Microfluidic-based biomimetic models for life science research

Keqiu Jiang , Chengyong Dong, Yakun Xu and Liming Wang*
Division of Hepatobiliary and Pancreatic Surgery, Department of General Surgery, The Second Affiliated Hospital of Dalian Medical University, 467 Zhongshan Road, Dalian, Liaoning 116027, China. E-mail: wangbcc259@sina.cn

Received 3rd March 2016 , Accepted 6th March 2016

First published on 8th March 2016


The advances in microfluidic technology have recently generated various microfluidic-based biomimetic models as novel three-dimensional (3D) models for life science research, offering some great advantages over conventional two-dimensional (2D) models, classical scaffold-free or scaffold-based 3D approaches and animal models. These biomimetic models could simulate the microenvironment of in vivo tissues and organs by controlling spatiotemporal gradients of chemical substances and imitating mechanical activities of living tissues and organs. They provide platforms for real-time observation of physiological and pathological processes, toxicology effect and drug effects in tissues and organs for life science research. Without a doubt, microfluidic-based biomimetic models would serve as a powerful tool for developing diagnosis and treatment methods for various diseases. In this study, we briefly summarized the fabrication of microfluidic-based biomimetic models and their use as 3D tissue/organ models.

1. Introduction

For decades, conventional 2D models (i.e. cell plate culture) have served as important platforms for life science research including cell biology, biochemistry, clinical laboratory medicine and other fields. The functions of various cells have been well-investigated via these 2D models through cultured cells or cellular products. However, conventional 2D models could not exactly simulate living tissue/organ conditions such as physiological phenotypes, the interaction and reaction of cells and so on. False positive or negative results from drug screening, failure of clinical trials and omission of effective drug candidates might be attributed to the incapability of 2D models used in drug screening.1 Traditional 3D models have been proposed to overcome the shortcomings of 2D models by culturing cells using scaffold-free or scaffold-based 3D approaches in vitro. Cells can migrate and grow in these traditional 3D models. Enhanced expression of cellular differentiation function and improved organizational structures have been observed when cells were cultured in extracellular matrix gels.2 Tumor cells grown in 3D scaffolds have different morphology,3 more diverse expression of cell surface receptors,4 various gene response of angiogenesis, cell migration and invasion,5–7 and different extracellular matrix synthesis8 other than those in conventional 2D approaches. However, these traditional 3D approaches continue to present some difficulty in simulating biological characteristics and functions of tissues and organs in vivo due to the lack of spatiotemporal gradients of important chemical substances, as well as mechanical activities of living tissues and organs.9 Time-consuming and costly animal experiments have to be conducted in order to simulate the processes of various diseases. Sometimes, the results from animal experiments could not be reproduced in human trials due to specie differences.10 Microfluidic-based biomimetic models have been viewed as the potential solution for issues observed for conventional in vitro models. Different micro-scale functional units (also known as organ-on-a-chip) have been designed and investigated on the basis of the advances in microfluidic technology, biomaterials and tissue engineering.

Microfluidic technology is also known as lab-on-chip (LOC) or micro total analysis system (μTAS). This technology has been widely used in various areas of life science research, including gene analysis, protein analysis, single cell analysis, drug screening, drug toxicology effect, multi-drug resistance, pathogen detection, cell culture and tissue engineering.11–13 Microfluidic technology and conventional 3D approaches have been integrated to form microfluidic-based biomimetic models via precision fabrication technology. A dynamic microenvironment could be created in a controllable manner in microfluidic-based biomimetic models via micro-channels or the micro fluid to provide stable nutrients and other chemical substances. Moreover, optical, acoustic, mechanical and electromagnetism technologies could be integrated in these models to simulate the mechanical activities of tissues and organs in vivo.9,12,14–16 Microfluidic-based biomimetic models have several unique features. Their micro-scale dimensions are consistent with the microstructure and microenvironment in vivo, thus easily creating micro-environmental conditions similar to the human body. Furthermore, biomimetic models can greatly reduce cost and improve detective efficiency due to the small amount of samples or reagents needed, thus serving as high throughput platforms for drug screening.11,17 Moreover, microfluidic-based biomimetic models make it possible to observe important physiological and pathological processes in real-time without using expensive animal models.

In a brief literature search using PubMed, 419 papers published since 2013 were found to have used microfluidic and tissue/organ models as their keywords. Fig. 1 summarizes the categories of these papers according to the tissue/organ models addressed and/or depicted in these papers. As shown in Table 1, microfluidic-based biomimetic models have been used in different areas in life science. Each model could be used for multifunctional applications. It is clear that there is a strong trend of shifting from conventional models to microfluidic-based biomimetic models. In this study, we reviewed literatures on the fabrication of microfluidic-based biomimetic models and current in vitro 3D tissues/organs generated from microfluidic-based biomimetic models in the field of life science. We briefly summarized the technologies and materials used for the fabrication of microfluidic-based biomimetic models and current in vitro 3D tissues/organs generated from microfluidic-based biomimetic models. For details on the specific applications of microfluidic-based biomimetic models such as drug screening, cancer research and so on, the readers should be referred to the previous review papers.18–21

