University of Illinois Urbana-Champaign

Breaking carbon-carbon bonds with cytochrome P450s from plants, animals, and bacteria

Miller, Justin C.

Content Files
Appendix G.sls-slas_ancestor-loganic-acid.pdb
Appendix F. Ancestral Sequences Site-by-Site Probabilities.txt
Appendix E. Multiple Sequence Alignment with All SLS, SLAS, 7DLH, Ancestral Sequences.xlsx
Appendix D. cyp72a730_loganic-acid.pdb
Appendix C. cyp72a565_loganic-acid.pdb
Appendix B. cyp72a564_loganic-acid.pdb
Appendix A. cyp72a1_loganin.pdb
MILLER-DISSERTATION-2022.pdf

Permalink

https://hdl.handle.net/2142/115446

Description

Title
Breaking carbon-carbon bonds with cytochrome P450s from plants, animals, and bacteria
Author(s)
Miller, Justin C.
Issue Date
2022-04-19
Director of Research (if dissertation) or Advisor (if thesis)
Schuler, Mary A.
Doctoral Committee Chair(s)
Hergenrother, Paul J.
Committee Member(s)
  • Sligar, Stephen G.
  • Lu, Yi
Department of Study
Chemistry
Discipline
Chemistry
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
  • cytochrome P450
  • Camptotheca acuminata
  • terpene indole alkaloid
  • enzyme catalysis
  • natural product biosynthesis
  • homology modeling
  • ancestral sequence reconstruction
  • bioinorganic chemistry
  • secondary metabolism
  • plant biochemistry
Abstract
The breaking of carbon-carbon (C-C) bonds facilitated by cytochrome P450s (CYPs, P450s) is remarkable for several reasons. Chemically, the general inertness of the C-C bond makes these reactions intriguing for the selectivity with which particular C-C bonds are oxidized and cleaved. Biochemically, the inclusion of such transformations in biosynthetic pathways enables the rearrangement of carbon scaffolds into different shapes without the requirement for developing a new set of C-C bond forming enzymes. Biologically, such P450s are included in pathways that process steroids and other hormones, synthesize essential cofactors, deactivate xenobiotics, and produce compounds to ward off microbial invaders or animal predators. This dissertation summarizes investigations into CYPs mediating C-C bond breaking reactions from the plant, animal, and bacterial world. First, I identified two secologanic acid synthases (SLASs; CYP72A564 and CYP72A565) from the medicinal plant Camptotheca acuminata responsible for producing secoiridoids from iridoids. After coupling these secoiridoids with tryptamine, the resulting strictosidinic acid core is diversified into species-specific terpene indole alkaloids (TIAs), a group of specialized metabolites many of which are used as medicines. Cloning the full-length coding sequences from cDNA, heterologous protein expression in E. coli, and in vitro assays of the full-length P450s confirmed that these Camptotheca SLASs are capable of breaking the C7-C8 bond of the iridoid glucosides loganic acid and loganin to make the secoiridoids secologanic acid and secologanin, respectively. Next, I investigated the structural bases of the broadened substrate scope of these Camptotheca SLASs compared to the secologanin synthases (SLSs; CYP72As) from other species that can only utilize loganin. A combination of ancestral sequence reconstruction, homology modeling, site-directed mutagenesis, and in vitro assays of the purified proteins revealed that the amino acid identity at two adjacent sites toggles the selectivity of the SLASs. The His131, His132 wild type proteins can turnover both loganic acid and loganin; a His132Asp mutation decreases loganic acid turnover and eliminates loganin turnover; a His131Phe mutant increases loganin turnover and conversely eliminates loganic acid turnover. The combined identification of two SLASs and mutants that toggle their substrate selectivity furnishes an improved set of secoiridoid-producing enzymes to use in heterologous expression of TIAs, replacing the yield-limiting SLS from Catharanthus roseus. Subsequently, I investigated a third Camptotheca SLAS candidate (CYP72A730) that, although closely related to the SLASs, did not consume either loganic acid or loganin. Observing a three amino-acid deletion in CYP72A730 relative to the Camptotheca SLASs, I hypothesized that this deletion was the reason for this P450’s inactivity. Site-directed mutagenesis to add these three amino acids to CYP72A730 and remove them from CYP72A564 and subsequent in vitro assays revealed that the inactivity of CYP72A730 and activity of CYP72A564 were unaffected. Consequently, it appears the inability of CYP72A730 to produce secoiridoids from loganic acid and loganin originates in some combination of the other 30 differences between it and the Camptotheca SLASs, most likely among the 11 more drastic changes in side chain size, charge, and flexibility. Discerning the mechanism of the C-C bond breaking catalyzed by these SLASs was another avenue of inquiry. My approach was the synthesis of an iridoid glucoside as a probe substrate interrogating the substrate-based requirements for C-C bond scission. Working from the readily available iridoid geniposide, troubles in the early portions of several synthetic routes slowed progress. After troubleshooting these early reactions and redesigning parts of the synthesis, the 7-deoxy-7-methyl-loganin compound is nearly complete. Once synthesized, a combination of in vitro reconstitutions with CYP72A564 and high resolution LC-MS for analysis will give insights into how the substrate’s composition influences C-C bond scission. The final series of studies described herein involve modeling the C-C lyase reaction catalyzed by the human, steroid-metabolizing CYP17A1 by reverse engineering this reactivity in bacterial P450s. Seeking a clearer understanding of whether this C-C bond breaking reaction occurs using a peroxyanion-mediated or a Compound I-mediated mechanism, the soluble, well-studied bacterial P450s present access to experiments not possible with CYP17A1. Using commercially available as well as specifically designed and synthesized α-hydroxyketones, neither Pseudomonas putida CYP101A1 nor Bacillus megaterium CYP102A1 could be coaxed into performing the C-C bond scission of interest. Molecular dynamics simulations with CYP101A1 suggested that the P450 positioned the α-hydroxyketone moiety too infrequently for catalysis by the peroxyanion-mediated mechanism. Subsequent work with another bacterial P450 (Rhodopseudomonas palustris strain HaA2 CYP199A4) and its benzoic acid substrates presented a system tested this hypothesis (i.e., that the C-C lyase reaction demands holding the α-hydroxyketone in a specific orientation over the heme for an extended period of time). For this, I synthesized an α-hydroxyketone onto a benzoic acid scaffold. This synthesis allowed the Bell and De Voss groups in Australia to provide early evidence of a P450-dependent C-C bond scission with this model substrate. With a functional system now in hand, future experiments will probe the reaction in CYP199A4 to ensure its comparability to that catalyzed by CYP17A1 and definitively determine the mechanism of its C-C lyase chemistry.
Graduation Semester
2022-05
Type of Resource
Thesis
Handle URL
https://hdl.handle.net/2142/115446
Copyright and License Information
© 2022 Justin C. Miller

Owning Collections

Graduate Dissertations and Theses at Illinois PRIMARY
Graduate Theses and Dissertations at Illinois