Less toxic yet still resistance evasive amphotericins and atomistic probing of the amphotericin B ion channel

Davis, Stephen

Permalink

https://hdl.handle.net/2142/78707 Copy

Description

Title

Less toxic yet still resistance evasive amphotericins and atomistic probing of the amphotericin B ion channel

Author(s)

Davis, Stephen

Issue Date

2015-01-22

Director of Research (if dissertation) or Advisor (if thesis)

Burke, Martin D.

Doctoral Committee Chair(s)

Burke, Martin D.

Committee Member(s)

• White, Maria C.
• Katzenellenbogen, John A.
• Tajkhorshid, Emad

Department of Study

Chemistry

Discipline

Chemistry

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• cystic fibrosis
• ion channel
• antimicrobial resistance
• antifungals
• Amphotericin B

Abstract

Amphotericin B (AmB) is a polyene macrolide natural product with two major functions: antifungal activity and ion channel formation. For over 50 years AmB has remained the last line of defense against systemic fungal infections. Remarkably, AmB has evaded the development of drug resistance during this extensive clinical lifetime. Unfortunately, AmB’s utility is impaired by severe toxicities. A less toxic AmB derivative thus stands to have a major impact on global human health. Similar to most small molecule medicines, AmB exerts its antifungal activity by binding and disabling a specific molecular target. In contrast, some diseases are caused by the malfunctioning of proteins, and thus lie outside this traditional paradigm. For example, treating the underlying cause of cystic fibrosis requires a small molecule capable of replicating the function of the CFTR chloride ion channel. AmB is the prototypical small molecule capable of ion channel formation. Although it has been known for over 40 years that AmB forms ion channels, the structure of these channels, as well as the functional groups responsible for ion conductance and selectivity remain unknown. Understanding these basic tenants of AmB mediated ion channel activity is the first step towards harnessing AmB to replace a missing or malfunctioning protein ion channel in a living system. There is thus a rich opportunity to harness synthesis to understand and optimize both the ion channel and antifungal functions of AmB. The polyol region of AmB has been predicted line channel interior, creating a hyrophillic environment for ions and water. Furthermore, the channel is proposed to be funnel shaped, with the narrowest region near the C3 hydroxyl. Based on this model, we hypothesized that removal of the C3 hydroxyl would impact ion channel conductance. The C3 alcohol was promptly excised, synthesizing C3deoxyAmB in only 9 steps from AmB. NMR characterization revealed that there were no significant alterations in the overall conformation of the of the AmB macrocycle upon removal of the C3 alcohol. Single ion channels of C3deoxyAmB in planar lipid bilayers revealed that C3deoxyAmB is capable of ion channel formation, however its conductance is significantly reduced relative to AmB. This is consistent with models that place the C-3 hydroxyl group at a critical point for ion conductance. During efforts to understand the molecular basis of ion selectivity, an efficient 3-step synthesis of a series of AmB urea derivatives was discovered. Although not useful for ion channel study, they were important probes to test an emerging allosteric modification model for non-toxic AmB derivatization. These derivatives selectively bound ergosterol (the primary sterol in yeast), but not cholesterol (the primary sterol in human cells), and maintained antifungal activity but were significantly less toxic to human cells. Additionally, these derivatives were more efficacious than AmB in a mouse model of disseminated candidiasis, and drastically less toxic in acute toxicity studies in mice. More selective pharmacological action is generally associated with decreased toxicity, but also with increased vulnerability to resistance. This creates an important question. Would a less toxic AmB derivative still evade drug resistance? The AmB urea derivatives were ideal candidates to evaluate this question. AmB urea resistant yeast were generated using mutagenesis and evaluated with a suite a fitness and genomic tests. Despite increased sterol selectivity and decreased toxicity, the development resistance to the AmB ureas was accompanied by significant fitness trade-offs, resulting in completely avirulent yeast. Therefore, it is possible to have a less-toxic yet resistance evasive antifungals. Based on their more selective, non-toxic, and resistance evasive profile, two of these compounds, AmBMU and AmBAU, are exceptionally exciting prospects as a clinical replacement for AmB. Degradative synthesis was vital in accessing both C3deoxyAmB and the AmB urea derivatives leveraged to probe AmB’s ion channel and antifungal functions. However, not all desired derivatives are accessible from this platform. For example, installing 13C labels into the backbone of the AmB framework would enable solid-state NMR studies capable of mapping the interaction between AmB and sterols at the atomistic level. An efficient and flexible total synthesis of AmB was designed grounded in the iterative cross coupling (ICC) strategy for small molecule synthesis. The synthesis was divided into three main phases, building block construction, ICC, and protecting group removal. Advances were made in all three phases. A scalable synthesis of building block one, and its cross coupling with building block two was developed. The planned final deprotection was additionally optimized. These advances contributed to the completion of the total synthesis of a fully protected doubly 13C-labeled AmB.

Graduation Semester

2015-5

Type of Resource

text

Permalink

http://hdl.handle.net/2142/78707

Copyright and License Information

Copyright 2015 Stephen Davis

Owning Collections