| dc.description.abstract | Structural, Functional, and Immunological Characterization of Recombinant Riboflavin Carrier Protein Expressed in Baculovirus System and Molecular Modeling of GnRH-Specific Single Chain Fragment Variable
This thesis is an account of studies carried out on the cloning, expression, and characterization of two recombinant molecules: riboflavin carrier protein and single chain antibodies.
Chapter 1 is the introduction to the first part of the thesis and describes water-soluble and fat-soluble vitamins, with special emphasis on the role of flavin nucleotides, which are vitamin B2 derivatives. Vitamin carrier proteins, which help in the transport of various water-soluble vitamins, are listed, of which information on the vitamin B2 (riboflavin) carrier protein is elaborated. This chapter describes the two domains of chicken riboflavin carrier protein (cRCP), the ligand binding as well as the receptor binding domains, which function in the binding and transport of riboflavin, in turn resulting in the deposition of the protein into the hen egg. Structural and functional differences exist in the RCP present in the egg yolk, egg white, as well as the plasma of egg-laying hens. The role of post-translational modifications and their functions, as well as the structural characterization, including the ligand binding pocket of the protein, are discussed. Immunological characterization of cRCP using polyclonal as well as monoclonal antibodies has been reported in literature. Recombinant RCP expressed in E. coli has been studied with respect to the ability of this protein to fold and maintain its function of ligand binding and immunological properties.
Chicken Riboflavin Carrier Protein (cRCP) is a 37 kDa phosphoglycoprotein, first identified in the avian system. This protein has dual loci of synthesis, i.e., liver and oviduct, depositing it in the egg yolk and egg white, respectively. Hence, the proteins have identical primary structures but differ in their glycan composition. The cRCP is a highly compact protein with 9 disulfide bonds, of which 8 are found in the ligand binding domain and a single disulfide bond bridges the ligand and receptor binding domains. Attempts to obtain recombinant RCP showed that the protein expressed in E. coli was obtained in the inclusion bodies. Refolding experiments on this protein yielded only 23% of the total protein. Of this fraction, only 25% showed riboflavin binding. It was then proposed that glycosylation could play an important role in the proper folding of the protein and hence E. coli-expressed protein is found in the inclusion bodies.
Chapter 2 describes the expression of cRCP protein in the baculovirus system, as this system permits the post-translational modifications of recombinant proteins. As the ligand binding domain is very compact in its structure and has minimal interaction with the receptor binding domain, the ability of the two domains to fold independently and maintain their respective functions was analyzed. Towards this, the full-length protein (rFL), the ligand binding domain (D1), and the receptor binding domains were cloned and overexpressed in the baculovirus system. The recombinant proteins, rFL and D1 domain, were secreted into the culture supernatant, indicating a proper folding of these proteins. The D2 domain was found in minute amounts in the intracellular fraction, indicating the inability of D2 to fold independently of the D1 domain. The D2 domain is probably degraded or unstable in the absence of the D1 domain.
Chapter 3 describes the characterization of the recombinant proteins in comparison to the native protein. The glycosylation of the rFL and D1 proteins and phosphorylation status of the serine residues in the rFL protein were determined. The riboflavin binding property of the rFL and D1 proteins was analyzed by the standard riboflavin fluorescence quenching and 14C-riboflavin binding assays used for the native protein. To determine conformational differences, tryptophan fluorescence quenching studies were carried out, and far-UV CD spectra were measured for rFL protein in comparison with the native protein. Immunological characterization was carried out with conformation-specific and sequence-specific monoclonal antibodies as well as polyclonal antibodies in inhibition studies using enzyme-linked immunosorbent assay (ELISA) as well as radioimmunoassay (RIA).
It was demonstrated that the expressed rFL protein is glycosylated as well as phosphorylated. The rFL protein binds riboflavin with a stoichiometry of 1:1, similar to that of the native protein. Riboflavin fluorescence quenching ability of the rFL protein indicates a similar stacking of riboflavin with residues tyrosine 75 and tryptophan 156 as in the native protein. However, incomplete tryptophan fluorescence quenching in the holoform of the rFL protein indicates that the microenvironment of the tryptophans near the ligand binding site might be slightly altered in the rFL protein. The secondary structure analysis with far-UV CD showed that the rFL protein also has a similar structure with the negative MRE values at 208 nm and 222 nm indicating the -helix pattern as well as minor differences seen in the apo and holo forms reported earlier for the native protein. The immunological characterization revealed that the conformation of the rFL protein was similar to that of the native protein except for one epitope. This difference in the epitope region does not affect the ligand binding by the rFL protein. Finally, our analysis of the soluble form of baculovirus-expressed cRCP indicates a possible role of glycosylation in attaining native conformation of cRCP.