image file: c6ra05691a-f1.tif
Fig. 1 Overview of the 419 articles published since 2013, which combined 3D cell culture with microfluidics.
Table 1 Biomaterials, features and applications of microfluidic-based biomimetic models reported in literature
Materials Characteristics Refs Models Applications Advantages Refs
Silicon High purity, relatively low cost, and capabilities for heat dissipation, intensity, and corrosion resistance; weak insulation, low transmittance, and low successful bonding rate; 11,24 and 29 Blood vessels Vascular network system, vascular barrier function, platelet function, hemostatis and thrombus formation, tumor extravasation, tumor migration, support effect of arteriovenous system, drug screening Low cost, mass production, high throughput, high efficiency, dynamic control, real-time observation 1,12,24,31–36
Nervous system Nervous damage, nervous regeneration, function of blood brain barrier, drug screening 31,37–41
Glass High strength, chemical inertness, better heat resistance, better transmittance and better insulation; poor oxygen permeability; Breast Drug screening, metastasis of breast cancer 42 and 43
Lung Lung injury and pulmonary edema, effect of macrophages in lung tumor 44–46
Polymer Low cost, easily processing molding, allowing mass production 1,17,19,23,24,28 and 30 Heart Myocardial tissue structure and function, drug screening, cardiac neural regulation 45,47–53
PDMS (polymer) Nontoxicity, good air permeability, biocompatibility and surface hydrophobicity, transmission of light, elasticity and stable temperature gradients Liver Metabolism, drug screening, toxicity effect, sinusoidal barrier function 1,9,14,54–58
Intestine Selective barrier function, epithelial–endothelial interface, ADMET, drug screening, toxicology 1,14,59–61
Hydrogel (polymer) 3D network hydrophilic polymer, high water contents and flexible structure, good biocompatibility, low immunogenicity, good maneuverability Kidney Renal reabsorptive barriers, renal toxicity, renal metabolomic, biological transport phenomenon, underlying disease mechanisms, drug toxicity 62 and 63
Tumor Tumor metastasis, circulating tumor cells behavior, anti-cancer drug screening, tumor treatment 55,64 and 65

2. Fabrication of microfluidic-based biomimetic models

Various biomaterials such as silicon, glass, polydimethylsiloxane and hydrogels have been used in the construction of microfluidic biomimetic models (Table 1). Depending on the biomaterials used, different processing technologies have been adopted for fabricating microfluidic biomimetic models. Etching is commonly used for silicon/glass-based microfluidic biomimetic models;22 while laser ablation, injection molding, hot-embossing, photolithography, soft lithography technology and 3D printing have been adopted for constructing polymer-based microfluidic biomimetic models.23–27 The use of 3D printing technology solves some challenges faced by etching, laser ablation, injection molding, hot-embossing, photolithography and soft lithography technology. Later technologies could only produce open channels with fixed depth and vertical walls in the bonding regions of substrates. Cylindrical channels with definitive volume and depth could be generated by adjusting the distance and time of 3D printing, thus minimizing the superficial area of channels and reducing chemical adsorption.28

2.1 Silicon/glass-based microfluidic biomimetic model

Silicon is the cornerstone for the modern electronic industry due to its high purity, relatively low cost, and capabilities for heat dissipation, intensity and corrosion resistance. Silicon materials have been used in manufacturing various micro-devices such as pumps, valves and moulds with a smooth finish for microfluidic biomimetic chips.29 Deep etching is a challenging issue for silicon due to its weak insulation, low transmittance and low successful bonding rate. Glass has been used as alternative material due to its high strength, chemical inertness, better heat resistance, better transmittance and better insulation. However, glass-based microfluidic biomimetic models are limited to research which only needs an anaerobic environment11,24 due to poor oxygen permeability in these models.

It is well-known that it is difficult to obtain a proper seal and attain larger width/depth ratios of channels in silicon/glass-based microfluidic biomimetic models. However, various mature separation methods and strategies used in traditional capillary electrophoresis could be directly applied for building these models.

2.2 Polymer-based microfluidic biomimetic models

Conventional polymers used in constructing polymer-based microfluidic biomimetic models could be classified into three types: thermoplastic polymers such as polyamide, polymethylmethacrylate, polycarbonate and polyethylene terephthalate; curing polymers such as polydimethylsiloxane (PDMS), epoxy resin and polyurethane; solvent evaporation-type polymers such as acrylic, rubber and fluorine plastics. Processing technologies for fabricating polymer-based microfluidic biomimetic models depend on the type of polymer used. Currently, extensively used processing technologies include laser ablation, injection molding, hot-embossing, photolithography and soft lithography technology. PDMS has been demonstrated to have good properties such as nontoxicity, good air permeability, biocompatibility and surface hydrophobicity. Both visible light and ultraviolet light are capable of passing through PDMS materials.17,24,30 Furthermore, PDMS polymers are integrated with outside components for better elasticity and stable temperature gradients. Moreover, monolayer or multilayer PDMS moulds are frequently fabricated in many microfluidic devices.28

Hydrogels are 3D hydrophilic polymer networks formed via physical or chemical crosslinking approaches. Hydrogels are also widely used in manufacturing microfluidic biomimetic chips due to their high water content and flexible structure. Among these hydrogels, polyethylene glycol (PEG) hydrogels are widely used due to good biocompatibility, low immunogenicity and good maneuverability. Functional PEG hydrogels are formed from PEG acrylic ester derivatives.

