Biosynthetic and Pharmacokinetic Approaches to Improve Steroid Therapeutics

Abstract

Steroid hormone analogs are clinically important, but their use is limited by severe side effects. In my dissertation, I present two approaches to improve steroid drugs. The first approach, discussed in Chapter 2, aims to biosynthesize novel steroids. Steroid derivatives are difficult to make via traditional organic synthesis, but many enzymes regio- and stereo-selectively process a wide variety of steroid substrates. I expressed seventeen of these enzymes in mammalian cells, from which I selected the human cytochrome P450 CYP7B1 for use as a biocatalyst. HEK293 cells stably expressing CYP7B1 processed two non-native substrates into three novel products. The cells 7α- and 7β-hydroxylated 17α-hydroxypregnenolone, and 11α-hydroxylated 16α-hydroxyprogesterone. Two of these reactions were unanticipated, as CYP7B1 was thought to exclusively 7α-hydroxylate steroids. I explored how these products could have arisen in Chapter 3 using a Rosetta docking model. The model suggested that these substrates’ D-ring hydroxyl groups prevents them from binding to CYP7B1 as neatly as the native substrate pregnenolone. This allows the non-native substrates to tilt, bringing different carbon atoms close to the active ferryl oxygen atom. In Chapter 4, I take a second approach to improving steroid drugs by considering whether a fusion protein could carry steroids to desired cells. I developed a multi-scale pharmacokinetics model to determine which features a glucocorticoid-binding antibody fusion protein requires to deliver steroid exclusively to leukocytes. The antibody’s target antigen must endocytose quickly upon protein binding, but endosomal release of steroid or antibody is not helpful. The model also showed that the fusion protein could direct endogenous cortisol to leukocytes, achieving immunosuppression without any synthetic glucocorticoid.