15. Medical Applications of Bioprocess Engineering

With increasing knowledge of cellular and molecular biology, the boundary between traditional bioprocess engineering and biomedical engineering has become increasingly fuzzy. In this chapter, we consider examples in which bioprocess principles are critical to solution of medical problems. Two important areas are tissue engineering and gene therapy using viral vectors. These are just two examples; other medical applications exist, and their number will undoubtedly increase. Indeed, we have already mentioned the use of transgenic plants as a source for edible vaccines and the techniques of microfabrication applied to make miniature process facilities for rapid DNA analysis to allow genome analysis in a physician’s office.

15.1. Tissue Engineering

An important contribution to medicine of bioprocess engineering is in regenerative medicine. Here the techniques we have discussed in the first 14 chapters can be used to generate replacement tissues for the human body. Building such replacement tissues is a challenge, but for some simple tissues, it is a commercial and clinical success.

15.1.1. What Is Tissue Engineering?

Tissue engineering has a primary focus on developing in vitro bioartificial tissues, typically based on cells derived from donor tissue. These tissues can be used as transplants to improve biological function in the recipient. Commercial examples include living skin tissue and chondrocytes implanted in a damaged knee for production of hyaline-like cartilage. An extracorporeal (outside of the body) artificial liver employing pig liver cells has been clinically tested.

Artificial tissues/organs for transplantation that are under active development include liver, pancreas, kidney, fat (for reconstructive surgery), blood vessel, bone marrow, bone, and neurotransmitter-secreting cell constructs. An alternative form of tissue engineering is in vivo alteration of cell growth and function. An example would be the use of implanted polymeric tubes with a controlled surface chemistry to encourage and guide reconnection of damaged nerves. Another use of artificial tissue constructs is for toxicological and pharmacological testing of potential new drugs. In this case, the artificial tissue or combination of tissues acts as a surrogate, reducing the need to use animals for such testing.

The primary difference between tissue engineering and protein production from mammalian cells lies in the constraints on cell selection. For protein production we prefer continuous, transformed cell lines. A single cell type is desirable. For tissue engineering the goal is to replicate the response of a living tissue. Cells removed from a tissue and cultured as a homogenous cell type in two dimensions (e.g., on a solid surface) often lose their authentic in vivo response. Since cell transformation and cancer are closely related, transformed cells cannot be used for a product for transplantation. Reconstruction of artificial tissues requires a deep understanding of the interactions of one cell with another, control of cellular differentiation processes, and knowledge of how cells interact with surfaces to which they attach. Most often, a polymer scaffold is used to guide and organize tissue growth. Maintaining the correct ratio of cell types can be difficult, since some cell types, such as fibroblasts, can “outgrow” others. The appropriate cell types must organize themselves into the appropriate three-dimensional configuration. With such tissues, function requires appropriate structure. Ideally, the cells can be multiplied from donor tissue for at least 10 passages and then assume the fully differentiated phenotype when the correct stimulus is applied. A major challenge is the routine, reproducible culture of such cells, and bioprocess engineers are in an excellent position to contribute to this technology.

Of particular importance to many tissue-engineered constructs is the formation of extracellular matrix, the interaction of cells with one another, and the interaction with an artificial surface. Anchorage-dependent cells must attach and spread on a substrate surface to proliferate and function. Cell adhesion is mediated by extracellular matrix proteins such as fibronectin and collagen. Synthetic substrates can be modified by adding synthetic peptide sequences (three to six amino acids) to a surface at the end of a synthetic polymer. The difference in strength of cell–substrate and cell–cell adhesion can greatly alter the three-dimensional organization of the cells.

The use of simulated microgravity reactors coupled with polymer scaffolds has been useful in some cases in the development of certain tissue constructs. The use of microfabrication techniques, where the investigator has control over placement of individual cells, is an intriguing technology for more authentic tissue constructs. An increasingly large array of tools are being developed for controlling the formation of tissue constructs with improved performance.

