MCF-7 breast cancer cell line grown in agarose culture for study of COX-2 inhibitors in three-dimensional growth system
MCF-7 breast cancer cell line grown in agarose culture for study of COX-2 inhibitors in three-dimensional growth system
David H. Kinder a and Amy L. Aulthouse b
a Department of Pharmacological and Biomedical Sciences, College of Pharmacy, Ohio Northern University, Ada, OH 45810, USA b Department of Biological Sciences, College of Arts and Sciences, Ohio Northern University, Ada, OH 45810, USA
Received 21 March 2003;
accepted 13 October 2003.
Available online 18 February 2004.
Summary: This paper reports a method for studying pharmaceutical COX-2 inhibitors in the MCF-7 breast cancer cell line where cells are grown in a three-dimensional agarose matrix. Cyclo-oxygenase (COX) has been shown to be elevated in several cancers and clinical studies have shown that certain cancer types can be prevented or at least minimized by patients who take cyclo-oxygenase inhibitors. In order to overcome the drawback of the insensitivity of cancer cell lines grown in monolayer format to the COX-inhibitors, this paper utilizes a three-dimensional growth pattern in which cells were grown in 3-D agarose culture in the same method used for human chondrocytes, allowing the system to be free of cellular attachment to plastic and allow in vivo tumor-like growth patterns.
The cell line chosen was the estrogen receptor positive breast adenocarinoma MCF-7, and the treatment was the COX-2 selective drug etodolac (Lodine). MCF-7 cancer cells were suspended in agarose and plated in a cell culture dish where they form small multicellular ‘tumors’ in the agarose. Then, they were treated with COX-2 inhibitors for 2.5 weeks to induce cell death (determined by trypan blue exclusion), and this was compared to cells grown in monolayer given the same drug concentration.
In 3-D culture, it was observed that there was increasing amount of dead MCF-7 cells (either single cells or colonies) with treatment of etodolac, supporting the concept that etodolac, and presumably other COX-2 inhibitors, were eliminating polyps or decreasing tumor formation by inducing a form of cell death, presumably apoptosis. Moreover, etodolac did not show any cytotoxicity or cell growth inhibitory effects in monolayer culture at 250ug/ml drug concentrations, yet it had a clear effect on the growth of cells grown in agarose cultures where 3-D colonies were formed at the same drug concentration.
Ultimately, the paper introduced a method for growing cancer cells that circumvents the growth characteristics of monolayer cultures by allowing 3-D growth habits.
Significance: Breast cancer is one of the most common type of cancer (10.4% of all cancer incidence) and occurs in both women and men worldwide. Although breast cancer in men is about 100 times less common than women, men with breast cancer have the same statistical survival rates as women (American Cancer Society). Being able to mimic the growth of the breast cancer cells in vitro through methods like a three-dimensional agarose matrix allows for a more progressive study on matters like the mechanism of cell death and growth in presence of certain kinds of drugs, ultimately to discover biochemical or physio-chemical factors that will improve or replace biological functions.
For my project in Bioe115, I hope to evaluate the affect of chemical stimuli or inhibitor at varying concentrations on MCF-7 breast cancer cells cultured in well-plates to ultimately measure the secretion of VEGF (vascular endothelial growth factors), which is believed to be the key mediator of angiogenesis in breast cancer. Hopefully, with the aid of this paper, I can perform my study of VEGF production with a 3-D agarose matrix that mimics the environment of a cancer growing in vivo.
19 comments:
Over 180,000 women are diagnosed with breast cancer in the United States every year so indeed, it is very important to conduct research into discovering novel therapies for the treatment of breast cancer. But, I don't understand the reasons why MCF-7 cells when treated with the COX-2 inhibitors do not react the same as in a 3D system. Is it due to contact inhibition/lateral inhibition of cells? Or diminished secretion of specific growth factors?
If COX is a regularly-occurring enzyme in all cells, I imagine etodolac drug delivery must be targeted to release in tumors specifically, which is a problem that has yet to be solved. Did the paper mention the effects of etodolac on regular cells?
This study exposed the cells to etodolac after 2.5 weeks; I wonder how this time corresponds to the time of diagnosis in a patient. If the tumors have more time to grow in a patient, then the dosage may have to increase, which may have cytotoxic effects on neighboring cells.
