Targeting endothelial-erythrocyte glycocalyx exchange for the novel diagnosis and treatment of renal disease

内皮细胞在我们的血管内形成连续的内膜。它们被一层叫做糖萼的保护层所覆盖。这一层是由蛋白质和糖混合而成的。它在细胞表面形成果冻状的覆盖物。这个糖萼有几个重要的功能。它调节细胞和蛋白质从血液进入组织的通道。它可以检测血液流动的速度（帮助调节组织血流量），并防止血液不必要地凝结。当糖萼不能发挥这些作用时，血管就不能发挥最佳功能，从而导致器官受损。脆弱的糖萼层损伤发生在几种人类疾病中。在糖尿病患者中，内皮糖萼损伤发生得很早，在易患肾脏疾病之前。在糖尿病的动物模型中，我已经证明，在糖萼受损的早期阶段进行干预（但在尿液中开始检测到蛋白质之前）可以预防肾脏疾病的发展。在人类中，并不是每个糖尿病患者都会出现糖萼损伤或肾脏疾病。我们认为，在病程早期出现糖萼受损的患者将最有可能随后发展为糖尿病性肾损害。识别这些高危患者将使临床医生能够优先考虑哪些患者应该更定期地就诊，哪些患者应该用现有药物进行更积极的治疗。目前，人类糖萼损伤不容易检测与临床试验。然而，我已经开发了一种新的血液测试来研究糖萼的变化，可以开发用于临床实践。在这个项目中，我将通过同时研究多个部位的内皮糖萼，来确认我的测试给了我们一个关于糖萼如何在全身变化的真实图景。这项研究需要在动物疾病模型中进行，以使我们能够获得所需的所有组织。在英国进行这项研究需要大量的糖尿病动物。然而，通过与洛杉矶的一个研究小组合作，我可以重新使用我在博士学位期间开发的测量糖萼深度变化的技术。利用他们的尖端显微镜，我可以对麻醉小鼠体内的血管进行成像，并测量随着糖尿病的发展，糖萼在器官和皮肤内是如何变化的。我可以每周测量这些部位的糖萼深度，并将我检测到的变化与我的血液测试进行比较，以证明我的测试反映了我们看到的变化。为了验证我的测试是否能预测哪些糖尿病患者的血管会继续受损，我将把我的测试纳入欧洲的既定研究“BEAt-DKD”中。这项研究目前正在招募参与者，并允许我将我的测试与目前内皮损伤的金标准测试进行比较，并收集后续数据来验证我的假设，即我们可以预测谁将发展为可检测的糖尿病肾病。我将与布里斯托尔大学的计算机科学家一起开发我的测试，以完善和自动化分析，同时将机器学习集成到编程中。机器学习应该让计算机在为我们执行分析方面变得越来越好，因为它从错误中学习，随着时间的推移变得更快、更可靠。最后，我将利用我通过研究内皮糖萼的变化所获得的信息，开发一种将糖萼的关键部分重新应用于血管壁的新方法。我的方法是针对循环中糖萼受损最严重的区域通过集中修复血液流动不顺畅的区域。这种新型的修复系统可用于急性糖萼损伤风险最高的患者，如创伤和败血症后，或慢性糖萼损伤患者，如血液透析患者。

Endothelial cells form a continual lining within our blood vessels. They are covered by a protective layer called the glycocalyx. This layer is made from a mix of proteins and sugars. It forms a jelly like covering on the cells surface. This glycocalyx performs several important jobs. It regulates the passage of cells and proteins from the blood into the tissues. It detects how fast the blood is moving (helping to regulate tissues blood flow), and it prevents blood from clotting unnecessarily. When the glycocalyx fails to perform these roles blood vessels cease to function optimally, which subsequently results in organs becoming damaged. Damage to the delicate glycocalyx layer occurs in several human diseases. In diabetes damage to the endothelial glycocalyx happens very early on, before predisposed patients develop detectable kidney disease. In animal models of diabetes, I have shown that intervening at this early stage when the glycocalyx is damaged (but before protein starts to be detectable in the urine) can prevent the development of kidney disease. In humans not everyone with diabetes develops glycocalyx damage or kidney disease. We believe that patients that develop a damaged glycocalyx early in the disease course will be the most likely to subsequently develop diabetic kidney damage. Identifying these high-risk patients would allow clinicians to prioritise which patients to see more regularly and which patients to treat more aggressively with existing medications.Currently, human glycocalyx damage is not easily detectable with a clinical test. However, I have developed a novel blood test to study glycocalyx changes that could be developed for clinical practice. In this project I will confirm that my test is giving us a true picture of how the glycocalyx changes throughout the body by simultaneously studying the endothelial glycocalyx at multiple sites. This study needs to be performed in animal models of disease to allow us access to all the tissue we need. To perform this study in the UK would need large numbers of diabetic animals. However, by collaborating with a research group in Los Angeles I can re-use a technique that I developed to measure glycocalyx depth changes during my PhD. Using their cutting edge microscope, I can image blood vessels deep inside anesthetised mice and measure how the glycocalyx is changing inside the organs and skin as diabetes develops. I can measure the glycocalyx depth at these sites every week and compare the changes I detect with my blood test to prove that my test is reflecting the changes we see.To check that my test can predict which diabetic patients will go on to develop damage to their blood vessels I will include my test in the established European study 'BEAt-DKD'. This study is currently enrolling participants and will allow me to compare my test to the current gold standard tests of endothelial damage and collect follow-up data to test my hypothesis that we can predict who will develop detectable diabetic nephropathy. I will simultaneously work on the development of my test by working with computer scientists at the University of Bristol to refine and automate the analysis whilst integrating machine learning into the programming. Machine learning should let the computer get better and better at performing the analysis for us as it learns from its mistakes to become a faster and more reliable with time.Finally, I will be use the information I learn by studying changes in the endothelial glycocalyx to develop a novel method of reapplying key parts of the glycocalyx to the blood vessel wall. My method will target the areas of the circulation where the glycocalyx is most damaged by focusing repairs on areas where the blood does not flow smoothly. This novel repair system could be used when patients are at the highest risk of acute glycocalyx damage e.g. after trauma and sepsis, or in patients with chronic glycocalyx damage e.g. patients on haemodialysis.