The word “chip” in organ-on-a-chip originates from the original manufacturing approach, an improved form of photolithographic etching used to produce computer microchips, which enables to control the shapes and sizes of surface features on the same scale [11]. Photolithography technology also has other usages in some areas, like patterning proteins and cells. Nevertheless, microfluidic culture systems are often made by “soft lithography”, and this technique is an alternative to photolithography which can be used to pattern surfaces needed in biochemistry and biology. Besides, soft lithography is cheap and experimentally simple, which can replicate patterns etched into silicon chips in more biocompatible and flexible materials [11] [12] [13]. In the process of soft photolithography, this special chip is done by pouring a liquid polymer, such as poly-dimethylsiloxane (PDMS), on an etched silicon substrate which is formed through soft lithography technology and polymerizing into an optically transparent, rubber-like material, thereby creating a “stamp” (Figure 1). Subsequently, both ends of the polymer block are opened for filling the fluids. Cells that flow into the channel adhere to the ECM substrate, and then the channel is continuously perfused with culture medium [11] - [17]. A pivotal feature of PDMS culture systems is their optical clarity, allowing real-time, high-resolution optical imaging of how cell responses to environmental cues [11].

Organ-on-chip platforms are designed to enhance preclinical models for analyzing responses to novel drug compounds in some organs [18]. With the increment of organoid models’ size and volume, the distance between the core and the surface which contacts the fresh medium is larger, and the growing cells cannot get enough oxygen and nutrients from the simple diffusion process. Also,

this process limits the amount of waste being removed from cells in the core. However, only cells in contact with fresh medium can survive. In these systems, the most important thing is to build structures and parameters that are similar to the conditions in the body. The use of microfluidics technology allows people to perform tissue culture in controlled and adjustable environments to optimize temperature, pH, nutrient supply, and waste disposal. Organ-on-chip systems can also be equipped with diverse sensors and actuators and integrated with them, so that they can more accurately monitor and control key parameters in the human body, various physical and electrical stimuli, nutrition exchange, etc. [19].

With the advance in modern techniques, various companies of the pharmaceutical industry have shown exponential achievements in developing more effective and lower cost drug discovery models just in recent years. And relying on this organ-on-chip technology, the integrated organ systems, such as heart, lungs, liver and so on, are generated to provide an optimal in vitro model for drug discovery [20] [21] [22] [23].

Take the lung organ-on-chips for example. Alveolar tissue and blood vessels are usually essential in the study of pulmonary drug discovery [20]. Therefore, in the lung models currently being studied for disease pathogenesis and drug screening, alveoli, the minimum functional unit of the lung, are used to simulate cyclic expansion by mechanical stretch to the gas-exchange surface [20] [24]. Researchers will use these chip models to address previous problems, such as a lack of interaction between the immune system and lung tissue, which may contribute to some compelling breakthroughs and could change the prospects for finding promising future treatments for diseases [20] [24]. Take the lung organ-on-chips for example. Alveolar tissue and blood vessels are usually essential in the study of pulmonary drug discovery [20]. Therefore, in the lung models currently being studied for disease pathogenesis and drug screening, alveoli, the minimum functional unit of the lung, are used to simulate cyclic expansion by mechanical stretch to the gas-exchange surface [20] [24]. Researchers will use these chip models to address previous problems, such as a lack of interaction between the immune system and lung tissue, which may contribute to some compelling breakthroughs and could change the prospects for finding promising future treatments for diseases [20] [24]. For instance, in 2017, Benam and other researches created a microfluidic device that interfaces a differentiated mucociliary bronchiolar epithelium, through engineering a small airway, an air-blood barrier, to study of physiological and pathophysiological mechanisms. This model avoids the recent main limitation of the airway mucosa, generalizing the roles of the true interactive immune system and endothelial tissues. And those tissues undergo shear stresses from blood flow, simulating as the airway. This improved lung-on-a-chip model makes it possible for further study of lung diseases related to immune systems, like comorbidities of chronic obstructive pulmonary disease (COPD) and asthma [20] [24] [25]. In other words, several the resulting lung-on-a-chip models contribute to the development of the biologically inspired OOCs we have today [3] [24] [25].