As noted in Chapter 9, “Operating Considerations for Bioreactors for Suspension and Immobilized Cultures,” 3-D printing of tissues is an emerging technology of increasing importance, particularly for mammalian cells. It is possible to print simple tissues (typically those that do not have a high level of vascularization) that are functional in the human body. The range of materials that are biocompatible with cells and still printable is a constraint, but rapid progress is being made in developing technologies that facilitate the use of a wider range of materials. The ability to print tissue constructs in any shape, on an individual basis, can allow the personalized replacement of missing or damaged tissues. For example, a replacement ear can be printed to match rather precisely the one that was lost in an accident. Skin is another organ that has been commercially manufactured by scaling up traditional tissue engineering approaches. The technologies for producing tissue-engineered constructs are expected to become a major component of regenerative medicine. However, the development of effective tissue constructs that are biologically complex, such as the liver, is a very difficult problem, and commercial production of transplantable organs is still futuristic. However, commercial production of simple tissues, particularly ones with low vascularization has been accomplished.

The two examples of commercial production of tissue-engineered products discussed in this chapter are for skin and cartilage.

15.1.2. Tissue-Engineered Skin Replacements

Skin replacements are needed for patients with burns and pressure ulcers (usually due to diabetes) and for reconstructive surgery. Cadaver skin is undesirable because of potential disease transmission, possible immunological rejection, inflammation of the wound bed, limited availability, and high cost due to labor-intensive procurement. Animal products would be problematic due to immunological reactions. An “artificial” human skin replacement that can be taken out of the freezer and used by physicians is highly desirable. TransCyte, a human fibroblast-derived temporary skin substitute, is such a product.

Artificial human skin is made from human neonatal foreskin fibroblasts, which are obtained from routine circumcision discards. These fibroblast cells are easily available and not subject to rejection by the body; they are also capable of replication, and numbers can be expanded in a straightforward manner. Donor tissue is enzymatically digested and seeded onto a three-dimensional bioresorbable polymer scaffold. The cell-seeded scaffold is placed in a bioreactor that provides a physiologically similar environment to promote cell growth and secretion of proteins and extracellular matrix material. The cells and extracellular matrix material result in a three-dimensional tissue construct that is functionally similar to natural skin and promotes wound healing.

The manufacturing process is centered on a bioreactor system that mimics in vivo conditions through control of temperature, pH, oxygen level, nutrient supply, waste removal, and fluid hydrodynamics. It must produce thousands of individual products of uniform and reproducible quality. The manufacturing system requires a bioreactor for product growth that also serves as final packaging. The system is sealed after assembly and remains sealed to maintain sterility. Figure 15.1 is a schematic of the process. The bioreactor is made of plastic with inlet and outlet ports. It is flat and just large enough to hold a polymer mesh about 13 by 19 cm. Twelve individual bioreactors are manifolded together. Groups of 12 are then operated in parallel to provide the desired lot size. Once the system is sterilized, it is opened only for cell seeding and nutrient replenishment. About 3 weeks are required for growth. At the end of this growth period, cryopreservative is pumped into the bioreactor module. After sealing inlet and outlet tubes, the bioreactor is packaged, frozen, and then stored or shipped. The product is thawed and used by the physician.


Figure 15.1. Conceptual process diagram for production of a human skin substitute.

This manufacturing process is similar to many discussed in this book but differs in that the product is discrete and not in solution. The process must meet all of the FDA and GMP (good manufacturing practice) requirements discussed earlier.

15.1.3. Chondrocyte Culture for Cartilage Replacement

Articulate or hyaline cartilage is a thin layer of tissue found at the ends of bones. Damage to articulate cartilage in the knee is a common health problem. Cartilage consists of three primary components: collagen fibrils, proteoglycans, and cells known as chondrocytes. These cells are present in low numbers (1% by volume) but are responsible for synthesis and release of these other compounds, forming cartilage. When chondrocytes are cultured as two-dimensional (monolayer) cultures, the cells are no longer differentiated and do not make normal hyaline matrix proteins. However, if cultured in three dimensions, such as a suspension in agarose, or in vivo, the cells redifferentiate and begin to manufacture hyaline-like matrix.

