two are not eclipsed but seem to be translated with respect to each other, unlike in case of H2TPPS-ConA complex. Seven water molecules are also found within the van der Waals distance of this porphyrin molecule. The nitrogen atom N8 of the porphine ring forms a hydrogen bond with a water molecule. Thus the nature of interactions involving H2TPPS with jacalin are substantially different from those with ConA although both cases show water mediated bonding with the protein. The preliminary studies in case of peanut lectin and H2TPPS suggests yet another mode of binding of H2TPPS with this lectin.

Publications

Original peer-reviewed articles

1.     Nair DT, Kaur KJ, Singh K, Mukherjee P, Rajagopal D, George A, Bal V, Rath S, Rao KV and Salunke DM (2003) Mimicry of native peptide antigens by the corresponding retro-inverso analogs is dependent on their intrinsic structure and interaction propensities. J Immunol 70:1362-1373.

2.     Nagpal S, Kaur KJ, Jain D and Salunke DM (2002) Plasticity in structure and interactions is critical for the action of indolicidin, an antibacterial peptide of innate immune origin. Protein Sci 11:2158-2167.

3.     Chakraborty S, Chakraborty N, Jain D, Salunke DM and Datta A (2002) Active site geometry of oxalate decarboxylase from Flammulina velutipes: Role of histidine coordinated manganese in substrate recognition. Protein Sci 11:2138-2147.

4.     Manivel V, Bayiroglu F, Siddiqui Z, Salunke DM and Rao KVS (2002) The primary antibody repertoire represents a linked network of degenerate antigen-specificities. J Immunol 169:888‑897.

5.     Gangatirkar P, Gangadharan S, Narendranath A, Nagpal S, Salunke DM and Karande AA (2002) Monoclonal antibodies to gonadotropin-releasing hormone (GnRH) inhibit binding of the hormone to its receptor. Hybrid Hybridomics 21:281-286.

6.     *Hemachand T, Gopalakrishnan B, Salunke DM, Totey SM and Shaha C (2002) Sperm plasma membrane associated glutathione S- transferases as gamete recognition molecules. J Cell Sci 115:2053-2065 (*in press last year, since published).