The final goal of almost all biomedical researches is to understand the pathology of human disease, thereby developing more effective patterns of diagnosis, prevention, and treatment. However, researchers cannot do basic diseases examinations with human bodies, due to kinds of reasons, so they have to rely on other alternative investigating models which have the same physiological environment as human disease [34]. The traditional preclinical drug development strategy that has been well acknowledged globally to predict the pharmacological parameters, and toxicological issues, is to utilize traditional simplified in vitro cell culture dishes and animal models [34] [35]. Though people have learned a lot about precious conclusions related to human disease from the disease models in animals, such as transgenic mice, some models fail to entirely and accurately reconstitute the human condition and exist several limitations separately [34]. As for the cells model, when the cells are taken out of their native environment, they will not work as to how they function in the human bodies, owing to the environmental differences between the dishes and in the body. In terms of animal models, the obtained data from animal models cannot be precisely extrapolated to humans because it sometimes results in the distinct safe starting doses for clinical trials due to vast differences in different species’ genomes.

In addition, traditional methods gradually cannot satisfy the growing demand for new and effective drugs. What’s more, during using traditional disease models, drug development will be time-consuming because of those inaccuracies caused by different species’ genes as well as raising ethical concerns [35]. In a word, traditional methods cannot reproduce the functions of the complicated biological structures, preventing people from getting timely and accurate analysis about diseases. These results have brought a heavy burden on the health care sector in the whole society and subsequently overwhelming the drug development progress [35].

Superior to the limitations of traditional drug discovery methods, organ-on-chip technology offers a new approach to study human physiology in an organ-specific context, and allows us to innovate in vitro disease models and replicates the native tissues and various cell-cell interactions via engineered biomaterials and microfluidic technologies, accelerating the assessment process of drugs [5] [35] [36].

In evaluating drug efficacy and safety, organs-on-chips have significant advantages over other in vitro cell culture models. 2D culture models are easier to be made and can provide a wealth of relatively inexpensive and non-time-consuming data. They contribute significantly to the early-stage assessments of drug toxicity, while these models fail to predict and analyze the response of real tissues to drugs and certain toxins in vivo. For 2D culture models poorly represent complex pathophysiology in patients, investigators need extra computer models and systematical biology approaches to analyze the drug response [1] [37] [38] [39]. Nevertheless, the advent of tissue engineering raises new possibilities for the research of complicated physiological and pathophysiological processes in vitro [38]. With the help of biomimetic 3D tissue structures with physiological barrier function, cell culture models recreate the interwoven set of biochemical and mechanical cues in the cellular microenvironment and recapitulate the complicated interactions between different cells in vivo, which can more accurately simulate and control drug delivery and infiltration in vivo [1] [38]. Therefore, unlike 2D culture models, organ-on-chips attempt to mimic complex in vivo conditions using a combination of microfluidics and biology, clinically relevant disease phenotypes and pharmacological responses that arise from structural and functional integration of multiple tissue types [1].

Compared with animal models, organs-on-chips also show advantages in the analysis of complex physiology issues. Based on recent systematic studies of the predictive value of traditional animal models, due to key genetic, molecular, immune, and cellular differences between humans and animals, the poor correlation of data between different species underlines the urgency for novel methods to simulate intricate human-related conditions [1] [40] [41] [42].

Biomaterials and structures used in organ-on-chips also are critical merits. Organ-on-chips adopt PDMS materials. These materials have outstanding optical transparency, which allows investigators to get direct real-time visualization and quantitative high-resolution analysis of various biological processes, which is impossible in animal models [1]. Besides, the special design of separation channels for co-culture of organ-on-chips, for example, the upper and lower channels are separated by thin porous membranes, enables fluid flow into different tissues independently in a single device, and combines with the techniques of microfluidics and bio-microelectromechanical systems (BioMEMS) to create platforms where the cellular microenvironments and microenvironmental factors can be precisely controlled [1] [43].

