Recombinant Growth hormone (GH) is now used widely for clinical and agricultural purposes. The clinical applications of human GH include growth stimulation for children of short stature, amelioration of catabolic states in GH-deficient adults, treatment of Crohn's disease, enhancement of bone fracture healing, and treatment of dilated cardiomyopathy. In livestock industries, porcine GH (pGH) and bovine GH (bGH) are used to increase lean meat production, improve feed conversion efficiency and to increase milk yields in dairy cattle. Not surprisingly, there has been great interest in the precise molecular mechanism of action of GH, with the final goal of producing these effects with small molecule mimics and higher potency analogues.
In this thesis, I sought firstly to gain insight into how a monoclonal antibody (MAb 263) can activate the GH receptor to stimulate a growth response independent of GH. Using a novel random PCR mutagenesis approach together with expression screening in yeast, I have mapped precisely the conformational epitope for this MAb on to the crystal structure of the GHR extra cellular domain (ECD). A library of 5200 clones of rabbit GH receptor ECD mutants was created, then screened for reactivity with 263 by immunoblot. This identified clones unable to bind MAb 263, whether the result of mutation of epitope residues or loss of ECD expression. In order to eliminate the latter, the library was rescreened with an anti-carboxyterminal-tag antibody to verify complete ECD expression. Sequencing for clones that expressed complete ECD but were not MAb 263 positive identified 20 epitope residues distributed in a discontinuous manner throughout the ECD. The major part of the epitope, as revealed after mapping onto the crystal structure model of the ECD molecule, was located on the side and upper portion of domain 1, particularly within the D-E strand disulfide loop 79- 96.
As this MAb is known to act as an agonist for the GH receptor in vitro and in vivo, it was important to determine if this agonist ability requires an intact bivalent antibody. Therefore, monovalent Fab fragments of MAb 263 were prepared to high purity and used to show that although they could compete for GH binding, they were unable to elicit agonist activity. However, Fab fragments could recover agonism if cross-linked with anti- Fab antibody. This showed that, as for GH, two receptors must be dimerized in correct orientation for receptor activation.
Molecular dynamics of a bivalent antibody of the same isotype as MAb 263 was used to dock the crystal structure of this antibody to its 1528-A2 epitope on the receptor ECD and to visualize the likely consequences of MAb binding. The minimized model enables the antibody to grasp two receptors in a pincer-like movement from opposite sides, facilitating alignment of the receptor dimerization domains in a manner similar to, but not identical with GH.
With the intention of developing a growth enhancing immune response, I undertook an immunization study in 48 GH-deficient dwarf rats using the sequence of the key epitope residues coupled to hemocyanin or bovine serum albumin. However, the synthetised epitope peptide did not result in any significant change in body growth although peptide antibodies were generated. This was most likely because the epitope residues are distributed in a discontinuous and conformation sensitive manner.
Following this, I sought to develop higher potency analogues of pGH and to further understand the relationship between GH binding affinity and ability to stimulate a biological response. GH has two nonidentical binding sites that bind two receptors in an ordered fashion, first through site 1 and then through site 2, resulting in the formation of an active receptor homodimer. Receptor dimerization and conformational change initiate phosphorylation of the Janus tyrosine kinase (JAK2) and activation of downstream signalling pathways, leading to increased cell proliferation and other actions. Based on simulation studies it was predicted that higher site 1 affinity would result in increased capture of GH at the cell surface, resulting in a lower EC50 for cell proliferation. Conversely, a higher site 2 affinity was proposed to enhance the ability of the GH to dimerize its receptor, resulting in a higher maximum response than wild-type GH. However, these approaches have been unsuccessful in improving the bioactivity of human GH (hGH). In past decade, a series of hGH analogues possessing higher receptor binding affinity at site 1 were produced by alanine-scanning mutagenesis or by monovalent phage display, with affinities improved up to 50-foId for site 1, site 2, or both. None of the analogues displayed a significant improvement in biopotency when tested in cell proliferation assays. Thus, the studies concluded that the binding affinity of hGH for its receptor surpasses that required for maximal cellular activity and its biopotency cannot be improved.
Based on our previous observation that hGH has 5 times higher potency than pGH in a cell proliferation assay dependent on the pGH receptor, I have created a series of pGH variants which introduced hGH-specific residues within the first receptor binding site of hGH into pGH. These variants were expressed in E.coli, refolded and purified by salt gradient ion exchange chromatography. Over 20 mutants were examined for site 1 binding affinity, and affinity for the dimeric receptor using Biacore 3000. Chips created by streptavidin coupling to S201C and S237C pGH receptor ECD were used for this purpose. The results show that the determinants responsible for the higher affinity of hGH are located in three of the four clusters of residues within the binding site 1 of hGH. These clusters are group 1 (G1) comprising R41K/ Inserted44F/ I45L; group 2 (G2) containing T61S/ G62N/ K63R/ D64E/ A66T, and group 4 (G4) involving del.182R/ F184S. Residues in group 2 were most effective, increasing site 1 binding affinity up to 24 fold relative to wild type pGH by decreasing off-rate kinetics. The determinants contributing to increased affinity in the three affinity-improved groups were combined together and analysed kinetically on the Biacore 3000. In addition, all mutants were examined in cell proliferation assay with BAF-B03 cells expressing the pGH receptor. For most of these, the EC50 for cell proliferation was substantially reduced. From these results it was possible to create mutants with 4-5 fold greater potency than wt pGH by introducing 5-10 hGH residues. Thus, pGH bio-potency can be substantially improved by increasing site 1 binding affinity, in contrast to hGH.
For regulatory reasons it was necessary to show that the higher potency pGH analogues possessing hGH binding determinants did not activate the hGHR. For this purpose, pGH mutants were examined in a BAF/B03 cell proliferation assay dependent on the hGHR. While most of the higher potency analogues did not cross-react with the hGHR, it was observed that mutants containing K165R and H169D can significantly activate the hGHR. To test whether their bioactivity correlates with the binding affinities, Biacore kinetic analysis of these mutants was carried out wnth both human GHBP and pig GHBP Biacore chips. This study showed that pGH K165 (corresponding to hGH R167) and H169 (corresponding to hGH D171) comprise the key primate specificity determinants.
Finally, in order to resolve the native of the conformational change in the ECD induced on GH binding, sufficient highly purified human GHR ECD was produced to enable determination of the crystal structure of the unliganded receptor ECD. This facilitated the collaborative development of a new model for GHR activation.
In conclusion, significant contributions have been made in (1) creating higher potency pGH analogues which should prove of economic significance to the livestock industry; (2) the first full mapping of an epitope for a GHR agonist MAb; (3) determining the major specificity determinants for primate GHR specificity; (4) enabling the formulation of a new mechanism for GHR activation. These advances change our understanding of the mode of action and use of this important hormone.