Because of immunological responses it is advantageous for patients to supply their own cells for expansion in number before the cells are reinjected into the knee. A manufacturing process for individual cell growth is needed. The supply and reuse of cells from an individual is termed autologous implantation.

The basic procedure entails biopsy of a patient to obtain a small number of chondrocytes, monolayer culture of primary chondrocytes and expansion of cell numbers, release of cells from monolayer and into suspension, assembly of cells in three dimensions, and injection into the patient’s own knee. This product is known as Carticel. The implanted cells produce hyaline-like cartilage and fill in defects in the patient’s knee. Patients can often reenter low-impact activities such as swimming or cycling in 6 months and vigorous physical activity in 12 months.

The primary manufacturing challenge here is the need for separate culture for each individual. To accomplish this for mass production is a challenge to any manufacturing process. There is expected to be a demand for other therapies (e.g., cancer treatment), where cells from an individual will need to be efficiently cultured and returned to the individual.

15.2. Gene Therapy Using Viral Vectors

Gene therapy is the transfer of one or more genes into cells for a therapeutic effect. It can be done ex vivo (outside the body) in tissues that are transplanted back into the patient or in vivo, where the genes, usually in a delivery vehicle such as a virus, are injected into the patient.

Gene therapy is intellectually connected to metabolic engineering. However, the complexity of humans presents an even greater challenge than the metabolic engineering of a single cell. Gene therapy is a quantitative problem that makes good use of the quantitative skills of engineers. Basically, the right gene needs to be delivered to only the right tissue target in the right amount, with the gene products being expressed at the right level at the right time for the right length of time. For gene therapy to be effective, many things have to go right. Clinical success with gene therapy has been minimal, as approaches have been qualitative and trial-and-error in nature. A rational analysis would be a useful tool.

Many methods can be used for gene delivery. These involve viral vectors, use of naked DNA, and liposome or particle-mediated gene delivery. In this chapter, we focus only on viral systems. Bioprocess technology is necessary for production of the viral vectors, and analyses arising from bioprocess studies are applicable to gene therapy.

The four primary virus vectors for gene therapy are retroviruses, lentiviruses, adenoviruses, and adeno-associated viruses (AAV). Retroviruses and lentiviruses are enveloped viruses, because they are encapsulated in a lipid bilayer membrane. The adenovirus and AAVs are nonenveloped viruses.

Typical concerns with the use of viral vectors include safety, low toxicity, genomic stability, and cell specificity. Safety to both patients and production workers is essential. Immune responses to viruses can be a significant issue and vary from patient to patient. One of the early trials, for example, with an adenoviral vector led to the death of patient. Some viral clinical vectors (e.g., lentiviruses) insert in random locations in the genome and can inactivate an essential gene in the host cell. AAV-based vectors are safer in that they insert at specific locations. AAV is not known to cause disease and elicits a mild immune response. For all of these viral vectors specificity can be broad, but targeting key receptors can increase specific delivery of genes. Some features of these viral vectors are summarized in Table 15.1.


TABLE 15.1. Comparison of Major Viral Gene Delivery Systems

15.2.1. Models of Viral Infection

Dee, Hammer, and Shuler proposed a model for the viral trafficking of Semliki Forest virus (SFV), an enveloped RNA virus that has been considered as a vector for large-scale production of heterologous proteins. However, this analysis, which was motivated by a bioprocess application, is applicable to retrovirus vectors for gene therapy (although SFV is not appropriate for gene therapy).

K. U. Dee, D. A. Hammer, and M. L. Shuler, Biotechnology and Bioengineering 46: 485–496, 1995.

As indicated in Figure 15.2, enveloped RNA viruses can enter cells through receptor-mediated endocytosis. The virus binds to specific receptor molecules on the cell’s surface. It is assumed (for this analysis) that attachment is irreversible and that the number of receptor molecules is much greater than the number of virus particles present. Consider the following circumstance:


Figure 15.2. The trafficking of an enveloped RNA virus is depicted. A mathematical model of these processes is given in the text.


