Engineering challenges in microphysiological systems

Modeling human physiology with the microphysiological systems In humans, tissues and organs are made of hierarchically assembled structures of mul­ tiple compositions to achieve biological func­ tions. The different tissues and organs are then organized in a specific order enabled by a circuitry of vascular network, further achieving physiological interactions. On the one hand, many of these complex elements cannot be readily reproduced on the con­ ventional planar, static cell culture systems already used in biology and medicine for over a century [1]. On the other hand, while pre­ clinical animal models are both biologically and physiologically capable, their relevancy to the human system remains question­ able, often leading to inaccurate clinical translation of assay results [2]. To this end, microphysiological systems, which are miniaturized biomimetic in vitro human tissue and organ models built from a combination of biological and engineering approaches, usually feature much higher per­ formance in recapitulating the functions of their in vivo counterparts [3–7]. A microphysio­ logical system usually consists of three main aspects: the organoid that is biologically relevant, often generated through principles including developmental biology [8], tissue engineering [9], bioreactor designs [10,11] and their combinations; the biophysical cues that represent the local niches, such as the sup­ porting matrix with tissue­matching proper­ ties [12], shear stress/interstitial pressure [13] and biomechanical strains [14]; and the cir­ culatory system that brings in physiological relevance, to enable communications among different organoids [5]. The ability to interconnect multiple organ­ oids together in a fashion so that the integral platforms reproduce their desired human physiology, forms the unique strength of the microphysiological systems. This is because of the fact that in the human body no single tissue or organ is isolated, where the function and behavior of one is closely dependent on the others through biophysical and biochem­ ical interactions. Sung and Shuler piloted the micro cell culture analog devices where up to ten types of organs were integrated to study their interactions [15]. Wikswo proposed the microphysiological systems with built­in valves to control the ‘blood flow’ among dif­ ferent organs [4]. More recently, a platform was developed to support long­term cocul­ ture of multiple human organoids [16], and a portable, reconfigurable multi­organ system with onboard microfluidic flow control has been reported [17]. Further development of Engineering challenges in microphysiological systems

"Microphysiological systems are emerging models of human physiology with broad applications in biomedicine, which are yet associated with several major engineering challenges that need to be addressed".

Modeling human physiology with the microphysiological systems
In humans, tissues and organs are made of hierarchically assembled structures of mul tiple compositions to achieve biological func tions. The different tissues and organs are then organized in a specific order enabled by a circuitry of vascular network, further achieving physiological interactions. On the one hand, many of these complex elements cannot be readily reproduced on the con ventional planar, static cell culture systems already used in biology and medicine for over a century [1]. On the other hand, while pre clinical animal models are both biologically and physiologically capable, their relevancy to the human system remains question able, often leading to inaccurate clinical translation of assay results [2].
To this end, microphysiological systems, which are miniaturized biomimetic in vitro human tissue and organ models built from a combination of biological and engineering approaches, usually feature much higher per formance in recapitulating the functions of their in vivo counterparts [3][4][5][6][7]. A microphysio logical system usually consists of three main aspects: the organoid that is biologically relevant, often generated through principles including developmental biology [8], tissue engineering [9], bioreactor designs [10,11] and their combinations; the biophysical cues that represent the local niches, such as the sup porting matrix with tissuematching proper ties [12], shear stress/interstitial pressure [13] and biomechanical strains [14]; and the cir culatory system that brings in physiological relevance, to enable communications among different organoids [5].
The ability to interconnect multiple organ oids together in a fashion so that the integral platforms reproduce their desired human physiology, forms the unique strength of the microphysiological systems. This is because of the fact that in the human body no single tissue or organ is isolated, where the function and behavior of one is closely dependent on the others through biophysical and biochem ical interactions. Sung and Shuler piloted the micro cell culture analog devices where up to ten types of organs were integrated to study their interactions [15]. Wikswo proposed the microphysiological systems with builtin valves to control the 'blood flow' among dif ferent organs [4]. More recently, a platform was developed to support longterm cocul ture of multiple human organoids [16], and a portable, reconfigurable multiorgan system with onboard microfluidic flow control has been reported [17] these biomimetic microphysiological systems are antic ipated to enable the construction of viable 'humanon achip' platforms, which together with the advances in stem cell technologies could make personalized preci sion medicine possible. However, several major engi neering challenges remain along the journey to achieve this ambitious goal.

