Scholarlink

<p>The synthesis of chiral amines is crucial since they are a key component in nearly 40% of top-selling drugs. ω-transaminases are promising biocatalysts for producing these chiral amines. This study reports the isolation of a wild-type <i>Bacillus</i> strain (<i>Bacillus inaquosorum</i> AGSP2) from a contaminated site at the Amlakhadi river, Gujarat, India. We optimized ω-transaminase production using both One Factor at a Time and Response Surface Methodology- Central Composite Design (RSM-CCD) techniques, with further model validation via an Artificial Intelligence (AI) tool, Support Vector Machine. The optimal medium, called Modified Luria–Bertani, contains fructose (12&#xa0;g/L), NaCl (7.5&#xa0;g/L), yeast extract (7.5&#xa0;g/L), peptone (12&#xa0;g/L), and α-methylbenzylamine (5&#xa0;mM). Optimization increased ω-transaminase production 2.8 times, reaching an activity of 6121.88 ± 42 U/ml at 37&#xa0;°C, pH 7, 120&#xa0;rpm, with 2% (v/v) inoculum. The RSM-CCD model had R2 = 0.95, predicted R2 = 0.78, RMSE = 0.2327, while SVM achieved R2 = 0.99, predicted R2 = 0.96, RMSE = 0.1327. AGSP2 catalyzed the biotransformation of acetophenone with (S)-α-methylbenzylamine, resulting in a 53.32% conversion rate. These findings demonstrate the potential of combining statistical and AI tools to improve biocatalyst production and applications, presenting a sustainable approach for chiral amine synthesis and highlighting ω-transaminase’s role in green biocatalysis.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Harnessing Bacillus inaquosorum AGSP2 for enhancing ω-transaminase production through classical and AI-supported statistical design

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月18日

查看全文
  • Shreya Pandya,
  • Urvish Chhaya,
  • Akshaya Gupte

摘要

The synthesis of chiral amines is crucial since they are a key component in nearly 40% of top-selling drugs. ω-transaminases are promising biocatalysts for producing these chiral amines. This study reports the isolation of a wild-type Bacillus strain (Bacillus inaquosorum AGSP2) from a contaminated site at the Amlakhadi river, Gujarat, India. We optimized ω-transaminase production using both One Factor at a Time and Response Surface Methodology- Central Composite Design (RSM-CCD) techniques, with further model validation via an Artificial Intelligence (AI) tool, Support Vector Machine. The optimal medium, called Modified Luria–Bertani, contains fructose (12 g/L), NaCl (7.5 g/L), yeast extract (7.5 g/L), peptone (12 g/L), and α-methylbenzylamine (5 mM). Optimization increased ω-transaminase production 2.8 times, reaching an activity of 6121.88 ± 42 U/ml at 37 °C, pH 7, 120 rpm, with 2% (v/v) inoculum. The RSM-CCD model had R2 = 0.95, predicted R2 = 0.78, RMSE = 0.2327, while SVM achieved R2 = 0.99, predicted R2 = 0.96, RMSE = 0.1327. AGSP2 catalyzed the biotransformation of acetophenone with (S)-α-methylbenzylamine, resulting in a 53.32% conversion rate. These findings demonstrate the potential of combining statistical and AI tools to improve biocatalyst production and applications, presenting a sustainable approach for chiral amine synthesis and highlighting ω-transaminase’s role in green biocatalysis.