Here Vex is the number of extracellular viruses per cell, C is the cell concentration, and ka is the attachment-rate constant. The value of ka can be estimated as


where kf is the intrinsic forward rate constant for the binding of a single viral attachment protein to a receptor, α is the number of viral attachment proteins per virus, and R is the number of available receptors per cell. If the ratio of R to Vex is large, R is approximately the total number of receptors.

Once the virus is attached, it randomly associates with “coated pits,” which internalize the virus through the process of endocytosis. Endocytosis is a process of invagination of the plasma membrane to form a vesicle in which the receptors and anything attached to the receptors are captured within the vesicle. Such vesicles are known as endosomes. The rate of endocytosis is assumed to be


where Vi is the number per cell of internalized virus, Vs is the surface concentration of virus (number per cell), and ke is the endocytosis rate constant.

The amount of surface virus can be determined from combining equations 15.1 and 15.3 to give


Endosomes are intermediates in recycle of plasma membrane components and transport of internalized materials to lysosomes for degradation. For the virus to replicate successfully, it must escape from the endosome before it is delivered to a lysosome. Endosomes undergo a biphasic pH change. In early endosomes, pH drops from about 7.5 to 6.0 within 5 to 10 min. In the second phase, pH drops slowly to about 5.2 in another 30 to 60 min. The pH drop is due to membrane proteins that pump hydrogen ions into the endosome.

The virus escapes through fusion of its membrane envelope with the endosome membrane. The fusion process is pH dependent. For SFV, the threshold pH for fusion is 6.2. This threshold is reached in early endosomes. The fusion process is quite rapid. However, not all virus undergoes fusion. The virus balance in the endosome is


where Vene is the number of virus in the early endosomes, kfus is the observed fusion-rate constant, η is the fraction of virus that can successfully fuse with endosomal membrane, and ktran is the rate constant for movement of inactive virus to the lysosome and degradation. The main cause of inactive virus is the fusion of virus with membrane fragments within the endosome instead of the endosomal membrane.

Virus that successfully fuses with the endosomal membrane is released into the cytosol. The protein coat is removed rapidly, and for this analysis, we assume that effectively all cytoplasmic virus is uncoated, so that the RNA is released and can be replicated.

The rate equation for uncoating is


where Vcyt is the cytoplasmic virus.

The amount of RNA synthesized per cell follows a saturation type response:


where km and Ksv are parameters dependent on the host cell. Equation 15.7 applies only to the early parts of the infection cycle. Virus that enters after RNA replication has been completed for early-entry virus probably no longer contributes to RNA synthesis, as host cell function begins to be lost.

Equations 15.1 to 15.6 allow for an analytical solution. If we assume that only free virus is present initially (t = 0), then the initial conditions are Vex = Vexo; Vs = 0; Vi = 0; Vene = 0; Vcyt = 0. The solutions in this case are as follows:


For wild-type SFV infecting baby hamster kidney (BHK) cells, the rate parameters have been estimated as ka = 5.2 × 10–9 ml/cell-min; ke = 0.039 min–1; and kfusα = ktran = 0.14 min–1 (based on 50% of internalized virus being uncoated). Such a model with these parameters fits time-course measurements from actual experiments and can also be used to make estimates of response if mutant virus forms are used.

These same equations can easily be applied to an ex vivo system for gene therapy. For in vivo use they apply in principle, but the presence of various natural compounds in the bloodstream can alter ka (and potentially other parameters), and multiple cell types are present which may bind, but not endocytose virus. A pharmacokinetic model (to predict time-dependent virus distribution at different positions in the body) would need to be coupled with virus/cell interaction models (of the type developed here) for each major cell type. It is easy to see why ex vivo therapy would be much easier to design than an in vivo protocol. For either the ex vivo or in vivo therapies to be successful, there must be a method to efficiently produce virus.

