National Science Foundation     |     Directorate for Engineering  (ENG)
Division of Chemical, Bioengineering, Environmental, & Transport Systems  (CBET)
CBET Research Highlights 
Notable Accomplishments from CBET Awards
5345 - Part A - An Engineered Tumor Model for Improving Drug and Gene Delivery Studies
Fan Yuan  -  Duke University

Outcome or Accomplishment:  Researchers from Duke University have developed a model for investigation of drug and gene delivery in tumor tissues without using animals.  They have used it to demonstrate that cellular structures are highly resistant to diffusion of small molecules and that genetically engineered bacteria can spread effectively in tumors.

Fan Yuan Image 1
    Figure 1.  The design of a 3D tumor model, established by culturing tumor cells in a microfluidic system.
Fan Yuan Image 2
    Figure 2.  Side channels perfused with various solutions of small, fluorescent markers,shows that most tumor cells were alive after overnight culture.
Fan Yuan Image Image 3
    Figure 3.  Resistance was higher for fluorescent molecules that could be taken up by tumor cells which demonstrates that cellular uptake could prevent the molecules from penetrating into deeper regions.
Credit for All Images:  Nelita Elliott and Fan Yuan, Department of Biomedical Engineering, Duke University

Impact:  The model established can facilitate development of new strategies for improving drug/gene distribution in tumor mass.  The engineered bacteria may inspire design of more efficient vehicles for targeted delivery of drugs and genes to tumors.

Explanation/background:  One of the challenges in cancer treatment is that therapeutic agents (e.g., drugs and genes) cannot reach the majority of tumor cells, because various barriers exist in tumors that block transport and spreading of these agents. As a result, tumors often grow back after treatment.  In the past, most studies on how to overcome the barriers require the use of tumors grown in animals.  That requirement can make the research complicated and inefficient.  To solve this problem, biomedical engineers at Duke University developed a new tumor model by growing tumor cells in a microfluidic channel to form three-dimensional (3D) structures.  The study, funded by the Biomedical Engineering Program at NSF, showed that the 3D structures could mimic cellular barriers to drug and gene delivery observed in animals.  To demonstrate that the model could be used for gene delivery studies, the researchers investigated active transport of genetically engineered bacteria.  Results from the study suggest that future gene delivery vectors need to be designed in such a way that they are capable to move actively in target tissues.

CBET Research Highlight - Part B - Engineering Technical Information

5345 - An Engineered Tumor Model for Improving Drug and Gene Delivery Studies

Fan Yuan - Duke University

Background:  Uniform distributing drugs and genes in tumor tissues is a challenge in cancer treatment.  At present, the distribution is limited to a narrow region near blood vessel wall after systemic administration or around drug release device for local delivery in tumor tissues.  The limitation is caused by physiological barriers that block transport and spreading of drugs and genes.  The limited distribution has been considered to be one of the leading causes of tumor recurrence after treatments.  To improve drug and gene distribution, various strategies have been investigated using animal models that can be complicated and inefficient.  Therefore, scientists have developed three-dimensional (3D) tumor models based on cell culture techniques.  In previous studies, 3D models have been established through spheroidal or multicellular layer (MCL) arrangement of cells.  Although these models are useful for drug screening and drug delivery studies, they are available only to certain types of tumors.  Additionally, it is hard to quantitatively determine transport of small, non-fluorescent drugs in tumor spheroids.  For the MCL model, visualization of fluorescent drug distributions or cellular responses to a treatment often requires disruption of the MCL, making this type of model unsuitable for real-time microscopic visualization.  Another approach to tumor model development is to culture cells in microfluidic systems.  However, the cell density in microfluidic channels reported in previous studies is significantly lower than that in tumors observed in animals and patients.  With low cell density, microfluidic tumor models cannot provide cellular structures that mimic the physiological barriers to drug/gene delivery, thereby not being suitable for transport studies.  To this end, Dr. Fan Yuan and his co-workers at Duke University in Durham, North Carolina have developed a new 3D tumor model, in which a microfluidic channel is densely packed with tumor cells.  Using this model, Dr. Yuan's lab, in collaboration with Dr. Lingchong You's lab at Duke University, investigated transport and proliferation of a genetically engineered Escherichia coli (E. coli) in tumors.  Non-pathogenic bacteria have unique capabilities that have made them attractive candidates for gene delivery carriers in cancer treatment.  However, their proliferation and transport behaviors in tumor tissues are still poorly characterized.  Therefore, the investigation funded by NSF is important.  Data from the study may inspire design of more efficient vehicles for targeted delivery of genes in tumors.

