The vascular system of the human body comprises arteries and arterioles constituting the arterial system, veins and venules constituting the venous system, and small capillaries that transport blood cells bringing oxygen and nutrients to the cells and carrying away carbon dioxide, lymph, and other waste ^[1][13]. High-pressure blood from the heart passes by the large arteries first, then small arteries and arterioles, before entering the capillaries, and then back to the heart by passing through the venules and veins ^[2][14]. Arterioles, capillaries, and venules form the multi-level vascular bed, which delivers nutrients and oxygen through the highly permeable vascular wall ^[3][15]. Lymph capillaries are the initial lymphatic vessels responsible for absorbing lymph, apoptotic cells, cellular debris, and circulating immune cells into the lymphatic system ^[4][16]. Endothelial cells (ECs) are common to all vessels and form the inner tunica intima layer that lines the vascular lumen. The middle layer, or tunica media, consists of circumferentially oriented smooth muscle cells (SMCs), and in large elastic arteries, multiple circular smooth muscle layers alternate with rings of elastic lamellae. Most blood vessels consist of three histologically distinct regions, each containing variable amounts of SMCs and elastin ^[1][13]. The cellular and extracellular matrix (ECM) composition of arteries, veins, and capillaries are different due to their function. Arterioles have fewer smooth muscle layers, and capillaries are covered by a discontinuous coat of pericytes (PCs) instead of SMCs. In contrast, veins have less smooth muscle and thinner walls than arteries and are more elastic ^[5][17].

The formation of new blood vessels, both in the embryo and postnatally, involves three major processes: (1) vasculogenesis, which is the formation of the first primitive vascular plexus in an embryo or postnatally de novo, (2) angiogenesis, consisting of the sprouting of new vessels from pre-existing ones, such as growing new capillaries out of postcapillary venules; and (3) arteriogenesis, defined as the rapid proliferation of pre-existing collateral vessels, induced by the change of blood flow during the obstruction ^[6][7][8][18,19,20]. Angiogenesis can be classified into two broad categories: sprouting angiogenesis and angiogenic remodeling ^[9][21]. After initial blood vessel formation, the vascular network is expanded, remodeled, and then undergoes maturation. Vascular tubes become covered by PCs on smaller capillaries or SMCs on larger vessels and ECM and become established functional blood vessels ^[10][11][22,23].

Angiogenesis was first observed in vitro by Folkman and Haudenschild, who observed the spontaneous organization of capillary ECs into capillary-like structures (CLS) and the formation of lumen ^[12][24]. Since these first angiogenesis assays, several achievements in engineering in vitro vasculatures have been done and reviewed previously ^[13][12]. In vitro vasculatures usually consist of three key elements: (1) human ECs, used to line the lumen of the bioengineered vascular structures; (2) human perivascular cells, used to support ECs function and/or provide perivascular stability to the networks; and (3), a scaffold that provides a physical space for the cells to interact and for the vascular network to develop, and the proangiogenic factors (VEGF, FGF and other various angiocrines) supplemented in the culture media or secreted by co-cultured stromal cells, such as fibroblasts ^[14][25].

The morphology and gene expression of the ECs is tissue-specific ^[15][26]. Endothelial cell phenotypes can differ not only between organs but they can also vary in the same organ between different areas of blood circulation, its neighboring ECs, and blood vessel subtypes ^[16][27]. This allows ECs to carry specific tissue functions ^[15][26]. For example, the brain endothelium and the BBB, respectively, consist of tight junctions ensuring low permeability and high transepithelial/transendothelial electrical resistance ^[17][28]. On the other hand, in order to provide a rapid molecular exchange, fenestrated endothelial layers can be found in kidney glomeruli and discontinuous endothelial layers in liver sinusoids ^[18][29].

The blood vessels’ differences in size, morphology, and function in different organs are long known, though distinct EC subpopulations are only recently being identified ^[18][29]. ECs express phenotypic heterogeneity depending on the location in the vascular tree ^[19][30]. Marcu and col. (2018) were able to point out different expression patterns of gene clusters involved in organ development and their functions. Data were collected mainly using isolated human fetal heart, lung, liver, and kidneys and their EC populations. The study has appointed that EC heterogeneity also translates into differences in their metabolic rates, with the heart ECs possessing the highest metabolic rate of the four aforementioned EC types ^[20][31]. Likewise, EC heterogeneity translates in the tumor environment. In a large study, lung tumor EC phenotypes of different species (human and mouse), patients, and in vivo/in vitro models were detailed by single-cell RNA sequencing (scRNA-seq) ^[21][32]. The scRNA-seq was performed on ECs of human/mouse and cultured lung tumor ECs, leading to the identification of previously unrecognized phenotypes and additional tip cell signatures.