15.2.2. Mass Production of Retrovirus

The efficient production of a highly concentrated solution of retroviruses (or the closely related lentivirus) that have been genetically modified to deliver a gene is difficult. Genetically engineered retroviruses, like all recombinant viruses, are produced by a two-part system consisting of a packaging cell line and the recombinant vector. This two-part system is necessary because the recombinant virus is unable to replicate itself. The recombinant virus is derived from a wild-type virus in which essential viral genes have been deleted and replaced with the therapeutic gene of interest. The enzyme reverse transcriptase converts RNA-encoded genes into DNA, which is transported into the nucleus and integrated into the cell’s chromosomal DNA.

The packaging cell line is genetically engineered to produce the essential, structural viral genes that have been deleted from the viral genome. Because the viral genes are encoded in the packaging cell’s chromosome, the resulting viral particles are incapable of causing disease but act as carriers for the desired, therapeutic genes. The resulting retrovirus vector can be used only with dividing cells, since cells must undergo mitosis before gene integration can occur. This feature limits in vivo use to cases such as cancer suppression but is not a limitation on ex vivo systems.

Another limitation on the retrovirus system is a bioprocess limitation; the production of high-titer virus and subsequent purification and concentration without loss of infectivity have been difficult. A major limitation to the effectiveness of many gene-therapy approaches, including the use of retroviruses, is that the average number of genes delivered to target cells is too low to achieve a beneficial therapeutic effect. This low efficiency is due to the bioprocess limitation on production of a concentrated, highly infectious retrovirus preparation.

The problems in producing a highly infectious preparation derive from two factors: rapid decay of virus and inhibition of viral infectivity by proteoglycans released by the packaging cell line. The proteoglycans have very high molecular weight. If the virus is concentrated by ultrafiltration, the proteoglycan is also concentrated by nearly the same factor. While the high virus concentration would be expected to increase the number of infected cells, the presence of concentrated proteoglycan leads to increased inhibition of infection, so that the number of cells infected does not increase significantly and in some cases even decreases with the concentrated viral preparation.

J. M. LeDoux, H. E. Davis, J. R. Morgan, and M. L. Yarmush, Biotechnology and Bioengineering 63: 654, 1999.

Another approach to increase viral titer is to produce virus at a reduced temperature (e.g., 28°C versus 37°C). While both the rate of virus production and viral decay decrease with temperature, the rate of decay is more sensitive to temperature. Thus the amount of potentially active virus is severalfold higher when produced at 28°C (e.g., 2 × 105 cfu/ml vs. 6 × 104 cfu/ml; cfu is colony-forming units and measures the number of active viruses in a dilute solution). However, the transduction efficiency did not increase, presumably due to changes in proteoglycan concentration.

These results should inspire us to consider bioreactor options beyond batch growth followed by ultrafiltration. Clearly, we would want to consider a bioreactor configuration in which the virus is removed as soon as it is produced and then purified and stored at low temperature to reduce decay. Further, a method of concentration of virus needs to be developed that does not also concentrate proteoglycan. Enzymatic digestion of proteoglycans has been suggested, but due to multiplicity of proteoglycan structures such a strategy is difficult to implement. Selective adsorption and desorption of virus is attractive, but the sensitivity of the virus to these processes is a concern. A packaging cell altered in its capability to produce proteoglycan is another possible strategy.

The primary point is that bioprocess technology is intimately connected to production of agents that can be effective in gene therapy. The solution to the biomedical problem requires a solution to the bioprocess problem.

15.3. Bioreactors

Mass production of cells for transplantation and the use of bioreactors as artificial hybrid organs are subjects of intense development in medicine. These are issues in which bioprocess engineers may make significant contributions.

15.3.1. Stem Cells and Hematopoiesis

Some animal cells are capable of extensive replication and self-renewal. Others are highly differentiated and perform specific functions; typically, differentiated cells cannot replicate (at least, replication is very limited). A stem cell is an undifferentiated cell capable of continuous self-renewal that can also produce large numbers of differentiated progeny, depending on extracellular factors.