Photolithography is often used to generate microchannels or micropores in polymer-based microfluidic biomimetic chips, for which specific masks are designed to have the desired micro-channels or micro pores that allow light to pass through. These masks are placed over hydrogel precursor layers. Microstructures with mask patterns are produced on hydrogel substrates after exposure to light irradiation.23 Through controlling exposure space and time, hydrogels with diverse structures could be produced in microfluidic biomimetic chips. In addition to photolithography, soft lithography and molding technology are also widely used in fabricating microfluidic biomimetic chips. Soft lithography can fabricate more complicated 3D microstructures than photolithography via the aid of PDMS masks with specific patterns.1 Substrates with irregular surface shapes and different chemicals could be used in soft lithography to meet the different chemical properties needed on the surface. Crosslinked hydrogels could be easily separated from PDMS moulds23 due to their difference in hydrophilicity and hydrophobicity.

Multi-functional hydrogel-based microstructures can be manufactured by using emulsion systems, flow lithography, and microfluidic fiber spinning technology.19 Complex 3D structures have been built in microgel form as a structure module for tissue engineering.23

3. Microfluidic-based biomimetic models as 3D tissue/organ models

3.1 Blood vessel model

Blood vessels are pipelines in the human body for transporting substances, and bridges for connecting different tissues and organs. Researchers have adhered endothelial cells to the channel extracellular matrix (ECM) wall, and further formed a confluent monolayer tube (Fig. 2A) for simulating the human vascular network.31,32 Resulting microfluidic-based biomimetic models that consist of independent compartments are capable of fluid function and infusions, and serve as models for investigating barrier function33 (Fig. 2B), platelet function, hemostasis and thrombus formation34 (Fig. 2C), as well as for observing the effects of drugs, toxins, growth factors, cytokines and other bioactive molecules.31 These blood vessel models have also been used in studying angiogenesis, the interference among perivascular cells, and the interaction between blood components and endothelial cells.12 Moreover, as barrier models, blood vessel models have been widely adopted in studies of the blood brain barrier, tumor escape, tumor extravasation35 (Fig. 2D) and tumor migration.36 Through the growth and self-assembly of endothelial colony forming cell derived endothelial cells and stomal fibroblasts in PDMS devices that consist of micro channels and tissue chambers, the perfused microvascular network models can successfully simulate the support effect of the arteriovenous system for tissues and organs1 (Fig. 2E). Blood vessel models can be combined with other special tissues to fabricate some major vascularization systems such as the heart. Through the heart model, heart failure, high blood pressure, low blood pressure, cardiac arrhythmias and normal cardiac function could be mimicked by changing the conditions of fluid flow in microfluidic channels. Microfluidic-based biomimetic blood vessel models as new research platforms have successfully advanced drug screening and clinical trial research for cardiovascular disease.24
image file: c6ra05691a-f2.tif
Fig. 2 Blood vessel models: (A) schematic of a microfluidic system used to mimic the stepwise interaction of endothelial and pericyte cells;32 (B) a schematic of a microfluidic vascular model for the 3D culture of tumor and endothelial cells under fluid flow;33 (C) a schematic of a hemostasis monitor device and a photograph of three hemostasis monitoring devices formed in a single piece of PDMS mould;34 (D) a schematic of vascularized organotypic microfluidic assays for studying breast cancer cell extravasation;35 (E) a microfluidic blood vessel model as a drug screening platform (individual modules can be connected by an inter-module connector, forming a continuous microvascular network that links all tissues and organs via connector valves).1 (The use of these figures, except for (D), was permitted by the publishers of the references cited. (D) was plotted according to the reference cited.)

3.2 Nervous system model

The nervous system consists of the brain, spinal cord, cranial nerve, spinal nerve, vegetative nerve and various ganglions. The nerve system coordinates the activities of different tissues and organs, and senses the outside environment. Thus, microfluidic-based biomimetic nervous system models are formed by connecting an axon chamber and a neuron chamber using microfluidic channels. Nervous system models provide new efficient platforms for investigating damage, degradation and regeneration in the nervous system. These models could be used to easily simulate chemical or physical nerve damage by controlling the system microenvironment or with laser or nano knife technology37 (Fig. 3A). On the other hand, regeneration models for the nervous system are created using microfluidic channels to block two independent cell culture chambers38 (Fig. 3B). Yin et al. studied the effective concentration of local tacrolimus for peripheral nerve regeneration by using a PDMS microfluidic device that consist of a pyramid-shape concentration gradient generator (CGG) and cell culture chambers39 (Fig. 4). Moreover, in vitro models of the blood brain barrier could be constructed by combining nervous system models and blood vessels structures to study the trafficking of drugs, nutrients, metabolite and other substances between blood and the central nervous system40,41 (Fig. 5). These microfluidic-based biomimetic models would benefit studies for nervous system diseases such as Alzheimer's disease and disseminated sclerosis due to dysfunction of the blood brain barrier.31,39
image file: c6ra05691a-f3.tif
Fig. 3 Nervous system models: (A) a schematic of nervous system damage models that consist of two isolated chambers, the axonal side and the somal side across micro-channels (in the circular microfluidic platform, green denotes somal compartments, blue denotes axonal compartments and red denotes micro-channels).37 (B) A schematic of a nervous system regeneration model that consist of two separate neuronal culture chambers (blue) interconnected through a series of asymmetrical micro-channels. (These micro-channels are interrupted by a central narrow channel that gives access to the central part of the axons [yellow]. Each chamber is individually perfused by two reservoirs.)38 (The use of these figures, except for (A), was permitted by the publishers of the references cited. (A) was plotted according to the reference cited.)

