Abstract:
Diabetic retinopathy (DR) is one of the most common diabetes mellitus complications that affects several hundred million people worldwide. DR disrupts the inner blood-retinal-barrier (iBRB), which protects the retina, causing visual impairment, therefore significantly decreases the quality of life. The development of new therapies, especially at the early stages of the disease is very important, since they could prevent retinal complications and vision loss. Current treatments are limited to therapies which alleviate the symptoms of the disease. Also, these therapies (e.g. intraocular anti-VEGF antibody injections) are targeting the later stages of the disease, when it is often late to prevent the pathological pathways in retinal blood vessels. Early stages of DR are associated with pathology of main vascular cells: the number of perivascular pericytes is gradually decreasing, endothelial cells undergo apoptosis and the basement membrane between pericytes and endothelial cells is becoming thicker. TGF- pathway is a critical regulator of endothelial cell-pericyte crosstalk during development and DR, and TGF- levels are elevated in diabetes.
Although the iBRB plays a critical role in retinal homeostasis, the iBRB function under physiological and pathological conditions is mostly unknown, due to the lack of suitable in vitro human iBRB models. To tackle this problem, we aimed to generate an in vitro human disease model for DR by using human induced pluripotent stem cell (hiPSC)-derived endothelial cells (ECs) and pericytes (PCs).
Initially, to recapitulate the iBRB, blood vessel components were generated from hiPSC-derived cells. Endothelial cells and pericytes, as the main components of iBRB were generated from the same hiPSC source. We have developed a novel protocol for endothelial cell generation which included the combination of a 3D suspension and 2D monolayer culture and used it for our model. Pericytes, the second crucial part of iBRB were generated from neural crest stem cells considering their embryonic resemblance to the retinal pericytes. Generated ECs and PCs were characterized based on morphology, and cell markers by immunocytochemistry and flow cytometry, and were proven to be functional. Co-cultured ECs and PCs in a physiological iBRB ratio (EC:PC ; 1:1) in the chip formed 3D blood vessel network within 24 hours and were viable for 7 days. Blood vessels were perfusable, allowing testing of blood vessel functionality. In order to mimic diabetes on blood vessels, hyperglycemic culture media (75 mM glucose) and advanced glycation end product (AGE)-precursor methylgloxal (MGO) treatment were applied in the chip. Diabetic vs. control blood vessels were quantitatively compared in terms of blood vessel diameter and shape. Interestingly, hyperglycemia treatment caused a slight decrease in blood vessel diameter, whereas MGO caused an enlargement in blood vessels. Blood vessel shape was not affected by diabetic conditions. I also investigated the effect of TGF- pathway on blood vessels, since TGF- levels are elevated in DR and its signalling is crucial for EC-PC crosstalk and vessel maturation. TGF- and its inhibitor (SB431542) both caused a significant decrease in vessel diameter, and SB431542 treatment resulted in the roundest vessel shape compared to control. Additionally, macroglial components of iBRB, Müller glia and astrocytes have been generated and characterized from hiPSCs.
Therefore, the main accomplishments of this thesis are:
1. The development of the first microvasculature on the 3D cell culture chip, using exclusively human iPSC-derived cells
2. Recapitulation of inner blood-retinal-barrier using hiPSC-derived functional endothelial cells and pericytes in vitro
3. Generation of a disease model for diabetic retinopathy by two different approaches (hyperglycemia and MGO) in blood vessels
4. Investigating the effect of TGF- treatment and its inhibition in the disease model
5. Quantitative analysis, which displayed an enlargement of blood vessel diameter in MGO-treated blood vessels.
6. Generation of glial components (Müller glia and astrocytes) of inner blood-retinal-barrier for further developments of the disease model
The DR disease model described here is a promising tool for studying the disease etiology of diabetic retinopathy in vitro. Co-culture of multiple human cell types in 3D to recapitulate a tissue or an organ allows disease etiologies to be studied in detail. Additionally, defining metabolites or altered gene expression patterns would allow for early diagnostic markers for DR to be discovered. Furthermore, additional co-cultures of blood vessels with generated and characterized hiPSC-derived glial cells can improve the recapitulation of iBRB. When upscaled, 3D cell culture and the use of hiPSC-derived cells are suitable for high-throughput analysis of blood vessel properties and potential drug candidates.
disease modelling
Print-on-Demand