The ultimate form of a stem cell is one that is pluripotent and can be differentiated into any type of cell in the body. Originally, pluripotent stem cells could be derived only from embryos. For cells derived from humans, this presented ethical as well as technical challenges. In 2006, Shinaya Yamanaka (2012 Nobel Prize winner) demonstrated it was possible to convert adult, specialized cells into induced pluripotent stem (iPS) cells by the introduction of only four genes. In principle, these cells then can be converted into any specialized cells with the correct extracellular cues. This technology offers great promise not only for use in regenerative medicine but also for uses such as drug testing and development. Progeny from these stem cells are genetically identical to the parental cells. There are subtle differences (epigenetic), and maturation into an adult phenotype can be challenging.

The scale-up of stem cell technology for the production of large quantities of cells is a bioprocess challenge. All of the GMP standards and regulatory requirements need to be met at a large scale. While single-use technology for cell cultivation (see Chapter 10, “Selection, Scale-Up, Operation, and Control of Bioreactors”) would appear to be a good option, fragile adherent cells (as stem cells are) are highly sensitive to their microenvironment, and cell harvest is a challenge. Multistack arrays based on two-dimensional growth on polystyrene plates have been proposed and may be practical, but they are still relatively inefficient. For this technology to move forward, more effective methods for mass production of cells must be sought.

Not all stem cells are pluripotent but can be precursors to a set of cell types, such as blood cells. The development of production systems for these cell types have been explored for many years, although none have been fully commercialized. Much of the early work has been focused on production of red blood cells from hematopoietic stem cells. This example requires the development of robust large-scale production systems for the process to become practical.

There are eight major types of fully mature blood cells in the human circulatory system. A hematopoietic stem cell gives rise to two types of progenitor cells. The progenitor cells are capable of self-replication but have a more restricted range of cells into which they may differentiate. In the hematopoietic system, one type of progenitor cell can give rise to the myeloid cells (e.g., red blood cells or erythrocytes, platelets, macrophages) and the other to lymphoid cells (e.g., T cells and B cells). The progression of differentiation depends on a large number of hematopoietic growth factors. Hematopoiesis, the process of generating these blood cells, takes place in the bone marrow. Ex vivo hematopoiesis is an alternative or supplement to bone marrow transplants. Hematopoietic stem and progenitor cells can be recovered from human blood, particularly umbilical cord blood.

Schemes for large-scale systems have been considered for 20 years. Coculture with stroma cells from the bone marrow is necessary to generate necessary growth factors, so bioreactors must accommodate adherent cell growth. Three basic reactor types are under development. Fluidized bed reactors with macroporous supports mimic structures much like that in bone. However, there are challenges in controlling surface chemistry and maintaining the appropriate mix of cell types. Cell sampling and control of reactor conditions can be problematic.

Flatbed reactors (e.g., modified T-flasks with continuous flow possible) can carry stroma and facilitate analysis and design, since the geometry is well defined. Direct microscopic observation of cells is possible. Automated flatbed systems have been used to generate cells for human clinical trials.

Membrane-based units, such as hollow-fiber reactors, are potential solutions. This is an efficient design that is fairly easy to characterize. However, cell observation and harvesting are considered problematic. Another alternative is the possible use of spheroid cultures (a natural aggregation of a mixed population of cells), which can be done in suspension-type bioreactors.

Bioreactors for production of other tissue types from other stem cells are a real possibility. There is far less experience with commercial-scale bioreactors for other stem cells/tissues, but the principles presented in this book should be applicable to these challenges.

15.3.2. Extracorporeal Artificial Liver

Liver failure is a major medical problem. The liver performs many metabolic functions (carbohydrate, fat, and vitamin metabolism; production of plasma proteins; conjugation of bile acids; and detoxification), of which detoxification is the most critical. In some cases, a failing liver may recover if the metabolic and detoxification demands on it can be reduced; an artificial liver may provide the respite necessary for self-repair of a liver. In other cases, an artificial liver may serve as a bridge to a transplant.