Biofabrication: making better organoids
Generating functional organoids typically requires deep biology, including those either derived from mature tissues or differentiated from stem cells with refined protocols [18,19]. Since the late 2000s, biologists have grown a wide variety of nearphysiological organ oids, such as cerebral cortex, intestine, optical cup, pituitary gland, kidney, liver, pancreas, neural tube, stomach, prostate, breast, heart and lung [19]. While they behave similarly to their in vivo counterparts, the processes of organoid generation based on principles of developmental biology and cell selforganization are usually lengthy at relatively low throughputs. On the other hand, tissue engineers have been able to adapt techniques used in engineering functional tis sues previously aimed for regeneration purposes to also build miniaturized tissue models [20], which are often faster and more convenient than biological meth ods, although the generated tissues are in general less functional. It is perhaps the convergence of these two complementary approaches combining biology with engineering techniques (e.g., micro/nanotopogra phy, photolithography and bioprinting) to impart bio mimetic architectures that would eventually benefit the faithful production of human organoids at shorter turnover times and higher throughputs.

Blood surrogate: keeping them all functional
To ensure proper functions of a microphysiological system, a common medium, termed the blood sur rogate, must be used to perfuse its entire circulation in an effort to maintain the system's homeostasis by transporting nutrient, oxygen and waste molecules to and from each organoid, and support the functional ity of each organoid [4,21]. The blood surrogates can be developed by carefully tuning the compositions of key components in the culture medium for each individual cell type, which is nontrivial, considering that every organoid is different in its cell types, every microphysiological system is distinctive in its organoid types and every configuration of the microphysiologi cal system is unique. The complexity for such develop ments exponentially expands as the number of organ oids to be connected in a microphysiological system is increased, further necessitating meticulous optimi zation processes. In addition, while most cell culture media contain serum, future advances in the blood surrogate are progressively trending toward the use of additives of welldefined compositions to improve the standardization of these formulations.

Scaling: ensuring the right proportions
Scaling is another critical consideration in the design of microphysiological systems [21,22]. This can be divided into two closely related levels: the scaling between the organoids and their human counterparts and each other, which determines whether the microphysiologi cal systems could accurately model the basic human physiology [21], as well as the scaling of the biophysical parameters including, for example, the compartmen talization of the organoids in a fluidic circuitry, the flow parameters and the intake/excretion pathways, where pharmacokinetics and pharmacodynamics of mol ecules may be simulated [16,22]. To date, several direct and allometric scaling laws have been proposed such as those by organ weight, volume and blood flow [21]. More recently, a multifunctional scaling approach was further introduced to design microphysiological sys tems based on mechanistic modeling of the biological mechanisms, specification of an objective function and identification of design parameters [23], which in com parison, could be more precisely tuned toward each specific application.

Bioanalysis: monitoring the responses
The microphysiological systems are selfcontained, similar to the human body. Under this scenario, while it is still necessary to conduct 'biopsies' as the (mini mally) invasive or endpoint assays, a more intuitive approach is to sample the 'blood' in situ for longterm analysis of the responses of the organoids, the same way how the patients are examined for labs in the hos pital. The most convenient approach falls to the design of a sampling outlet that can be activated to sample certain amounts of medium at predetermined time intervals for offchip assessments such as the conven tional enzymelinked immunosorbent assay (ELISA) and mass spectrometry (MS). Alternatively, it would make the procedures more physiologically relevant by integrating onboard biochemical sensors with the microphysiological systems for monitoring the analyte levels produced by the organoids, to reduce labor and variations due to human errors [24]. Physical sensing units should further be included to allow for precise assessment and reverse modulation of physical micro environments in which the organoids reside [25]. The capability to introduce various imaging abilities that enable noninvasive and in situ characterization of the organoids is of equal importance to understand their behaviors [24].

Conclusion
Microphysiological systems have emerged as enabling platform technologies for modeling human physiol ogy, featuring structural and functional analogy to ensure biological relevance as well as physiological simulation through compartmentalized microfluidic interconnections. As such, they are potentially much more realistic than the conventional planar, static cell culture models and closer to the human system than animals. It is anticipated that, by further overcom ing several major challenges associated with building the microphysiological systems, they will eventually find widespread biomedical applications across a vari ety of fields spanning from basic cell biology to drug development.

Financial & competing interests disclosure
The author acknowledges support from the National Cancer Institute of the National Institutes of Health Pathway to Independence Award (K99CA201603) and the LUSH Prize. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.

Open access
This work is licensed under the Creative Commons Attribution 4.0 License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/