In organ-on-chip, scientists will sample from specific human tissues to culture in conditioned media to analyze their metabolites and other secreted products, which may help drug testing [1] [11] [43]. However, each patient has unique genes, thereby having diverse tissue types. The different genes will cause patients to respond differently to drugs. Although patients exhibit similar symptoms, they may have different underlying causes for the disease. Yet, generally, based on their similar symptoms, they are more likely to receive the same medication, which could lead to the significant risk of taking the wrong therapy [44] [45]. Owing to drug developments related to various aspects, except individual genetic variations, many factors could hamper and change the way of drug development, such as epigenetic and environmental factors. These problems would have tremendous impacts on a patient’s response to treatment, which remains to an extent unknown. Recently, OOC technology has drawn attention to be used to replace traditional drug discovery models and become personalized disease models by reflecting the genetic characteristics of cells in each patient [36]. While OOCs nowadays have generated different types of chips from many types of human and animal stem cells, researchers still face the challenge of predicting the personal safety risks of new heterogeneous organisms, in part due to the constrained availability of human cells to assess tissue-specific toxicity [36] [44]. But the advance in the production of human-induced pluripotent stem cells (hiPSCs) has the prospect to fill this gap. Since hiPSCs which are gained from patient’s tissue or be directly obtained as pathogenic cells from patients can be cultured and differentiate into major and different cell types. These cells will be cultured in the devices (OOCs) that are made by various engineered biomaterials and 3D microfabrication techniques. The drug candidates will be tested and analyzed through a series of studies and in vivo tests in animals. After that, based on the responses received from the disease models to the animals, we determine the type of drug and dosage, therefore developing and generating personalized drugs for specific patients [36]. iPSC technology provides resources derived from the patients themselves for disease-specific drug development models, and hiPSC-integrated OOCs give the possibilities of imitating human physiology for each individual, which establishes personalized drug testing platforms and generates a much better predictive tool (Figure 2) [36] [44] [45].

Despite significant progress in creating organ-on-chip and microengineered tissues, effective drug toxicity testing actually involves the implementation of each organ and its interactions. Thus, all organs are functionally needed to be integrated into the human body in the future, establishing a microfluidic circulatory system, and researchers still have a lot of work to do in the development of complex and complete models that rebuild the metabolism and physiology of the entire organs [5] [43].

Thus, researchers have further put forward the concept of “human-on-chip” models. Interconnected isolation chambers compose the whole model, each compartment containing different cell types representing different human organs, and are connected by the microfluid circulation system in a microfabrication bioreactor. It can be closer to the physiological fluid flow conditions, realistic size ratios and multi-tissue interactions in ideally, which may help in developing associated with the physiological pharmacokinetic model [5] [53] [54]. The concept of man on the chip is shown in Figure 3. Biomimetic microsystems representing different organs are integrated into a microfabrication bioreactor and connected in a physiologically relevant way to simulate complicated and dynamic drug processes of absorption, distribution, metabolism, and excretion, allowing for more reliable evaluation of drug efficacy and toxicity [5].

The primary requirement for organ-on-chip systems in the future to be the major new drug screening and discovery methods is to accurately characterize the response of these microchips to pharmacological modulation and to verify their ability to predict the response of drugs with well-defined characteristics to humans. A key challenge will be to determine the best source of human cells for accurate response in the body. The second important issue lies in combining the scale-up of complex techniques with the integrated engineering systems, such as microengineered organ simulators with sensors to detect and measure various light, chemical, electrical and mechanical signals from cells to analyze their structures and functions. For another example, more reliable operations should rely on the development of automated control of flow and pressure by microfluidic

devices equipped with microengineered valves and pumps, which paves the way for multiplexing these microsystems for high-throughput analysis and drug screening [5] [55] [56].

At present, laboratories are working hard to link more mechanically active organ-on-chip models through microfluidics technology to provide more complete biomimetic techniques for physiology, therefore achieving the goal of human-on-chip [5].