Due to the spatial and metabolic complexity of the liver, an implantable artificial liver is a distant objective. However, an extracorporeal device to serve as a temporary assist device is realistic (such a system is in clinical trials). A promising design is a hollow-fiber system using porcine (pig) hepatocytes. Such cells are relatively easy to obtain in large quantities and maintain a satisfactory level of differentiated cellular activity in regard to detoxification. A disadvantage is due to limited lifespan (and proliferative ability). The membrane that separates these cells from the blood provides protection against adverse immune reactions. The issues in the design of such hollow-fiber reactors are similar, whether the reactor is to be used in a bioprocess operation or in a biomedical application.

15.3.3. Body-on-a-Chip Systems

Preclinical drug development is a difficult process (over several billion dollars per approved drug and typically 12 to 15 years). This effort typically requires the use of animals to test the potential safety and efficiency of drug candidates before beginning human clinical trials. However, these tests on animals are not good predictors of human response due to differences between human and animal physiology and metabolism. Consequently, about 10% of the drug candidates that pass preclinical animal testing succeed in becoming approved drugs. Because human clinical trials are very expensive, this inefficiency increases significantly the cost of drugs.

Currently, technology is being developed to construct human surrogates for use in preclinical evaluation of drugs. These surrogates are often known as microphysiological, body-on-a-chip, or multiorgan-on-a-chip systems. These systems depend heavily on ideas from bioreactor systems and tissue engineering or immobilized cell technology. Because drug development requires many tests done on nearly identical systems, these human surrogates are constructed on a microscale using technologies originally developed for electronics.

These systems make use of a physiologically based pharmacokinetic (PBPK) model of the human body to guide the development of these microscale systems. The PBPK models mathematically the uptake, distribution, metabolism, and physiological response of various organs and tissues to the parental drug and its metabolites. However, these models are limited because our understanding of drug fate and metabolism is incomplete. By replacing the differential equations in the PBPK model with living tissue/organ mimics, we can observe responses from various “organs” that would not be anticipated otherwise, thus providing a more accurate prediction of response. However, because the mathematical model and the device are easily linked and parameters can be determined relatively precisely with the device, the extrapolation of data from the device to prediction of human response is possible.

No in vitro model will ever fully capture the complexity of humans, but an improvement in preclinical testing from 10% success to 30% success would have a tremendous impact in reducing costs and increasing the number of drugs available to treat disease.

While a detailed description of this technology is beyond the scope of this book, references to further information are provided at the end of this chapter. This example demonstrates how a bioprocess knowledge of bioreactors and mass transfer can be coupled with an understanding of human physiology to make useful predictions on the potential safely and effectiveness of drug candidates (and also exposure to chemicals and cosmetics).

15.4. Summary

The principles from bioprocess engineering are not restricted to use for biomanufacture of chemicals, pharmaceuticals, and other biological compounds. These same principles have many applications in medicine, as well. Indeed, as biomedical engineering has become more oriented to molecular and cellular systems and bioprocess engineering more toward animal cells, the boundary between these two activities has become very porous.

Important concepts in this chapter include ideas of how to produce tissue constructs on a commercial scale. Also, the efficient production of virus for use in gene therapy is an unresolved bioprocess problem. The basic understanding of the dynamics of viral infection of cells is important both in design of biomanufacturing processes and for gene therapy. Quantitative approaches to describing such processes can lead to rational design both for the biomanufacturing process and for therapeutic strategies. Bioreactor design is also an important component of design of artificial organs and reactors to generate functional tissue from stem cells and to the development of human surrogates for drug development.

The list of examples of the intersection of bioprocessing and medicine is a growing one that will present exciting career possibilities.

Suggestions for Further Reading

DEE, K. U., D. A. HAMMER, AND M. L. SHULER, A Model of the Binding, Entry, Uncoating, and RNA Synthesis of Semliki Forest Virus in Baby Hamster Kidney (BHK-21) cells, Biotechnol. Bioeng. 46: 485–496, 1995.

