癌细胞除了通过重塑代谢以存活和增殖外，还能适应并操纵其他细胞的代谢，这一特性表明，仅使用体外单细胞培养模型进行临床肿瘤代谢研究可能存在局限性，难以发现能转化为临床治疗的新型代谢靶点。尽管这一认识日益增强，并且解决这个问题的工作正在变得常规化，但仍有很多尚不明了。例如，关于癌细胞如何操纵非癌细胞的代谢的生化机制方面的知识以及对它们的存活和增殖产生的影响的了解仍然有限。此外，不同癌症类型和进展阶段的这些过程的变异以及它们对治疗的意义也仍然大部分未被探索。本研究采用跨学科的方法，利用数学建模的预测能力丰富实验结果。我们开发了一个功能性的多细胞体外模型，用于定性和定量分析骨髓间充质干细胞和骨髓瘤细胞系体外共培养模型所引发的代谢网络。为了获得这个模型，我们设计了一个定制的人类基因组约束性重构工作流程，结合了传统的mCADRE和Metabotools算法、新颖的redHuman算法以及13C代谢通量分析。我们的工作流程将最新的人类代谢网络矩阵(Recon3D)转化为两个细胞特异性模型，并结合一个跨越共同生长培养基的代谢网络。当将我们的体外模型与体外模型进行交叉验证时，我们发现体内模型成功地再现了其体外对应物的重要代谢行为，包括细胞生长预测、呼吸速率，以及对表明细胞之间红氧化代谢物互相转运的观察提供支持的结果。版权：© 2023 Vera-Siguenza等。本文是根据创作公共许可条款的开放获取文章，允许在任何媒体中进行无限制使用、分发和复制，只要原作者和出处被标明。

It is increasingly apparent that cancer cells, in addition to remodelling their metabolism to survive and proliferate, adapt and manipulate the metabolism of other cells. This property may be a telling sign that pre-clinical tumour metabolism studies exclusively utilising in-vitro mono-culture models could prove to be limited for uncovering novel metabolic targets able to translate into clinical therapies. Although this is increasingly recognised, and work towards addressing the issue is becoming routinary much remains poorly understood. For instance, knowledge regarding the biochemical mechanisms through which cancer cells manipulate non-cancerous cell metabolism, and the subsequent impact on their survival and proliferation remains limited. Additionally, the variations in these processes across different cancer types and progression stages, and their implications for therapy, also remain largely unexplored. This study employs an interdisciplinary approach that leverages the predictive power of mathematical modelling to enrich experimental findings. We develop a functional multicellular in-silico model that facilitates the qualitative and quantitative analysis of the metabolic network spawned by an in-vitro co-culture model of bone marrow mesenchymal stem- and myeloma cell lines. To procure this model, we devised a bespoke human genome constraint-based reconstruction workflow that combines aspects from the legacy mCADRE & Metabotools algorithms, the novel redHuman algorithm, along with 13C-metabolic flux analysis. Our workflow transforms the latest human metabolic network matrix (Recon3D) into two cell-specific models coupled with a metabolic network spanning a shared growth medium. When cross-validating our in-silico model against the in-vitro model, we found that the in-silico model successfully reproduces vital metabolic behaviours of its in-vitro counterpart; results include cell growth predictions, respiration rates, as well as support for observations which suggest cross-shuttling of redox-active metabolites between cells.Copyright: © 2023 Vera-Siguenza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.