Definition of an Index Parameter To Screen Highly Functional Enzymes Derived from a Biochemical and Thermodynamic Analysis of Ancestral <i>meso</i>‐Diaminopimelate Dehydrogenases

Araseki, Hayato; Kawasaki, Mayu; Kambe, Akira; Kozuka, Kohei; Ito, Sohei; Nakano, Satoshi

doi:10.1002/cbic.202200727

Cited by 2 publications

(2 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this study, we attempted to assign new SrtEs optimal for SML through in silico enzyme screening, following methodologies from earlier research. 41,42 The reconstruction of ancestral SrtEs was contemplated using the entirety of the From an initial pool of 5000 sequences, 800 homologue sequences of SavSrtE were identified via following two steps of the curation: (a) homologue sequences were downloaded using Blastp, inputting the SavSrtE sequence with an E-value of "1.0 × 10 −6 " and selecting the "nonredundant" database. (b) Sequences deviating by more than 90% in length from SavSrtE, or those showing >90% identity with other sequences, were excluded.…”

Section: ■ Results and Discussionmentioning

confidence: 99%

“…In this study, we attempted to assign new SrtEs optimal for SML through in silico enzyme screening, following methodologies from earlier research. , The reconstruction of ancestral SrtEs was contemplated using the entirety of the sequences in the curated library. However, selecting a representative subset of ancestral SrtEs from hundreds of reconstructed sequences poses a significant challenge.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Design of Ancestral Sortase E that is Applicable in Protein Biomaterial Synthesis

Miyata,

Chisuga,

Kambe

et al. 2024

ACS Catal.

Self Cite

View full text Add to dashboard Cite

Protein biomaterials would have the potential to address global challenges in health and environment. Numerous production methods of the biomaterials exist, with sortase-mediated ligation (SML) being one of the representative technique. SML facilitates the site-specific conjugation of two compounds: donor peptides or proteins with a cell wall sorting signal at their C-terminus and nucleophiles that have oligoglycine at their N-terminal. In our research, we reconstructed an ancestral sortase E, named AcSE5, through a combination of sequence data mining and ancestral sequence reconstruction. AcSE5, a Ca 2+ independent sortase, recognizes donors with LAETG at their Ctermini and can employ both peptides bearing GGG-or GAA-at N-terminus and straight-chained alkylamines as nucleophiles. The enzyme can achieve efficient peptide conjugation, exceeding 70% under optimal conditions. With AcSE5, we synthesized two protein conjugates: Venus-conjugated shark variable new antigen receptor (VNAR) and dual-conjugated VNAR via poly(ethylene) glycol diamine. Direct enzyme immobilization to amino-terminated polystyrene beads was also achieved using AcSE5. The resultant beads, when conjugated with hyper-thermostable ancestral L-amino acid oxidases (HTAncLAAO2), can be employed for deracemization of various racemic amino acids into D-form. For three of phenylalanine derivatives, preparative-scale (100 mg scale) deracemization can be achieved. This process provides high enantiopurity (>99% ee) and isolation yields (>64%) through chemoenzymatic reactions. The immobilized HTAncLAAO2 showed complete resistance to 10 repeated reactions for a total of 240 h. AcSE5 is an excellent enzyme for SML applications.

show abstract

Section: ■ Results and Discussionmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Design of Ancestral Sortase E that is Applicable in Protein Biomaterial Synthesis

Miyata,

Chisuga,

Kambe

et al. 2024

ACS Catal.

Self Cite

View full text Add to dashboard Cite

show abstract

Leveraging ancestral sequence reconstruction for protein representation learning

Matthews,

Spence,

Mater

et al. 2023

Preprint

View full text Add to dashboard Cite

Protein language models (PLMs) convert amino acid sequences into the numerical representations required to train machine learning (ML) models. Many PLMs are large (>600 M parameters) and trained on a broad span of protein sequence space. However, these models have limitations in terms of predictive accuracy and computational cost. Here, we use multiplexed Ancestral Sequence Reconstruction (mASR) to generate small but focused functional protein sequence datasets for PLM training. Compared to large PLMs, this local ancestral sequence embedding (LASE) produces representations 10-fold faster and with higher predictive accuracy. We show that due to the evolutionary nature of the ASR data, LASE produces smoother fitness landscapes in which protein variants that are closer in fitness value become numerically closer in representation space. This work contributes to the implementation of ML-based protein design in real-world settings, where data is sparse and computational resources are limited.

show abstract

Definition of an Index Parameter To Screen Highly Functional Enzymes Derived from a Biochemical and Thermodynamic Analysis of Ancestral meso‐Diaminopimelate Dehydrogenases

Cited by 2 publications

References 44 publications

Design of Ancestral Sortase E that is Applicable in Protein Biomaterial Synthesis

Design of Ancestral Sortase E that is Applicable in Protein Biomaterial Synthesis

Leveraging ancestral sequence reconstruction for protein representation learning

Contact Info

Product

Resources

About