One of the interesting tools used for the analysis of gene expression and mRNA translation in in vivo targeted cells is ribosome tagging (RiboTag). Targeted polyribosomes are labeled by hemagglutinin A and then isolated by immunoprecipitation, followed by qRT-PCR or RNA-seq. For example, Jambusarie and col. (2020) used RiboTag to isolate tissue-specific mRNA to analyze organ-specific ECs from collected brain, heart, and lung tissues to study the translatome patterns of gene clusters during homeostasis ^[22][33]. Furthermore, by exposing mice to endotoxin lipopolysaccharide, the study was also able to show tissue-specific gene expression related to vascular barrier function, metabolism, and substrate-specific transport ^[15][26]. Another study, presented by Cleuren and col. (2019) employed endothelial-specific translating ribosome affinity purification (EC-TRAP) and high-throughput RNA sequencing analysis. They demonstrated methods of in vivo snapshot and expression profiling and identified 82 shared genes among five vascular beds as well as pan-EC markers ^[2][14]. Understanding the differences and roles of individual EC populations can help improve novel treatment options and should also be considered when developing in vitro models ^[18][19][29,30].

Monolayers have been widely used in research over the years; however, the disadvantages of testing on (two-dimensional) 2D cultures have become more evident. The absence of hierarchical structure, nutrition gradients, limited cell–cell interactions, and cell organization in 2D cultures often result in higher sensitivity of the cells to drug treatments. Furthermore, the cells in (three-dimensional) 3D models behave differently than in 2D due to retained ECM signals. Therefore, 3D models often represent more physiologically relevant, in vivo-like microenvironments ^[23][24][25][34,35,36]. The recent progress in creating 3D cultures has brought several manufacturing methods, such as multicellular spheroids, organoids, scaffolds, 3D hydrogel scaffolds, organs-on-chips, and 3D bioprinting ^[26][37]. Due to a lack of proper nutrition and gas exchange in the thicker constructs without vasculature, the cells might experience necrosis. The efforts and progress in creating 3D cultures that would recapitulate the human organs are limited by missing vascularization within the structures. Therefore artificial blood vessels or vascular networks are created using various engineering techniques and biomaterials. It involves designing and constructing functional vascular systems that mimic the structure and function of natural blood vessels found in living organisms. Engineered vasculature holds great potential for a wide range of applications, such as the creation of functional tissue grafts for transplantation and tissue regeneration, the development of complex organs-on-a-chip for drug testing, and the provision of vascular support to bioengineered organs. At the same time, vasculature plays a pivotal role in many diseases, such as cancer metastasis, tumor angiogenesis, or atherosclerosis. The development of 3D models for drug testing, toxicology assays, in vitro imitation of pathological states, and progress in their applications for regenerative medicine are currently dependent on vascularization strategies ^[23][34].

3. Application of Engineered Vasculatures in Cancer Research and Drug Delivery

Successful cancer treatment entails novel drugs and delivery approaches ^[27][101]. Drug development requires various screening models, including cell culture on a plate and animal models ^[28][29][102,103]. However, cell culture on a plate is limited in replicating the in vivo microenvironment and can only be used for early toxicity testing ^[30][31][104,105]. Animal models can replicate in vivo drug responses but are limited in their ability to predict drug side effects and effectiveness in humans ^[32][33][76,106]. The tumor microenvironment (TME) promotes tumor growth and tends to be highly inflamed, characterized by the presence of various immune cells and the growth of blood vessels ^[34][107]. Moreover, the function of blood and lymphatic vessels in the TME is aberrant, as they are more tortuous, disorganized, and non-functional and contribute to tumor immunosuppression and resistance to treatment. Therefore, one strategy to treat tumors is combining standard therapy and anti-angiogenesis agents to normalize tumor vasculature transiently, enhancing the efficacy of anticancer drugs delivered during the normalization window ^[35][108]. As biomimetic in vitro systems can create a bond between 2D in vitro and animal models by imitating the 3D design of in vivo tissues, vascularized tumor models are engineered to test anticancer drugs ^[36][109]. A common practice is to culture vascular ECs with tumor spheroids to create a network of blood vessels ^[37][110]. These vascularized tumor models are employed to study (i) the delivery of nutrients and oxygen to the tumor, (ii) the efficacy of drug delivery systems, and (iii) tumor development and progression. 