DING, Y-T., AND X.-L. SHI, Bioartificial Liver Devices: Perspectives on the State of the Art, Front. Med. 5: 15–19, 2011.

ESCH, M. B., T. L. KING, AND M. L. SHULER, The Role of Body-on-a-Chip Devices in Drug and Toxicity Studies, Annu. Rev. Biomed. Eng. 13: 55–72, 2011.

ESCH, M. B., A. SMITH, J. M. PROT, C. OLEAGA, J. HICKMAN, AND M. L. SHULER, How Multiorgan Microdevices Can Help Foster Drug Development, Adv. Drug Deliv. Rev. 69–70: 158–169, 2014.

LE DOUX, J. M., H. E. DAVIS, J. R. MORGAN, AND M. L. YARMUSH, Kinetics of Retrovirus Production and Decay, Biotechnology and Bioengineering 63: 654–662, 1999.

LU, L., H. M. ARBIT, J. L. HERRICK, S. G. SEGOVIS, A. MARAN, AND M. J. YASZEMSKI, Tissue Engineered Constructs: Perspective on Clinical Translation, Ann. Biomed. Eng. 43: 796–804, 2015.

REN, S., J. I. IRUDAYAM, D. CONTRERAS, D. SAREEN, D. TALAVERA-ADAMS, C. N. SVENDSEN, AND V. ARUMUGASWAMI, Bioartificial Liver Device Based on Induced Pluripotent Stem Cell-Derived Hepatocytes, J. Stem Cell Res. Ther. 5(2): 1000263, 2015.

SINGH, V. K., M. KALSAN, N. KUMAR, A. SAINI, AND R. CHANDRA, Induced Pluripotent Stem Cells: Application in Regenerative Medicine, Disease Modeling, and Drug Discovery, Front. Cell Dev. Biol. 3: Article 1–17, 2015.

vAN VEEN, T., AND J. A. HUNT, Tissue Engineering Red Blood Cells: A Therapeutic, J. Tissue Eng. Regen. Med. 9: 760–770, 2015.


15.1. Many experiments for virus trafficking are done with prebound virus. Virus attaches at low temperature to suppress endocytosis, and unbound virus is washed away. Cells are then warmed to 37°C to initiate endocytosis. Derive equations in this case for the time-dependent concentration of internalized virus and virus that enters the cytosol and uncoats.

15.2. A mutant Semliki Forest virus (SFV) has values of kfusη = 8.3 × 10–3 min–1 and ktran = 1.8 × 10–2 min–1. The wild-type virus has values of kfusη = 6.4 × 10–2 min–1 and ktran = 1.1 × 10–1 min–1. For both wild-type and mutant SFV the value of ke = 0.074 min–1. Prebound virus were used (see Problem 15.1) and Vs0 was the same for both cases. At 20 min, what is the ratio of uncoated virus for the wild-type virus to the mutant?

15.3. A value of kaC is measured as 0.021 min–1. Assume kfusη = ktran = 0.14 min–1 and ke = 0.04 min–1. The initial inoculum was 500 viruses/cell. At 2 hours after inoculation, unbound virus was removed. Calculate Vs, Vcyt, and Vene.

15.4. You wish to produce active retrovirus, and you are investigating the effect of temperature on the process. Active virus is subject to decay with a rate constant, kd = 2.2 day–1 at 37°C and 0.76 day–1 at 31°C. The production rate of virus from a packaging cell line is kp = 3.3 virus/cell-day at 37°C and kp = 2.9 virus/cell-day at 31°C. Assume that there are 1 × 106 cells, and the volume of liquid medium is 2 ml. The initial number of virus in solution is zero. How many viruses are there per ml 12 hours after initiation of virus production?

15.5. Mass-transfer limitations are often critical in design of devices such as the hollow-fiber bioartificial liver for use as an extracorporeal assist device. Consider removal (detoxification) of a slightly hydrophobic compound in a patient’s blood. Draw a diagram and describe potential mass-transfer limitations on the rate of detoxification by intracellular enzymes.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.