Engineering of a Novel Substrate Specificity of Biotin Carboxylase by Machine Learning Assisted Directed Evolution

Visa enkel post
dc.contributor.authorLeal Garza, Luis M.
dc.contributor.departmentChalmers tekniska högskola / Institutionen för biologi och biotekniksv
dc.contributor.examinerEngqvist, Martin
dc.contributor.supervisorEngqvist, Martin
dc.date.accessioned2021-08-10T13:03:00Z
dc.date.available2021-08-10T13:03:00Z
dc.date.issued2021sv
dc.date.submitted2020
dc.description.abstractThe engineering of more efficient CO2 fixation mechanisms is an important target that has been addressed by different approaches. Many of these approaches attempt to tackle the photorespiration process. In plants, photorespiration is a consequence of an oxygenation reaction catalysed by RuBisCO (main enzyme in charge of carbon fixation) instead of a carboxylation. This oxygenation occurs due to atmospheric O2 competing with CO2 for the active site of RuBisCO. In this project, the first step of a multi domain RuBisCO engineering approach is proposed. This first step consists in a substrate walk engineering on biotin carboxylase (accC) from Escherichia coli. Biotin-dependent carboxylases are attractive because they show high specificity for carboxylation and no interaction with atmospheric O2. Biotin-dependent carboxylases have been the target of many different engineering endeavours, most of which focus in the carboxytransferase domain (accA), instead biotin carboxylase (accC). Additionally, many of these approaches focus on random directed evolution. This project uses a combination of Machine Learning-assisted and rational single-site mutagenesis directed evolution approaches to introduce a new-to-nature 2-imidazolidone carboxylase activity. The work presented here represents one of the first attempts to engineer the biotin carboxylase subunit (accC), one of the few to do machine learning assisted evolution, and the only known to the date to attempt a substrate walk in the biotin carboxylase subunit. The Machine Learning Assisted phase of the project managed to incorporate two simultaneous mutations (G163L and G164H) showing for the first time a 2-imidazolidone carboxylase activity 1.5-fold greater than the background. Results show that improvements in the methodology could have reduce the bias introduced into the machine learning models. After an additional round of site-saturation mutagenesis, one subsequent mutation (G83Y) increased the 2-imidazolidone carboxylase activity to be around 4-fold greater than the background. Unexpected ATPase activity was observed in the final mutant (G83Y, G163L, G164H, F279L) leading to an estimated futile ATP use of 4.4 molecules for each 1 molecule of carboxylated 2-imidazolidone. Future work on this enzyme should monitor both, spectrophotometrically and radiometrically to better control the mutant selection. Additionally, computational models could be used to pre-select the region of interest for mutagenesis.sv
dc.identifier.coursecodeBBTX60sv
dc.identifier.urihttps://hdl.handle.net/20.500.12380/303878
dc.language.isoengsv
dc.setspec.uppsokLifeEarthScience
dc.subjectMachine Learningsv
dc.subjectDirected Evolutionsv
dc.subjectBiotinsv
dc.subjectBiotin Carboxylasesv
dc.subjectDesthiobiotinsv
dc.subject2-imidazolidonesv
dc.titleEngineering of a Novel Substrate Specificity of Biotin Carboxylase by Machine Learning Assisted Directed Evolutionsv
dc.type.degreeExamensarbete för masterexamensv
dc.type.uppsokH
local.programmeBiotechnology (MPBIO), MSc
Ladda ner
Original bundle
Visar 1 - 1 av 1
Bild (thumbnail)
Namn:
Master's Thesis_Luis M Leal Garza.pdf
Storlek:
1.21 MB
Format:
Adobe Portable Document Format
Beskrivning:
Engineering of a Novel Substrate Specificity of Biotin Carboxylase by Machine Learning Assisted Directed Evolution
Ladda ner
License bundle
Visar 1 - 1 av 1
Bild (thumbnail)
Namn:
license.txt
Storlek:
1.51 KB
Format:
Item-specific license agreed upon to submission
Beskrivning:
Ladda ner
Samling
Examensarbeten för masterexamen