基于图形的超分辨蛋白质-蛋白质相互作用的空间邻近性预测单细胞中的癌症药物反应。

Graph-Based Spatial Proximity of Super-Resolved Protein-Protein Interactions Predicts Cancer Drug Responses in Single Cells.

作者信息

Zhang Nicholas, Cai Shuangyi, Wang Mingshuang, Hu Thomas, Schneider Frank, Sun Shi-Yong, Coskun Ahmet F

机构信息

Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA USA.

Interdisciplinary Bioengineering Graduate Program, Georgia Institute of Technology, Atlanta, GA USA.

出版信息

Cell Mol Bioeng. 2024 Oct 6;17(5):467-490. doi: 10.1007/s12195-024-00822-1. eCollection 2024 Oct.

DOI:10.1007/s12195-024-00822-1

PMID:39513000

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11538221/

Abstract

PURPOSE

Current bulk molecular assays fail to capture spatial signaling activities in cancers, limiting our understanding of drug resistance mechanisms. We developed a graph-based super-resolution protein-protein interaction (GSR-PPI) technique to spatially resolve single-cell signaling networks and evaluate whether higher resolution microscopy enhances the biological study of PPIs using deep learning classification models.

METHODS

Single-cell spatial proximity ligation assays (PLA, ≤ 9 PPI pairs) were conducted on EGFR mutant (EGFRm) PC9 and HCC827 cells (>10,000 cells) treated with 100 nM Osimertinib. Multiplexed PPI images were obtained using wide-field and super-resolution microscopy (Zeiss Airyscan, SRRF). Graph-based deep learning models analyzed subcellular protein interactions to classify drug treatment states and test GSR-PPI on clinical tissue samples. GSR-PPI triangulated PPI nodes into 3D relationships, predicting drug treatment labels. Biological discriminative ability (BDA) was evaluated using accuracy, AUC, and F1 scores. The method was also applied to 3D spatial proteomic molecular pixelation (PixelGen) data from T cells.

RESULTS

GSR-PPI outperformed baseline models in predicting drug responses from multiplexed PPI imaging in EGFRm cells. Super-resolution data significantly improved accuracy over localized wide-field imaging. GSR-PPI classified drug treatment states in cancer cells and human lung tissues, with performance improving as imaging resolution increased. It differentiated single and combination drug therapies in HCC827 cells and human tissues. Additionally, GSR-PPI accurately distinguished T-cell stimulation states, identifying key nodes such as CD44, CD45, and CD54.

CONCLUSION

The GSR-PPI framework provides valuable insights into spatial protein interactions and drug responses, enhancing the study of signaling biology and drug resistance.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s12195-024-00822-1.

摘要

目的

当前的大量分子检测方法无法捕捉癌症中的空间信号传导活动，限制了我们对耐药机制的理解。我们开发了一种基于图的超分辨率蛋白质-蛋白质相互作用（GSR-PPI）技术，以在空间上解析单细胞信号网络，并评估更高分辨率的显微镜技术是否能使用深度学习分类模型增强对蛋白质-蛋白质相互作用的生物学研究。

方法

对用100 nM奥希替尼处理的EGFR突变（EGFRm）PC9和HCC827细胞（>10,000个细胞）进行单细胞空间邻近连接分析（PLA，≤9个蛋白质-蛋白质相互作用对）。使用宽场和超分辨率显微镜（蔡司Airyscan，SRRF）获得多重蛋白质-蛋白质相互作用图像。基于图的深度学习模型分析亚细胞蛋白质相互作用，以对药物治疗状态进行分类，并在临床组织样本上测试GSR-PPI。GSR-PPI将蛋白质-蛋白质相互作用节点三角测量为三维关系，预测药物治疗标签。使用准确率、AUC和F1分数评估生物学判别能力（BDA）。该方法还应用于来自T细胞的3D空间蛋白质组学分子像素化（PixelGen）数据。

结果

在从EGFRm细胞的多重蛋白质-蛋白质相互作用成像预测药物反应方面，GSR-PPI优于基线模型。超分辨率数据比局部宽场成像显著提高了准确率。GSR-PPI对癌细胞和人肺组织中的药物治疗状态进行了分类，随着成像分辨率的提高，性能也有所改善。它区分了HCC827细胞和人体组织中的单一和联合药物治疗。此外，GSR-PPI准确区分了T细胞刺激状态，识别出关键节点，如CD44、CD45和CD54。