Secretion of the ligand binding domain into the culture supernatant indicated its ability to fold independently of the D2 domain. The ligand binding ability was low as compared to that of the rFL and native proteins in the 14C-riboflavin binding assay. However, the D1 protein was unable to quench riboflavin fluorescence. The results of immunological assays showed that except for two mAbs 6A4D7 and 5A2E6, all the other epitope regions are unaltered for binding. These results indicate that the recombinant D1 protein has a similar fold as that of the D1 domain present in the recombinant full-length protein. However, from the ligand binding assays, it can be concluded that the riboflavin binding pocket is not folded completely to near native state, and the orientation of the tyrosine 75 and tryptophan 156 residues in the ligand binding pocket are not maintained in the recombinant D1 protein. It is evident from our study that the D2 domain is essential for proper folding and formation of the ligand binding pocket of the D1 domain. On the other hand, the D2 domain was unable to fold or is unstable in the absence of D1 domain. Overall, the two domains of cRCP are required for mutual stability, proper folding, and function.
The second part of the thesis deals with the Construction and Computational Analysis of Single Chain Fragment Variable (ScFv) of Anti-GnRH Monoclonal Antibodies
Gonadotropin Releasing Hormone (GnRH) plays a pivotal role in the pituitary-gonadal axis. A single gene located on chromosome 8 of the human genome encodes the peptide. The pre-pro-GnRH is a 92 amino acid protein, which upon post-translational modification forms the decapeptide, GnRH. Additionally, post-translational modifications of the N and C-termini of the peptide to pyroglutamate and glycinamide, respectively, are known to be crucial for receptor binding and activity of GnRH. GnRH, secreted in a pulsatile fashion from the hypothalamus, acts on the pituitary gonadotropes, which in turn results in the normal functioning of the hypothalamic-pituitary-gonadal axis and the reproductive system. A defective functioning of GnRH leads to reproductive disorders, which are treated by using GnRH analogs. The design of better analogs depends on the knowledge of the structure of GnRH. The structure of GnRH is not deciphered yet owing to its highly flexible conformation as evident from the NMR studies. The NMR studies in the presence of lipids in solution revealed a p-turn conformation of GnRH. Computational analysis from two different studies proposed the conformation of GnRH to be a type II -turn with slight differences in the overall conformations. An alternative to crystallization is co-crystallization of GnRH with its receptor. GnRH receptor being a membrane-associated protein, co-crystallization is technically difficult. The use of monoclonal antibodies that bind GnRH in an active conformation can be used for co-crystallization.
With the above in mind, anti-GnRH monoclonal antibodies (mAbs) were raised in our laboratory. Two of these mAbs, namely F1D3C5 and E2D2H12, were characterized for their affinity to bind the antigen and the epitopes for these mAbs were mapped. The ability of these antibodies to bind GnRH was analyzed on primary pituitary cultures and gonadotrope cell lines. The mAb F1D3C5 showed complete inhibition of GnRH activity, while E2D2H12 showed only 50% inhibition, indicating that the mAb F1D3C5 probably recognizes GnRH in its active conformation in vitro. Co-crystallization studies with the Fab fragments of these two mAbs with GnRH were not successful due to the non-homogenous crystals obtained
In Chapter 4, the role of these monoclonal antibodies (mAbs) in recognizing GnRH in vivo was analyzed using a mouse model system. The F1D3C5 mAb arrested the estrous cycle in the mice, while mAb E2D2H12 had no effect. For use in co-crystallization studies, the construction of the single-chain fragment variable (ScFv) of the two mAbs was attempted. An ScFv is the minimal part of an antibody that binds the antigens. An ScFv contains only the variable regions of the heavy and light chains while maintaining the paratope region intact. The two domains in the ScFv are connected via a linker region. The ScFvs of the two mAbs can be used to determine the differences in the conformations of GnRH recognized by them. The periplasmic expression of E2D2H12 ScFv in the AD494 E. coli strain was standardized. The purified ScFv was analyzed for binding to GnRH in inhibition ELISA.
From the generated ScFvs of the two mAbs, the sequence of the variable regions was determined, and the derived amino acid sequence of the two ScFvs was used for computational analysis. The complementarity-determining regions (CDRs) of the two mAbs were identified, compared, and three-dimensional models were generated for the ScFvs of the two mAbs. The GnRH peptide was docked into the antigen-binding region of the modeled ScFvs, and the interactions between the GnRH and the ScFv were analyzed. Comparison of the docked ScFv-GnRH complexes of E2D2H12 and F1D3C5 shows the probability of F1D3C5 ScFv forming a more stable complex with GnRH as compared to the complex of E2D2H12 ScFv.
Taken together, our study demonstrates that mAb F1D3C5 recognizes GnRH in the active conformation, and the generated ScFv can be used for co-crystallization in order to determine the structure of the decapeptide, GnRH. | |