4. Vascular Genesis and Angiogenesis for Tissue Engineering and Regenerative Medicine

Therapeutic vascular genesis and angiogenesis present an interesting approach in regenerative medicine, for example, in the treatment of ischemic diseases. Ischemia, or a restriction of oxygen supply to a tissue, has various etiologies, such as cardiovascular and hematologic diseases, trauma, graft failure/rejection after organ transplantation, etc. Between these etiologies, cardiovascular ischemia is the most common and presents in many clinical forms, of which the most critical are summarized in Table 1. The common denominator in the revascularization options is ‘time is tissue’. The sooner an arterial occlusion is resolved, the more chance the ischemic tissue has to recuperate and survive. Delayed revascularization often leads to irreversible damage and tissue loss. Nonetheless, reperfusion of the ischemic tissues is not the end of the story. Essential in the treatment of ischemia is to also address potential ischemia–reperfusion (I/R) injury. This pathophysiological phenomenon exists of an exacerbated inflammatory response after reperfusion/revascularization, which injures the ischemic tissue even more. Hitherto, therapeutic initiatives to counter I/R injury have not been successful in a clinical setting ^[45][182]. Thus, patients at risk for critical ischemic diseases could benefit from angio/vasculogenesis-stimulating treatments in a secondary (i.e., prevention in chronic ischemia) or tertiary (after reperfusion) setting. One novel treatment option is to manipulate the angiogenesis program, such as the VEGF, angiopoietin, and FGF pathways. These pathways are crucial for EC proliferation, migration, survival, and the stabilization of formed vascular networks ^[46][183]. Unfortunately, the administration of angiogenic cytokines as recombinant protein or gene therapy has not yet succeeded in clinical trials, as reviewed in ^[47][48][49][184,185,186]. Furthermore, vasculo- and angiogenesis research potentially has far more applications than ischemic disease treatment. As mentioned earlier, an integrated mature vascular network is indispensable for any relevant healthy tissue and cancer model, both for in vitro studies and translation into in vivo models. This means the vascularity needs to develop in an integrative manner with other cell types, avoiding the creation of a static coculture ^[50][187]. In this regard, each tissue requires its own specific vascular network. Anatomical organ-specific vascularity has been established for a while, but only recently have studies focused on the structural and molecular differences. These studies demonstrate the heterogeneity in endothelial structure in bone, brain, cardiac, liver, kidney, gut, skin ^[51][188], and pancreatic ^[19][30] capillaries. Where large vessels, brain, lung and heart tissue present a continuous endothelium with strong adherence junctions ^[52][189], the lung, kidney and gut are built of fenestrated endothelium. Even more permeability is allowed in discontinued endothelium, such as in liver tissue. Compared to structural differences, molecular heterogeneity is understudied in humans due to the lack of access to multi-organ samples. Existing studies are based on murine samples, showing a large variety of angiogenic and endothelial markers between different organs ^[20][53][31,190]. It seems important to take these organ-specific structural and molecular differences into account when assessing vascular development in engineered constructs since the vascular structure often plays a key role in organ function. To create these organ-specific vasculatures, researchers mostly look at autologous PSC (pluripotent stem cell) and EPC (endothelial progenitor cell) as cell sources due to their proliferation and endothelial differentiation potential, their accessibility and neutral immunogenic status. PSC can be differentiated into ECs in three different ways ^[54][191]. Firstly, through self-aggregation in suspension and the formation of embryoid bodies (EBs). Secondly, endothelial differentiation can be stimulated by coculturing PSC with stromal cells (so-called feeder cells) ^[55][192]. Thirdly, and currently most applied, is seeding PSC on substrate-coated (e.g., Matrigel, gelatin, fibronectin) culture plates and subsequentially adding recombinant growth factors ^[56][57][193,194].