Results:  The design of 3D tumor model, established by culturing tumor cells in a microfluidic system, is shown in Figure 1.  The system contained three parallel channels; and the tumor cells were loaded only into the central channel.  The side channels were constructed to mimic blood vessels in tumors.  To characterize the tumor model for drug delivery studies, a mouse skin cancer cell (B16.F10) was loaded into the central channel with the cell density being similar to that observed in the same tumor grown in mice.  When the side channels were perfused with various solutions of small, fluorescent markers, it was observed that most tumor cells were alive after overnight culture (see Figure 2).  The space between cells was small, which accounted for only 22% of the total volume in the central channel.  The low extracellular space made the tumor structures highly resistant to diffusion of the fluorescent markers.  The resistance was higher for fluorescent molecules that could be taken up by tumor cells (see Figure 3), demonstrating that cellular uptake could prevent the molecules from penetrating into deeper regions.
To investigate bacterial proliferation and transport in tumors, the E. coli was engineered to co-express invasin, a ligand for b1 integrin receptor in tumor cells, and mCherry, a fluorescent marker for bacterial visualization in tumors, so that both bacteria spreading and bacterial-tumor cell interactions could be investigated simultaneously.  In the study, the 3D tumor model was established using two different types of tumor cells: the mouse skin cancer cell (B16.F10) and a mouse breast cancer cell (EMT6), with different levels of b1 integrin receptor.  The E. coli expressing mCherry alone was used as a non-invasive control.  The study showed that the bacteria could spread throughout the entire central channel after overnight culture (see Figure 4).  The spreading was presumably driven by forces generated by bacterial cell division, because there was a dramatic difference in the amount of bacteria during the process of spreading.  The non-invasive bacteria proliferated to a higher extent than the invasive ones; and the rate of proliferation appeared to be tumor cell type dependent.  These data together suggested that tumor cells could secrete inhibitors for bacterial proliferation, and more inhibitors were secreted when tumor cells were treated with invasive E. coli.  Furthermore, an inverse correlation between bacterial cell density and tumor cell density was observed, suggesting that the growth of bacteria could deplete nutrient supply to tumor cells.  The findings described above could aid researchers in developing novel strategies for bacteria mediated cancer treatment.

Scientific Uniqueness:  The tumor model established is versatile for studies of transport and cellular uptake of drugs and genes.  It is not limited to the cells used in the studies mentioned above.  Any tumor cells can be loaded into the microfluidic channel to form 3D structures.  The cell density in the model is comparable to that in tumors grown in animals, which is novel and crucial for transport studies.  A novel application of the model is the investigation of proliferation and transport behaviors of genetically engineered E. coli in tumors.

Strategic Outcome Goals include:
- 1Discovery:  A 3D tumor model has been developed that can be used as a platform for studying drug and gene transport in tumor tissues.  The model itself can be useful in translational cancer research.  The second outcome is the demonstration that genetically engineered, non-pathogenic E. coli is an effective vehicle for gene delivery in tumors.
- 2Learning:  This project has provided research experiences to both undergraduate and graduate students at Duke University.  Specifically, several Duke undergraduate students, including one minority student, did their independent studies in this project.  In addition, three underrepresented undergraduate students did summer research related to the NSF-funded project.  The project has also trained two Ph.D. graduate students.

Intellectual Merit:  The merit is two-fold.  First, it developed a new tumor model for investigation of drug/gene transport and cellular responses to anti-cancer treatment without using animals.  Second, the study demonstrated novel findings on bacterial proliferation and spreading as well as bacterial-tumor cell interactions in the tumor model.

The Broader Impacts of this research include:
- 1Benefits to society:  This NSF-funded research has led to the development of a new tumor model that can be used to facilitate studies of drug and gene delivery to tumor cells, which is currently a challenge in cancer treatment.  It is well known that distributing drugs to all tumor cells is a prerequisite for success in cancer treatment.  Therefore, any improvement in drug and gene delivery can directly enhance clinical outcomes of cancer treatment.  The improvement can also decrease toxicity of drugs and genes in normal tissues, and thus improve the quality of life of cancer patients.
- 2Broadening participation of underrepresented groups:  About 50% of students involved in this NSF-funded project were female.  The project has also provided research experiences to two black undergraduate students and one deaf undergraduate student.  Furthermore, it has trained one black female Ph.D. student.
- 3Advancing discovery and understanding while promoting teaching, training, and learning:  The project has led to three research publications in peer reviewed journals and several presentations at scientific conferences.  Meanwhile, it has been used to train undergraduate and graduate students to perform drug and gene delivery research.  The engineered bacterial cells have also been used in teaching a wet lab project in a required BME undergraduate course at Duke University.
- 4Results disseminated broadly to enhance scientific and technological understanding:  Results from the study will be published in scientific journals and presented in scientific conferences for dissemination.

Program Director:
Semahat Demir
CBET Program Director - Biomedical Engineering
NSF Award Number:   0828630
Award Title:   Delivery of Bacterial Therapeutics to Solid Tumors
Principal Investigator:   Fan Yuan
Institution Name:   Duke University
Program Element Code:   5345
CBET Research Highlight:   Fiscal Year 2012
Approved by CBET on:   23 March 2012

Top of Page

This Research Highlight was Updated on 12 April 2012.