image file: c6ra05691a-f4.tif
Fig. 4 A microfluidic device designed for screening effective concentrations of FK506 is shown. The microfluidic device consisted of an upstream concentration gradient generator and a downstream cell culture chamber. The medium could be gradually diluted and forms a special gradient of FK506 via inlet 1 and inlet 2. Schwann cells (SCs) were loaded into the device and cultured for nine days in the presence of FK506 solution39 (the use of this figure was permitted by the publisher in ref. 39).

image file: c6ra05691a-f5.tif
Fig. 5 A neurovascular microfluidic bioreactor (major components, cell types and their spatial arrangement are indicated in the schematic of the neurovascular unit) for recreating blood-brain barrier physiology and structure on chip40 (the use of this figure was permitted by the publisher in ref. 40).

3.3 Breast model

Novel microfluidic-based biomimetic models have been constructed as high throughput platforms for anti-cancer drugs screening, and as models for breast cancer metastasis-related research. Choi et al. proposed a breast tumor globule by culturing ductal epithelial cells and fibroblasts in 3D ECM scaffolds. This breast model successfully functioned as ductal carcinoma in situ (DCIS) for evaluating the effectiveness and toxicology effects of anti-cancer drugs42 (Fig. 6A). Furthermore, Bersini et al. cultured ectomesenchymal stem cells and bone marrow cells in microfluidic-based ECM scaffolds that cover vascular endothelial cells in microfluidic channels. By combining with breast cancer tissues, this model could simulate breast cancer bone metastasis43 (Fig. 6B).
image file: c6ra05691a-f6.tif
Fig. 6 Breast model: (A) a breast cancer-on-a-chip micro-device that consist of upper and lower cell culture chambers separated by an ECM-derived membrane (DCIS spheroids were embedded in the mammary epithelium in the upper channel, while the fibroblast-containing stromal layer was cultivated on the other side of the membrane).42 (B) A schematic of a bone metastases model of breast cancer (first, hBM-MSCs [brown] were cultured for 14–21 days using osteogenic medium in microfluidic devices; second, the ECM [yellow filaments] started to be deposited in the devices; third, a monolayer covering the media channel was created by seeding ECs [red] after three days; finally, cancer cells [green] were introduced after three additional days, and their extravasation ability and micrometastasis generation were monitored for 1–5 days).43 (The use of these figures was permitted by the publishers in ref. 42 and 43.)

3.4 Lung model

Serious pulmonary diseases have become major killers for humans, worldwide. Huh et al. reconstructed a human alveolar capillary interface in vitro using microfluidic technology. This microfluidic-based biomimetic model could simulate the normal breathing process by controlling air, fluid and cyclic stress in micro-channels.44 By culturing lung epithelial cells, endothelial cells and immune cells on microfluidic chips, microfluidic-based biomimetic lung injury and pulmonary edema models could be fabricated to simulate the development of pulmonary edema in vivo, including the alveolar fluid accumulation, fibrin deposition and gas exchange function damage.45 Microfluidic-based biomimetic lung models are currently used in exploring pathogen, toxicology, the self-defense process and pulmonary disease treatment. Furthermore, Li et al. created a biomimetic airway model by adding a permeability membrane between two PDMS layers with bronchial epithelial cells growing on the top of membrane and macrophages growing on the opposite side of the membrane.46 Circulating sterile air flowed in the upper channel, and the medium of culturing cells flowed in the lower channel. These biomimetic lung models were used to study the behavior and effect of macrophages in lung tumors46 (Fig. 7).
image file: c6ra05691a-f7.tif
Fig. 7 Biomimetic airway model: the biomimetic airway model consists of three vertical compartments with two different cells chambers (two polydimethylsiloxane layers) separated by porous membranes. Bronchial epithelial cells were grown on the top of the membrane, and macrophages were grown on the opposite of the membranes. Circulating sterile air flows in the upper channel and the cell culture medium flows in the lower channel46 (the use of this figure was permitted by the publisher in ref. 46).

3.5 Heart model

Kim et al. reported the simulation of the structure and function of natural myocardial tissues by exploiting hydrogels.47,48 The constructed ECM structure was similar to the resident environment of cardiac cells in the human body, offering an engineered tool for heart tissue engineering research and stem cell-based therapy for heart diseases. However, this model had some shortcomings, in which it could not be used for evaluating cell source variability and parameters such as mechanical, soluble and electrical stimulation.45 Legant et al. created mini-engineered heart tissue (m-EHTs) by implanting artificial heart tissue within the collagen/fibrin structure.49–51 They were able to successfully use the m-EHTs for low-dose and high-throughput drug screening. Furthermore, Agarwal et al. simulated heart functional responses to cardiac drugs using microfluidic-based heart-on-a-chip for high-throughput pharmacology research, and explored the relationships of their structure and function.52 By co-culturing myocardial cells and sympathetic nerve cells, the functional synaptic structure between myocardial cells and neurons could be formed in the resulting biomimetic model.53 These microfluidic-based biomimetic heart models provide novel platforms for preclinical studies.

3.6 Liver model

Livers are involved in the biological transformation of many substances such as drugs, poisons and some metabolites in vivo. Many chemicals could produce serious side effects such as hepatotoxicity and liver damage, which could even lead to severe diseases. In microfluidic-based biomimetic liver models, hepatic cultures have a semblable 3D tissue microstructure and cell/nutrient ratio. Oxygen gradient, fluid rate and mass transport could be precisely controlled in the models.54 These models provide efficient, accurate, reliable and low-cost platforms for studying liver tissues and cell metabolism, drug discovery and toxicity.55,56 The challenges for manufacturing these models are to determine a means of maintaining the metabolism activity of liver tissues and cells for a long time, and to determine a means of enabling the co-existence of different cells. The solutions for these challenges might reply on stable nutrient perfusion and controlled oxygen gradient and shear stress through the use of micro pumps and micro valves.14,54 These models can simulate normal liver tissue structures, and can also be used to create disease models for monitoring the developmental and therapeutic progress of liver diseases in real-time.57 Lee et al. designed and fabricated an artificial endothelial-like barrier to simulate liver sinusoid structures and mimic sinusoidal barrier functions58 (Fig. 8). Due to the existence of various special enzymes in the liver, oxygen concentration in liver tissues has significant regulation effects. By directly controlling the supply rate and concentration gradient of oxygen in micro-channels, microfluidic-based biomimetic liver models can be used to replicate the drug metabolism process in the body,1 and simulate physiological variations of the forms and function of liver cells along the liver sinusoids, which is believed to be dependent on regional variations in oxygenation.9
image file: c6ra05691a-f8.tif
Fig. 8 Microfluidic liver sinusoid (plotted according to ref. 58): (A) schematic diagram of the hepatocyte microenvironment in liver tissues; (B) a microfluidic liver sinusoid composed of a flow inlet, a flow outlet and a cell inlet (the nutrient transport channel indicated in red color was separated by the hepatocyte culture area indicated in green color, while a set of parallel microfluidic channels indicated in blue connected the transport channel and hepatocyte culture area).58

3.7 Intestine model

The absorption of nutrition into the human body mainly relies on the intestine. Intestinal tissues comprise of three different glands that can absorb useful substance and prevent the intake of waste substances via the selective barrier function. Microfluidic-based biomimetic intestine models can filter drug compounds into the microenvironment by mimicking the selective barrier of intestines.1 Intestinal models have central channels that consist of a selective ECM-coated membrane and intestinal epithelial cells. This membrane mimics intestinal motility in the body by circularly pumping to vacuum chambers59 (Fig. 9). Kimura et al. successfully created a biomimetic intestine model by combining a special membrane and vascular fluid, simulating the intestinal epithelial barrier and epithelial–endothelial interface structure.60 Based on this model, Mahler et al. further developed a gastrointestinal tract microscale cell culture analog to study drug toxicity.61 Basic intestinal functions such as absorption, distribution, metabolism, elimination and toxicity (ADMET) could be simulated through the replication of some key cell types such as intestinal epithelial cells, goblet and vascular epithelial cells, the special structure of the villi and mucus, and the characteristics of intestine motility.14
image file: c6ra05691a-f9.tif
Fig. 9 Gut-on-a-chip: gut-on-a-chip made of two vacuum chambers and two micro-channels with basal nuclei (blue) and apical mucin (magenta) separated by a porous membrane, on which undulating epithelial cells formed villi-like structures59 (the use of this figure was permitted by the publisher in ref. 59).

3.8 Kidney model

Preclinical research on renal toxicity has mainly used 2D approaches and animal experiments. However, these methods could not fully reveal the reactions of drugs in the human body, thus leading to a high failure rate at the clinical stage for selected drug candidates. Recently, microfluidic-based biomimetic kidney models have been exploited for addressing such challenges faced by high-throughput drug screening, biomarkers of renal toxicity and renal metabolomics. For example, Jang et al. established a model by covering selective filtration membranes with renal tubular epithelial cells, and used this model for simulating the transport of drugs and metabolites.62 Frohlich et al. also created a biomimetic kidney model using topographically-patterned porous membranes for simulating the function of renal reabsorptive barriers. These kidney models would help researchers understand the biological transport phenomenon that underlies disease mechanisms and drug toxicity63 (Fig. 10).
image file: c6ra05691a-f10.tif
Fig. 10 A model of renal reabsorptive barriers: a multiple-channel microfluidic device was assembled using a topographically-patterned membrane with an intact pore structure for studying cell culture and transport63 (the use of this figure was permitted by the publisher in ref. 63).

3.9 Tumor model

For a long time, the areas of anti-cancer drugs and targeted cancer therapy relied heavily on traditional experimental models. With the advances in microfluidic-based biomimetic models, tumor cells could be cultured on a 3D scaffold structure to form spheroid, hollow fiber and layer models, thus simulating heterogeneous blood supply conditions.55 Biomimetic models can mimic the process of tumor metastasis,64 and provide efficient platforms for anti-cancer drug screening and tumor treatment. With the microfluidic-based biomimetic tumor model, Jedrych et al. investigated the specific effects of photodynamic therapy (PDT) for cancer treatment.65

4. Concluding remarks and future perspectives

Compared with traditional approaches, microfluidic-based biomimetic models can successfully simulate the microenvironment of tissues and organs in vivo (including spatiotemporal chemical gradients, mechanical stimulation, and so on), and realize the dynamic manipulation and real-time observation of important physiological and pathological processes. These biomimetic models have been used as new efficient platforms/tools for life science research such as drug discovery and developments through efficacy and toxicity studies, as well as for novel diagnostic tests that can predict treatment outcome in patients. However, challenges continue to exist in maintaining the long-term stability of various cultured human cells in biomimetic models. Currently, cell survival in biomimetic models could only be maintained for a short time. In order to address this issue, further advances are required on microfluidic technology, biological materials, fabrication technology and cell culture technology. With such advances in the future, these models are expected to be able to mimic more multifunctional tissues and organs that are consistent with the human body, and even replace damaged tissues and organs. Nevertheless, microfluidic-based biomimetic models can provide solutions to many thorny problems of clinical trials, and save a large amount of cost and time before it is absolutely needed for use in animal models. We believe that microfluidic-based biomimetic models are promising for investigating the diagnosis and treatment of all kinds of human diseases.

Conflict of interest

The authors declare no conflict of interest and no financial disclosures.


This work was supported by the funds from the National Natural Science Foundation of China (81272368, 81471755) and the clinical capability construction project for Liaoning provincial hospitals (LNCCC-B03-2014).


  1. C. Heylman, A. Sobrino, V. S. Shirure, C. C. Hughes and S. C. George, Exp. Biol. Med., 2014, 239, 1240–1254 CrossRef PubMed.
  2. F. Pampaloni, E. G. Reynaud and E. H. Stelzer, Nat. Rev. Mol. Cell Biol., 2007, 8, 839–845 CrossRef CAS PubMed.
  3. C. Feder-Mengus, S. Ghosh, A. Reschner, I. Martin and G. C. Spagnoli, Trends Mol. Med., 2008, 14, 333–340 CrossRef CAS PubMed.
  4. F. Wang, V. M. Weaver, O. W. Petersen, C. A. Larabell, S. Dedhar, P. Briand, R. Lupu and M. J. Bissell, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 14821–14826 CrossRef CAS.
  5. M. Valcarcel, B. Arteta, A. Jaureguibeitia, A. Lopategi, I. Martinez, L. Mendoza, F. J. Muruzabal, C. Salado and F. Vidal-Vanaclocha, J. Transl. Med., 2008, 6, 57 CrossRef PubMed.
  6. M. A. Wozniak, K. Modzelewska, L. Kwong and P. J. Keely, Biochim. Biophys. Acta, 2004, 1692, 103–119 CrossRef CAS PubMed.
  7. D. Yamazaki, S. Kurisu and T. Takenawa, Oncogene, 2009, 28, 1570–1583 CrossRef CAS PubMed.
  8. K. A. Beningo, M. Dembo and Y. L. Wang, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 18024–18029 CrossRef CAS PubMed.
  9. D. Huh, G. A. Hamilton and D. E. Ingber, Trends Cell Biol., 2011, 21, 745–754 CrossRef CAS PubMed.
  10. A. K. Capulli, K. Tian, N. Mehandru, A. Bukhta, S. F. Choudhury, M. Suchyta and K. K. Parker, Lab Chip, 2014, 14, 3181–3186 RSC.
  11. X. J. Li, A. V. Valadez, P. Zuo and Z. Nie, Bioanalysis, 2012, 4, 1509–1525 CrossRef CAS PubMed.
  12. M. L. Kovarik, D. M. Ornoff, A. T. Melvin, N. C. Dobes, Y. Wang, A. J. Dickinson, P. C. Gach, P. K. Shah and N. L. Allbritton, Anal. Chem., 2013, 85, 451–472 CrossRef CAS PubMed.
  13. L. Ying and Q. Wang, BMC Biotechnol., 2013, 13, 76 CrossRef PubMed.
  14. A. M. Ghaemmaghami, M. J. Hancock, H. Harrington, H. Kaji and A. Khademhosseini, Drug Discovery Today, 2012, 17, 173–181 CrossRef CAS PubMed.
  15. A. Mathur, P. Loskill, S. Hong, J. Lee, S. G. Marcus, L. Dumont, B. R. Conklin, H. Willenbring, L. P. Lee and K. E. Healy, Stem Cell Res. Ther., 2013, 4(suppl. 1), S14 CrossRef PubMed.
  16. Y. Shao and J. Fu, Adv. Mater., 2014, 26, 1494–1533 CrossRef CAS PubMed.
  17. T. James, M. S. Mannoor and D. V. Ivanov, Sensors, 2008, 8, 6077–6107 CrossRef CAS.
  18. N. S. Bhise, J. Ribas, V. Manoharan, Y. S. Zhang, A. Polini, S. Massa, M. R. Dokmeci and A. Khademhosseini, J. Controlled Release, 2014, 190, 82–93 CrossRef CAS PubMed.
  19. K. E. Sung and D. J. Beebe, Adv. Drug Delivery Rev., 2014, 79–80, 68–78 CrossRef CAS PubMed.
  20. X. Xu, M. C. Farach-Carson and X. Jia, Biotechnol. Adv., 2014, 32, 1256–1268 CrossRef CAS PubMed.
  21. M. R. Carvalho, D. Lima, R. L. Reis, V. M. Correlo and J. M. Oliveira, Trends Biotechnol., 2015, 33, 667–678 CrossRef CAS PubMed.
  22. F. He, H. Xu, Y. Cheng, J. Ni, H. Xiong, Z. Xu, K. Sugioka and K. Midorikawa, Opt. Lett., 2010, 35, 1106–1108 CrossRef PubMed.
  23. N. Annabi, A. Tamayol, J. A. Uquillas, M. Akbari, L. E. Bertassoni, C. Cha, G. Camci-Unal, M. R. Dokmeci, N. A. Peppas and A. Khademhosseini, Adv. Mater., 2014, 26, 85–123 CrossRef CAS PubMed.
  24. M. L. Kovarik, P. C. Gach, D. M. Ornoff, Y. Wang, J. Balowski, L. Farrag and N. L. Allbritton, Anal. Chem., 2012, 84, 516–540 CrossRef CAS PubMed.
  25. D. K. Maurya, W. Y. Ng, K. A. Mahabadi, Y. N. Liang and I. Rodriguez, Biotechnol. J., 2007, 2, 1381–1388 CrossRef CAS PubMed.
  26. S. Yang and D. L. Devoe, Methods Mol. Biol., 2013, 949, 115–123 CAS.
  27. C. M. Ho, S. H. Ng, K. H. Li and Y. J. Yoon, Lab Chip, 2015, 15, 3627–3637 RSC.
  28. M. K. Gelber and R. Bhargava, Lab Chip, 2015, 15, 1736–1741 RSC.
  29. S. Amer and W. Badawy, Curr. Pharm. Biotechnol., 2005, 6, 57–64 CAS.
  30. M. Zhan, L. Chingozha and H. Lu, Anal. Chem., 2013, 85, 8882–8894 CrossRef CAS PubMed.
  31. A. Tourovskaia, M. Fauver, G. Kramer, S. Simonson and T. Neumann, Exp. Biol. Med., 2014, 239, 1264–1271 CrossRef PubMed.
  32. J. Kim, M. Chung, S. Kim, D. H. Jo, J. H. Kim and N. L. Jeon, PLoS One, 2015, 10, e0133880 Search PubMed.
  33. C. F. Buchanan, S. S. Verbridge, P. P. Vlachos and M. N. Rylander, Cell Adhes. Migr., 2014, 8, 517–524 CrossRef PubMed.
  34. A. Jain, A. Graveline, A. Waterhouse, A. Vernet, R. Flaumenhaft and D. E. Ingber, Nat. Commun., 2016, 7, 10176 CrossRef CAS PubMed.
  35. J. S. Jeon, S. Bersini, M. Gilardi, G. Dubini, J. L. Charest, M. Moretti and R. D. Kamm, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 214–219 CrossRef CAS PubMed.
