Scholarlink

<p>The emerging crisis of antimicrobial resistance has renewed interest in bacteriophage (phage) therapy as a promising alternative. The efficacy of phage therapy primarily depends on the specific interaction between the phage’s receptor-binding proteins (RBPs) and receptors on the bacterial surface. RBPs are critical for host recognition and infection, and engineering RBPs to alter host range and tropism is a key strategy for improving phage-based treatments. Recently, artificial intelligence (AI)-based tools have emerged as enabling technologies for protein engineering, ranging from predicting RBP structures using tools like AlphaFold to optimizing binding interactions through directed evolution, chimeric design, and deep learning-assisted host range prediction. In this review, we summarize the distinct structural features of RBPs and the main engineering strategies employed to modify them. We further evaluate the application of AI-driven approaches in RBP engineering, discussing current methodologies, including structure prediction, directed evolution, and deep learning-assisted host range prediction, along with current challenges such as data scarcity, model interpretability, bottlenecks in high-throughput experimental validation, and biosafety and ethical concerns. Finally, we outline future directions for leveraging the integration of experimental and AI-driven approaches to advance the rational design of next-generation phage therapeutics with tailored host specificity and improved efficacy against multidrug-resistant bacterial infections.</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

From structure to design: experimental and AI-driven approaches in receptor-binding protein engineering for reprogramming phage host range

  • 期刊论文

  • 发表日期：

    2026年3月17日

查看全文
  • Shuo Xu,
  • Shuqi Yang,
  • Xin Jiao,
  • Jiaqi Cai,
  • Jiahui Wu,
  • Jinjuan Qiao

摘要

The emerging crisis of antimicrobial resistance has renewed interest in bacteriophage (phage) therapy as a promising alternative. The efficacy of phage therapy primarily depends on the specific interaction between the phage’s receptor-binding proteins (RBPs) and receptors on the bacterial surface. RBPs are critical for host recognition and infection, and engineering RBPs to alter host range and tropism is a key strategy for improving phage-based treatments. Recently, artificial intelligence (AI)-based tools have emerged as enabling technologies for protein engineering, ranging from predicting RBP structures using tools like AlphaFold to optimizing binding interactions through directed evolution, chimeric design, and deep learning-assisted host range prediction. In this review, we summarize the distinct structural features of RBPs and the main engineering strategies employed to modify them. We further evaluate the application of AI-driven approaches in RBP engineering, discussing current methodologies, including structure prediction, directed evolution, and deep learning-assisted host range prediction, along with current challenges such as data scarcity, model interpretability, bottlenecks in high-throughput experimental validation, and biosafety and ethical concerns. Finally, we outline future directions for leveraging the integration of experimental and AI-driven approaches to advance the rational design of next-generation phage therapeutics with tailored host specificity and improved efficacy against multidrug-resistant bacterial infections.