Philip D. Blood and Dr. William G. Pitt, Chemical Engineering
A major limitation to the effectiveness of cancer chemotherapy is the toxic side effects it has on normal tissues. Thus, it would be highly advantageous to develop a form of treatment that allows higher doses of drugs to be delivered to tumors without increasing the amount of drug absorbed by healthy tissues. To this end, various controlled drug delivery systems are being devised which aim to preferentially target cancerous tissue.
One such system involves encapsulation of cancer drugs in spherical carriers called micelles. These micelles consist of a group of molecules with hydrophobic and hydrophilic ends. In an aqueous solution such as blood, and at certain threshold concentrations, these molecules spontaneously congregate to form a sphere with their hydrophilic ends facing out, and their hydrophobic ends facing in. Small, hydrophobic molecules, such as doxorubicin (DOX) and many other common cancer drugs, become sequestered within the hydrophobic core of the micelles. If these micelles can then be caused to release the drug only at tumor sites, preferential targeting of tumor cells is achieved.
The drug delivery system investigated in this report involves the use of low frequency ultrasound to release drugs from micelles made of the polymer Pluronic P-105. In previous work, when human leukemia cells were incubated with DOX and P-105 in the absence of ultrasound, the cells were completely protected from the DOX; however, when ultrasound was applied to the same solution, more cells were killed than by DOX alone.1 Thus, this drug delivery system has the potential to increase the effectiveness of chemotherapy on solid tumors in two ways: first, by protecting healthy tissues from the toxic effect of the drugs, and second, by increasing the toxicity of the drug when it is released by ultrasound at the tumor site.
The objective of this study is to determine whether this treatment can be effective on cells that are attached to a surface (anchorage-dependent cells). The leukemia cells used previously grow in suspension and serve only as a general model of a cancer cell. This treatment could not actually be used to treat leukemia because ultrasound can only be focused on localized solid tumors. Therefore, we must show that this treatment also works on anchorage-dependent cells like breast cancer cells.
To test this system, we grew MCF-7 breast carcinoma cells in plastic 96-well plates. The cells were subjected to combinations of the following variables: high (10μg/ml), medium (1μg/ml), or low (0.1μg/ml) concentrations of DOX; presence or absence of P-105; treatment with or without ultrasound; and varying treatment durations.
To treat the cells with ultrasound, we floated the 96-well plates in an ultrasound bath. We had to be careful to maintain the sterility of the plate during the ultrasound treatment. After trying to seal the plate with various substances (silicon, hot glue, silly putty) we finally discovered a polyurethane membrane that we used to cover the plate. This membrane permits gas exchange, but keeps out microorganisms.
This discovery was significant because it helped us solve another perplexing problem: overheating of the wells during sonication. We found that the temperature in the wells could reach temperatures in excess of 50° C during ultrasound treatment. The polyurethane membrane alleviated this problem since it allowed the heated water vapor to escape. However, to completely solve the problem we had to fill the unused wells and the spaces between the wells with sterile water and maintain the ultrasound bath at 30° C.
The plate had to be placed in the same location within the bath to keep the ultrasound levels consistent between experiments. We designed a Styrofoam mould to keep the plate in position, and then we measured the ultrasound level in each well with a hydrophone. This allowed us to choose wells of similar ultrasound level in which to treat our cells.
To treat the cells, we removed the normal growth medium and replaced it with medium containing the various combinations of DOX and/or P-105. Following the various treatment times, we washed the cells to remove the DOX and P-105. We then replaced the normal growth medium and allowed the cells to grow for a few days. We assessed the effectiveness of each treatment by measuring the total cellular protein content of each well after the designated incubation period. This was done using the bicinchoninic acid (BCA) protein assay. Previous work demonstrated a linear relationship between total cellular protein and cell number as measured by the BCA assay.2 Thus, the lower the protein content of a given set of wells following incubation, the greater the effectiveness of the treatment.
Although we used a poly-amino acid to attach the cells firmly to the wells, we discovered that we were losing cells during the washing steps. By lowering the number of washes and improving our washing technique, we were able alleviate this problem, but we have not yet solved it. For this reason, a significant amount of experimental error was introduced into our results.
Because of these difficulties, the only conclusion the data support so far is that P-105 micelles protect anchorage-dependent cells from low and medium concentrations of DOX in the absence of ultrasound. Unexpectedly, the data suggest that sonicated free DOX at low and medium concentrations kills anchorage dependent cells more effectively than sonicated DOX with P-105. We are working on reducing the experimental error in order to reliably determine whether ultrasound-activated release of DOX from P-105 micelles can enhance killing of anchorage-dependent cells compared to unsonicated free DOX.3
References
- G.A. Husseini, R.I. El-Fayoumi, K.L. O’Neill, N.Y. Rapoport, W.G. Pitt, DNA damage induced by micellardelivered doxorubicin and ultrasound: comet assay study, Cancer Lett. 154 (2000) 211-216.
- A.M. Hall, V. Croy, T. Chan, D. Ruff, T. Kuczek, C. Chang, Bicinchoninic acid protein assay in the determination of adriamycin cytotoxicity modulated by the MDR glycoprotein, J. Nat. Prod. 59 (1996) 35-40.
- I am indebted to the BYU Cancer Research Center and ORCA for funding this study, as well as to Dr. Kim O’Neill and Dr. Gregory Burton for kindly allowing me to use their research facilities.