Your paper mentioned that cancer cells were grown "in 3-D agarose culture in the same method used for human chondrocytes." As I know that dynamic processes are needed to culture chondrocytes since they need some hydrostatic pressure, compression, and other mechanical stimuli to maintain their functions. Do you know if the breast cancer cells need the same condition? Did the paper talk about any details about how they cultured the cells in agarose to maintain the same properties as in human body?
I understood from the paper that there was increased death of MCF-7 in the 3d gel. There were little mention of the controls for the experiments. For instance, was there any experiment that showed that MCF-7 can grow in agarose long term without inducing stress responses? How did the author show that the cells died from the drug and not from stresses induced by being in a 3d culture?
I am also curious about the question that Leland asked. In addition, I was wondering how cyclo-oxygenase inhibitors work in controlling cancer growth? Since the cancer cells produce cyclo-oxygenase at an elevated level, wouldn't the inhibitor have a greater effect on normal cells?
I was wondering how applicable is the MCF-7 line in regards to comparing it to breast cancer in general. Would this only be able to treat a sub-set of breast cancer patients, or are the inhibitory actions taking place able to work in most types of breast cancer.
Could you also explain or go more into the mechanism of cell death, or what the authors predicted caused the cell death?
What causes the insensitivity to a monolayer growth?
Ummm…interesting!! We should have grown our MCF-7 cells in 3D agarose matrix. But here is what I think about their results though. I think it’s harder for cells to grow in 3D agarose matrix than 2D monolayer in general. If I assume the cells in 3D matrix is getting nutrient through diffusion from the media they put above the 3D agarose matrix like we did in our lab (since you didn’t mention it in summery), it’s much harder for cells to get nutrient in this way than in 2D culture. So even without the inhibition effect of COX-2, I expect less cell proliferation/ more cell death in 3D culture than 2D culture.
Ok, I basically agree with the Rina and Aron’s comments about the possible contact inhibition on MCF-7 cells. I actually found a paper that they used MCF-7 cells just released from contact inhibition, which means (from my understanding) MCF-7 cells are/can be contact inhibited.
I tried to found this paper because we were going to seed enough MCF-7 cells so that we can start with almost confluent (more cells from beginning means more VEGF secretion so we can at least detect some results even from low sensitive dot blot.) But, here comes the question…what if the MCF-7 is contact inhibited? That means they will stop growing by not secreting much VEGF (growth factor)? This is why I agreed to start with non-confluent state even that will introduce another variable (cell growth over time) that makes harder to quantify the VEGF secretion rate (amount/cell/day)…we failed to do it anyway >___<””
Again, we should have run the comparison of cell proliferation in 2D & 3D culture without any stimulus/ inhibitor to confirm the level of contact inhibition of MCF-7 cells.
Wait, this might not be enough to conclude the level of contact inhibition…
How about changing the stiffness (concentration) of agarose as variable?
If there is much less cell proliferation in higher concentration agarose matrix, we might be able to conclude that there is contact inhibition/ and its level…?
I don’t know if I’m making sense…
But, nice choice of paper, which gave me another insight for future work for our project =P
Mine is not question by the way~~
I think this paper idea of growing in 3D culture would be very helpful for future MCF-7 experiments to better mimic the human breast microenvironment. However, what I don't understand is why did they need to grow MCF-7 in an agarose culture? Did they compare their results from 3D growth system to what would happen in 2D? Was there a noticeable difference between results in 2D and 3D culture that would impact a study of COX-2 affect on MCF-7 cells? Was this experiment specifically designed due to special properties of COX-2 inhibitors or would it work for all hormones (e.g. estradiol, tamoxifen)?
Hey Audrey,
This paper addresses a crucial concern of growing cells in environments which more accurately mimic the conditions in vivo. However, as others have commented above, uncontrollable and unmeasured factors such as stress caused by the agarose gel shows that even this novel approach is somewhat lacking. I was curious if the paper hypothesized as to what particular properties of the 2D and 3D system caused such a drastic difference in the effect of COX-2 inhibitors on MCF-7 cells.
Rina:
That is definitely a quesion to be addressed. So the paper's purpose was to prove that, from knowing COX-2 inhibitors promote cell death in vivo, a 3-D model, which better mimics an in vivo environment compared to 2-D models, would show the cell death activity from etadolac while 2-D models would not.
This method is actually for the study in determining mechanisms of cell death caused by COX-2 inhibition. