Micro Cell Culture Analogs and What They Can Contribute to the Drug Screening Process

Home / New Perspectives / Emerging Technologies / Micro Cell Culture Analogs and What They Can Contribute to the Drug Screening Process

Emerging Technologies

Micro Cell Culture Analogs and What They Can Contribute to the Drug Screening Process

By Michael L. Shuler & Mandy B. Esch, Cornell University

Published: February 24, 2009

About the Author(s)
Michael L. Shuler is the James M. and Marsha McCormick Chair of Biomedical Engineering at Cornell University, as well as the Samuel B. Eckert Professor of Chemical Engineering. His research focuses on applying chemical reaction engineering principles to biological systems. He has been a member of the Cornell faculty since 1974.

Shuler received the Amgen Award in biochemical engineering and has been elected to the National Academy of Engineers and the American Academy of Arts and Sciences. He has served on scientific advisory boards for Hurel Corporation, Princeton University, University of Texas, and at Carnegie Mellon for the Carnegie Institute of Technology and the Department of Chemical Engineering.

Among his other honors are an Excellence in Teaching Award (Cornell Society of Engineers), the 1986 Marvin J. Johnson Award for the MBT division of ACS, the 1989 AICHE Food, Pharmaceutical & Bioengineering Division Award, the 1991 Professional Progress Award, the 2003 Warren K. Lewis Award, and the Bailey Award in 2005 from the Society for Biological Engineering.

Shuler received a bachelor’s degree in chemical engineering from the University of Notre Dame and a Ph.D. in chemical engineering from the University of Minnesota.

Michael L. Shuler, Ph.D.
115 Weill Hall
Cornell University
Ithaca, NY 14853
Email: mls50@cornell.edu

Dr. Esch is a Research Associate at the Department of Biomedical Engineering at Cornell University. She develops microfluidic in vitro models of the microvasculature and the gastrointestinal tract. She received an M.S. degree in Biology in 1998 and a PhD in Biotechnology in 2001, both from the Julius Maximilians University in Wuerzburg, Germany.

Mandy B. Esch, Ph.D.
250 Duffield Hall
Cornell NanoScale Science and Technology Facility
Ithaca, NY 14853-2700
Email: mbe2@cornell.edu

The development of pharmaceuticals is an expensive endeavor that produces only a handful of new drugs each year. In 2008, the Food and Drug Administration has approved 21 new drugs – only three more than in 2007, and little more than half as many as ten years ago.1 In the search for these successful drugs, each year pharmaceutical companies screen through thousands of an estimated 1060 imaginable small molecule compounds with in vitro assays and pre-clinical trials. Of the many compounds tested, some are successful and are tried on humans. But even though these compounds have cured diseases in animals, about 89% fail to produce the desired results in humans,2 confirming that human disease mechanisms and the general metabolism of humans are quite different from those of animals. It seems, the closer the initial in vitro and in vivo drug screening assays are to accurately modeling the human organism, the better is the quality of data produced, and the more resources can be focused on the best drug candidates during the elaborate and expensive pre-clinical and clinical trials. Likewise, better models of the human system would allow us to more realistically simulate the toxicity of newly developed chemicals and to determine safe exposure limits for work environments and consumer products.


Researchers spend much effort on developing screening assays that address the relevant cell types and the reactions of the metabolism path targeted. In vitro screening assays are typically conducted in 96-, 384-, or 1536-well plates in which immortalized cells grow. Using these, a compound library containing thousands of compounds can be screened during the course of a few days. The assay output shows either cell viability in response to a compound, or the influence of the compound on particular processes such as its catalytic effects on an enzymatic reaction, its binding to a surface receptor, or the activation or up-regulation of a gene. Promising compounds are tested with animal models, in which both the pharmacology and the biological efficacy can be assessed. The usefulness of data obtained with these in vivo experiments is debatable because the human body and human diseases are complex and are not always accurately modeled in rodents or other animals, even if transgenic animals with added or deleted genes pertaining to the disease in question are studied.
Micro Cell Culture Analogs

Experimenting with realistic micro cell culture analogs (µCCAs) of human organs provides an alternative to model a drug’s complex effects on the human organism in vitro. Cell culture analogs are physical representations of mathematical physiologically based pharmacokinetic (PBPK) models.3 PBPK models are typically used to theoretically estimate a drug’s uptake, distribution, metabolism and excretion in the human body.4 These models treat each organ as one “compartment” with its characteristic parameters such as compound residence time within the vascular network, the interstitial spaces, and within the cells of the organ. The models also take flow limitations, drug transport mechanisms, and the involved metabolic steps into account.

Following the anatomically guided compartmental flow charts of PBPKs, cell culture analogs consist of several actual physical chambers and connecting channels (Figure 1). Nutrients and drugs can be supplied to the organ compartments at their characteristic perfusion rate. In well-engineered devices, a drug’s rate of uptake, residence time, and release rate per volume unit are similar to those measured in the human body. Surrogate organs, though, are usually scaled down to measure only a few cubic millimeters.

Modeled after the human body, a cell culture chip can, for example, contain a liver cell compartment sized 1.7 mm3, and a kidney compartment only one seventh of that size.5 Since in flow-limited systems such as a microfluidic chip the amount of reaction is controlled by the residence time of a compound within an organ compartment, the scaling to small size is acceptable as long as the fluid residence time and transport are accurately represented.6 This technology has been referred to as “Body-on-a-Chip”.7 It can be used for drug screening as well as to study the toxic effects of chemicals used in the work-place and other environments.

Figure 1: Flow chart for a simplified PBPK model that contains five organ compartments and the corresponding µCCA chip.5 The model was used with digestive tract and liver cells to simulate the first pass metabolism of acetaminophen.


The most utopian version of a µCCA might be a silicon chip with many chambers built up on top of each other in which many tissue-engineered, three-dimensional organ compartments are interconnected through a vascular-like network and that models most of the human organism. While conventional multi well plate models only measure the response of a single cell type, µCCAs capture the reactions of the organ system as a whole. Metabolites resulting from a drug’s influence on one organ can reach other organs and exert their effects, positive or negative. Even reactions so far unknown to us will take place within the system and contribute to the authenticity of the test results.

The use of micro- and nanofabrication in cell culture analog device development contributes significantly to the realization of their many advantages. Interconnecting channels and the volumes of organ compartments can be designed so that the distribution of fluid flow to the compartments is the same as the distribution of flow within the body. Other flow parameters, such as fluid residence times and shear stresses, can be adjusted to approximate those seen in vivo.

Nanofabrication also lets us modify the surfaces of organ chambers so that they are favorable towards a cell’s needs. Endothelial cells that line the inner walls of blood vessels, for example, grow on rounded surfaces and arrange their cytoskeleton accordingly. Such surface clues and cytoskeletal arrangements may affect intracellular transport mechanisms and the communication between adjacent cells. The further development of semiconductor materials and fabrication techniques for cell culture devices will likely soon lead to the availability of three dimensional scaffolds that will better enable the construction of three-dimensional surrogate tissues.

Recent Research

Authentically behaving liver surrogates are of special interest to µCCA device development, because the liver transforms toxins and metabolizes many drugs. Like all organs, the liver consists of several different cell types that are together responsible for its function. In vitro tests with two types of liver cells that were co-cultured alongside each other have shown “crosstalk”, which enabled them to survive in a microchip environment without being artificially stimulated with growth factors.8 Such a two-cell system behaves more realistically than each cell type alone. Sung et al. have shown the liver’s metabolizing function using a multi-organ µCCA device that contained a liver compartment and a compartment for colon cancer. Tegafur, a prodrug that is metabolized in the liver and acts as a chemotherapeutic agent for colon cancer, exerted measurably higher toxic effects on cancer cells when liver cells were present in the device.9 This effect of Tegafur on colon cancer has previously only been seen in animal experiments or studies involving humans, but not in vitro.

Another microscale organ surrogate that uses co-cultures of several cells is the digestive system module developed by Mahler et al.10 The lining of the intestinal tract, which is responsible for nutrient and drug absorption, can be modeled with absorptive epithelial cells and mucous-producing goblet cells.11 This combination of cells has been used for the in vitro digestive system, with which the digestion of acetaminophen was studied.5 Metabolites resulting from acetaminophen digestion are toxic for liver and lung cells, an effect that could be seen with the developed module.

Tissues such as those found in the digestive system present barriers, which drugs have to cross to reach other organs. This barrier function is especially important to modeling a drug’s uptake in vitro since it changes the concentration at which a drug becomes effective or toxic. Drugs can also enter the body via inhalation. In this case the lung epithelium presents a barrier. Huh et al. have developed a “lung” device in which lung epithelial cells are cultured at a liquid air interface.12 The researchers were able to acoustically detect injuries inflicted by liquid plugs in the “airway” of the lung. It is conceivable that, in conjunction with other organ compartments, this lung surrogate could be used to test drugs against airway diseases.


We envision that µCCAs will reduce the number of animals needed to validate a drug pre-clinically. PBPK models and in vitro data determine the starting doses for in vivo experiments. Data obtained from µCCAs may provide more realistic concentration and absorption values that can be used in conjunction with PBPK models to better predict the human response and to determine better initial dosing scenarios. Because of the correct flow rates and residence times of a drug in organ surrogates in the µCCA, the resulting dose estimations should be more accurate than those obtained with static multi well plate assays. Potentially, fewer animal tests need to be carried out to establish the safety of a drug or a chemical, because a smaller range of doses will need to be tested.

Current research efforts focus on validating results obtained with µCCAs. One way to do this is to compare data obtained with them with previously published data from preclinical trials. If the µCCA results are validated, the models will become increasingly important in drug development and toxicity testing.

CCAs might also be used to compose multi-drug prescriptions for individual cancer patients. A patient’s tissue samples can be cultured on several devices and their response to many different drug combinations can be evaluated. Tatosian and Shuler have recently published experimental results that demonstrate the feasibility of this approach.13 Since microfluidic chips are inexpensive to make, and only very small tissue samples are required, this may be practically feasible.


Combined with PBPK models, µCCAs offer a real possibility to realistically model the uptake, distribution, metabolism and excretion of pharmaceuticals or toxins in vitro. They eliminate the necessity to know a priori all metabolic reactions involved and have the potential to reveal metabolic reactions and pathways that are so far unknown to us. The engineering of authentic tissue constructs, however, is still challenging. Co-culture and three-dimensional organ models need to be further developed and combined with each other to build multi-organ systems with the long-term goal of developing whole organism models. Nanofabrication can be used to build µCCA chips with integrated sensor elements. Integrated optical waveguides and electrodes can be used to measure optical, electrical, and electrochemical signals that report cellular responses such as free radical formation or mitochondrial activity.

©2009 Michael Shuler & Mandy Esche

  1. Hughes, B. 2007 FDA drug approvals: a year of flux. Nature Reviews Drug Discovery.2008, 7, 107-109.
  2. Kola, I. and Landis, J. Can the pharmaceutical industry reduce attrition rates? Nature Reviews Drug Discovery.2004, 3, 711-715.
  3. Shuler, M.L. and Xu, H. Novel Cell Culture Systems: Nano and Microtechnology for Toxicology. In “Computational Toxicology: Risk Assessment for Pharmaceutical and Environmental Chemicals”. S. Ekins, Ed., J. Wiley & Sons, Inc, Hoboken, N.J., 2007, Chapter 25, 695-723.
  4. Poulin, P. and Theil, F.-P. Prediction of pharmacokinetics prior to in vivo studies. 1. Mechanism based prediction of volume of distribution. J. Pharm. Sci.2002, 91, 129-156.
  5. Mahler, G.J., Esch, M.B. and Shuler, M.L. Characterization of a gastrointestinal tract microscale cell culture analog used to predict drug toxicity, submitted to Biotechnology & Bioengineering.
  6. Freedman, D.H. Your body on a chip. Newsweek.2005, October 10th, 59.
  7. Fogler, H.S. Elements of Chemical Reaction, Fourth Edition, 2006, Person Education Inc., Upper Saddle River. N.J.
  8. Hwa, A.B., Fry, R.C., Sivaraman, A., So, P.T., Samson, L.D., Stolz, D.B. and Griffith, L.G. Rat liver sinusoidal endothelial cells survive without exogenous VEGF in 3D perfused co-cultures with hepatocytes. FASEB J.2007, 21, 2564-2579.
  9. Jong Hwan Sung, Taek-il Oh, Donghyn Kim, Michael Shuler “A microfluidic cell culture analog system (CCA) for testing drugs for colon cancer”, 232nd American Chemical Society National Meeting, Fall 2006, San Francisco, CA, USA, Poster session, BIOT 393.
  10. Mahler, G.J., Chang, J.Y., Glahn, R.P. and Shuler, M.L. Development of a Gastrointestinal Tract Microscale Cell Culture Analog to Predict Drug Transport. MCB.2008, 5(2), 119-13.
  11. Mahler, G.J., Shuler M.L., and Glahn, R.P. Characterization of Caco-2 and HT29-MTX cocultures in an in vitro digestion/cell culture model used to predict iron bioavailability. J Nutr Biochem.2008, in press.
  12. Huh, D., Fujioka, H., Tung, Y.-C., Futai, N., Paine, R., Grotberg, J.B. and Takayama, S. Acoustically detectable cellular-level lung injury induced by fluid mechanical stresses in microfluidic airway systems. Proc Natl Acad Sci USA.2007, 104, 18886-18891.
  13. Tatosian, D. and Shuler, M.L. A novel system for evaluation of drug mixtures for potential efficacy in treating multidrug resistant cancers. Biotechnol Bioeng. 2008, in press.