  36. X. Y. Wang, Y. Pei, M. Xie, Z. H. Jin, Y. S. Xiao, Y. Wang, L. N. Zhang, Y. Li and W. H. Huang, Lab Chip, 2015, 15, 1178–1187 RSC.
  37. R. Siddique and N. Thakor, J. R. Soc., Interface, 2014, 11, 20130676 CrossRef PubMed.
  38. B. Deleglise, B. Lassus, V. Soubeyre, A. Alleaume-Butaux, J. J. Hjorth, M. Vignes, B. Schneider, B. Brugg, J. L. Viovy and J. M. Peyrin, PLoS One, 2013, 8, e71103 CAS.
  39. B. S. Yin, M. Li, B. M. Liu, S. Y. Wang and W. G. Zhang, Exp. Ther. Med., 2015, 9, 154–158 CAS.
  40. J. A. Brown, V. Pensabene, D. A. Markov, V. Allwardt, M. D. Neely, M. Shi, C. M. Britt, O. S. Hoilett, Q. Yang, B. M. Brewer, P. C. Samson, L. J. McCawley, J. M. May, D. J. Webb, D. Li, A. B. Bowman, R. S. Reiserer and J. P. Wikswo, Biomicrofluidics, 2015, 9, 054124 CrossRef PubMed.
  41. H. Cho, J. H. Seo, K. H. Wong, Y. Terasaki, J. Park, K. Bong, K. Arai, E. H. Lo and D. Irimia, Sci. Rep., 2015, 5, 15222 CrossRef CAS PubMed.
  42. Y. Choi, E. Hyun, J. Seo, C. Blundell, H. C. Kim, E. Lee, S. H. Lee, A. Moon, W. K. Moon and D. Huh, Lab Chip, 2015, 15, 3350–3357 RSC.
  43. S. Bersini, J. S. Jeon, G. Dubini, C. Arrigoni, S. Chung, J. L. Charest, M. Moretti and R. D. Kamm, Biomaterials, 2014, 35, 2454–2461 CrossRef CAS PubMed.
  44. D. Huh, B. D. Matthews, A. Mammoto, M. Montoya-Zavala, H. Y. Hsin and D. E. Ingber, Science, 2010, 328, 1662–1668 CrossRef CAS PubMed.
  45. K. H. Nam, A. S. Smith, S. Lone, S. Kwon and D. H. Kim, J. Lab. Autom., 2015, 20, 201–215 CrossRef CAS PubMed.
  46. E. Li, Z. Xu, H. Zhao, Z. Sun, L. Wang, Z. Guo, Y. Zhao, Z. Gao and Q. Wang, Oncotarget, 2015, 6, 8900–8913 CrossRef PubMed.
  47. D. H. Kim, E. A. Lipke, P. Kim, R. Cheong, S. Thompson, M. Delannoy, K. Y. Suh, L. Tung and A. Levchenko, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 565–570 CrossRef CAS PubMed.
  48. D. H. Kim, Kshitiz, R. R. Smith, P. Kim, E. H. Ahn, H. N. Kim, E. Marban, K. Y. Suh and A. Levchenko, Integr. Biol., 2012, 4, 1019–1033 RSC.
  49. T. Boudou, W. R. Legant, A. Mu, M. A. Borochin, N. Thavandiran, M. Radisic, P. W. Zandstra, J. A. Epstein, K. B. Margulies and C. S. Chen, Tissue Eng., Part A, 2012, 18, 910–919 CrossRef CAS PubMed.
  50. W. R. Legant, A. Pathak, M. T. Yang, V. S. Deshpande, R. M. McMeeking and C. S. Chen, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 10097–10102 CrossRef CAS PubMed.
  51. A. Hansen, A. Eder, M. Bonstrup, M. Flato, M. Mewe, S. Schaaf, B. Aksehirlioglu, A. P. Schwoerer, J. Uebeler and T. Eschenhagen, Circ. Res., 2010, 107, 35–44 CrossRef CAS PubMed.
  52. A. Agarwal, J. A. Goss, A. Cho, M. L. McCain and K. K. Parker, Lab Chip, 2013, 13, 3599–3608 RSC.
  53. A. Takeuchi, S. Nakafutami, H. Tani, M. Mori, Y. Takayama, H. Moriguchi, K. Kotani, K. Miwa, J. K. Lee, M. Noshiro and Y. Jimbo, Lab Chip, 2011, 11, 2268–2275 RSC.
  54. S. S. Bale, L. Vernetti, N. Senutovitch, R. Jindal, M. Hegde, A. Gough, W. J. McCarty, A. Bakan, A. Bhushan, T. Y. Shun, I. Golberg, R. DeBiasio, O. B. Usta, D. L. Taylor and M. L. Yarmush, Exp. Biol. Med., 2014, 239, 1180–1191 CrossRef PubMed.
  55. N. T. Elliott and F. Yuan, J. Pharm. Sci., 2011, 100, 59–74 CrossRef CAS PubMed.
  56. P. M. van Midwoud, E. Verpoorte and G. M. Groothuis, Integr. Biol., 2011, 3, 509–521 RSC.
  57. C. T. Jones, M. T. Catanese, L. M. Law, S. R. Khetani, A. J. Syder, A. Ploss, T. S. Oh, J. W. Schoggins, M. R. MacDonald, S. N. Bhatia and C. M. Rice, Nat. Biotechnol., 2010, 28, 167–171 CrossRef CAS PubMed.
  58. P. J. Lee, P. J. Hung and L. P. Lee, Biotechnol. Bioeng., 2007, 97, 1340–1346 CrossRef CAS PubMed.
  59. J. H. Tsui, W. Lee, S. H. Pun, J. Kim and D. H. Kim, Adv. Drug Delivery Rev., 2013, 65, 1575–1588 CrossRef CAS PubMed.
  60. H. Kimura, T. Yamamoto, H. Sakai, Y. Sakai and T. Fujii, Lab Chip, 2008, 8, 741–746 RSC.
  61. G. J. Mahler, M. B. Esch, R. P. Glahn and M. L. Shuler, Biotechnol. Bioeng., 2009, 104, 193–205 CrossRef CAS PubMed.
  62. K. J. Jang, A. P. Mehr, G. A. Hamilton, L. A. McPartlin, S. Chung, K. Y. Suh and D. E. Ingber, Integr. Biol., 2013, 5, 1119–1129 RSC.
  63. E. M. Frohlich, J. L. Alonso, J. T. Borenstein, X. Zhang, M. A. Arnaout and J. L. Charest, Lab Chip, 2013, 13, 2311–2319 RSC.
  64. E. W. Young, Integr. Biol., 2013, 5, 1096–1109 RSC.
  65. E. Jedrych, Z. Pawlicka, M. Chudy, A. Dybko and Z. Brzozka, Anal. Chim. Acta, 2011, 683, 149–155 CrossRef CAS PubMed.


These authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2016