X-Ray Crystallographic Studies of Human Chorionic Gonadotropin

Biochemistry, Analytical chemistry

Harris, Deborah Clare

University:University of Glasgow

关键词: Biochemistry, Analytical chemistry;   

Others  :  http://theses.gla.ac.uk/76371/1/13834116.pdf

来源: University of Glasgow

【 摘 要 】

Human Chorionic Gonadotropin (hCG) is a glycoprotein hormone released by the trophoblast to stimulate progesterone production by the corpus luteum during the early stages of pregnancy. It exists as a heterodimer consisting of noncovalently bound alpha and beta-subunits. The alpha-subunit is common with the other members of this glycoprotein family, the pituitary hormones Luteinizing Hormone, Follicle Stimulating Hormone and Thyroid Stimulating Hormone. The alpha-subunit consists of 92 amino acid residues including ten cysteine residues which form five disulphide bridges. There are two N-linked oligosaccharide moieties bound to Asn 56 and Asn 82. The beta-subunit, which confers the specific biological role of each hormone, consists of 145 residues in hCG. There are twelve cysteine residues, conserved throughout the proteins, all involved in disulphide linkages. The beta-subunit is also heavily glycosylated with two long chain N-linked oligosaccharides at Asn 13 and Asn 30 and four O-linked oligosaccharides at Ser 121, Ser 127, Ser 132 and Ser 138. From circular dichroic work, hCG consists largely of aperiodic structure, 25-30% beta-sheet and 0-12% alpha-helix. There is a limited knowledge of the regions involved in the subunit interface and receptor binding areas, but many questions remain unanswered. This thesis describes my work in attempting to elucidate the three-dimensional structure of human Chorionic Gonadotropin by X-ray crystallography. HCG was partially deglycosylated by treatment with anhydrous Hydrogen Fluoride to produce the species HF-hCG. The tertiary structure was largely unaffected as HF-hCG was still capable of binding to receptor and had a similar secondary structure content as measured by Circular Dichroism (Keutmann et aL (1983a)). Crystals were grown of these HF-hCG species using the hanging drop method. Optimisation of the conditions led to crystals with sharp hexagonal bipyramidal morphology growing up to 0.4mm in each direction. The crystals were characterized and found to be of the space group P6i22 or enantiomer with cell dimensions a = b = 88.68A and c = 177.2A. The multiple isomorphous replacement method was employed for phase determination. Many potential heavy atom derivatives were screened using either visual comparison of precession photographs or by collection of a low resolution dataset on a Xentronics area detector and calculation of differences and difference Pattersons. Several derivatives were discovered, but unfortunately they shared one of the three major heavy atom binding sites. Data were collected on a Xentronics area detector, on film at station PX7.2 at the S.R.S., Daresbury and on the FAST at station PX9.6 at the S.R.S., Daresbury. Of the various datasets the film data were of the highest resolution, with diffraction observed to 2.5A but problems with radiation damage meant that more than one crystal was required to collect a full dataset. The best data collected were on the Xentronics area detector where a whole dataset could be collected with a single crystal. However, the maximum diffraction observed using the rotating anode source was 3.0A resolution. The data were processed using both the XENGEN and XDS packages, the latter giving far more accurate estimates of the intensities of the higher resolution data. The first derivative solved was K2Pt(CN)4 using a combination of difference Patterson maps and direct method techniques. S.I.R. phases calculated from this derivative were used to phase difference Fouriers for other derivatives. The anomolous data were used to determine the correct hand, and the true space group was found to be P6s22. MIR electron density maps were calculated and the solvent and protein regions were clear. However, it was not possible to distinguish the separate subunits. Phase refinement and phase extension methods were employed in an attempt to improve and extend the higher resolution phases. Solvent flattening, histogram matching and maximum entropy techniques all produced improved maps in which ?-sheet like density was apparent. Attempts were made to fit the sequence to the electron density, but this was made difficult both by the quality of the maps, and the extensive network of disulphide linkages. Also the unequivocal assignments of these linkages had not been made. Segments of poly alanine chain were fitted where possible and the partial structure used in phase combination in a "bootstrap" attempt to solve the structure.