A human 3D bone-sensory nerve co-culture model to investigate bone-derived pain

背景：由癌症转移、不可愈合骨折、骨质疏松性骨折或骨关节炎(OA)等疾病引起的骨源性疼痛是出了名的难以管理和普遍存在的疾病，目前全球估计有150万骨转移癌、2200万骨不连骨折、890万骨质疏松性骨折和5.3亿骨关节炎患者。骨骼和神经之间存在着复杂的信号，它们将健康和疾病中的病理和疼痛联系在一起。感觉神经的侵袭和敏化与骨骼疾病有关，并与动物模型和患有骨关节炎、骨折和骨转移的人类患者的疼痛反应有关。该项目的目标是(I)开发人源化的3D骨-感觉神经共培养模型，(Ii)使用该模型来研究骨结构变化与局部痛觉之间的联系机制，以及(Iii)测试该模型是否可以筛选出两种已知的预防疼痛和影响骨结构的药物。由于这些神经在矿化骨中的位置，因此研究骨-神经的相互作用是具有挑战性的。目前，尚不存在人类衍生的感觉神经-骨界面模型。已经产生了一些骨痛模型，使用来自脊椎的神经，从动物和人类被称为背根节(DRG)，但在技术上具有挑战性，并未被广泛采用。动物模型被广泛用于骨源性疼痛的研究。2019年发表了650多篇原创性研究论文，利用动物模型研究骨源性疼痛的病理学，其中4%是包括犬类在内的大型动物，平均每项研究使用23种动物，总计超过14000只动物。一家制药公司在测试治疗骨痛的药物时，每年可能会使用1000多只动物。在最近一份关于整形外科动物试验的综述15中，发现在9年的时间里，40%的整形外科动物学术研究从未发表过，几乎一半的已发表研究从未被引用过，总共有近10,000个动物生命，对整形外科知识没有任何合理的影响。骨性关节炎的动物模型每组使用10只动物来揭示对骨骼结构和疼痛的影响。例如，在骨关节炎研究中，132-198只小鼠或360只大鼠将被用来测试一种药物。我们的新的骨-感觉神经模型将促进低成本的研究，以确定新的分子靶点和预先筛选止痛药。它将减少早期临床前研究动物的使用，并取代目前用于识别和测试缓解骨源性疼痛的新治疗方法的模型。我们将通过在国内和国际会议和研讨会(例如，骨科研究学会临床前模型分会、英国骨科研究学会、国际骨性关节炎研究学会)上传播，以及通过我们目前制药行业合作伙伴的药物验证，促进该模型在动物试验中的使用。

Background Bone-derived pain from conditions such as cancer metastasis, non-healing fracture, osteoporotic fracture, or osteoarthritis (OA) are notoriously hard to manage and prevalent, with an estimated 1.5 million bone cancer metastases, 22 million non-union factures, 8.9 million osteoporotic fractures and 530 million OA sufferers currently worldwide. Complex signalling exists between bone and nerves which couple pathology and pain in health and disease. Sensory nerve invasion and sensitisation are associated with bone diseases, and intrinsic to pain responses in animal models and human patients with osteoarthritis, fracture and bone metastasis. This project aims to (i) develop a humanised 3D bone-sensory nerve co-culture model, (ii) use the model to investigate mechanisms linking changes in bone structure to local pain sensation and (iii) test whether the model can screen 2 drugs known to prevent pain and influence bone structure.Investigating bone-nerve interactions is challenging due to the position of these nerves within mineralised bone. Currently, no human derived models of the sensory nerve-bone interface exist. Some models of bone pain have been produced that use nerves from the spine, called dorsal root ganglia (DRG) from animals and humans, but are technically challenging and not commonly adopted. Animal models are extensively used for the investigation of bone derived pain. In 2019, over 650 original research papers were published investigating bone derived pain in pathology using animal models, 4% of which were large animals including canines, with an average of 23 animals typically used per study, totalling over 14000 animals. A pharmaceutical company testing drugs for bone pain may use more than 1000 animals p.a. In a recent review on orthopaedic animal testing15, it was revealed that over a 9-year period, 40% of academic orthopaedic animal studies were never published and almost half of those published were never cited totalling nearly 10,000 animal lives with no justifiable impact on orthopaedic knowledge. Animal models of osteoarthritis use >10 animals per group to reveal effects on bone structure and pain. For example, in osteoarthritis research 132-198 mice or 360 rats would be used to test 1 drug. Our new bone-sensory nerve model will facilitate low-cost research to identify new molecular targets and pre-screening of drugs for pain. It will reduce early-stage preclinical research animal usage, and replace models currently used to identify and test new treatments to ease bone derived pain. We will promote the use of this model over animal testing through dissemination at national and International conferences and workshops (e.g. Orthopaedic Research Society Preclinical Models Section, British Orthopaedic Research Society, Osteoarthritis Research Society International), and by validation with drugs from our current collaborators in the pharmaceutical industry.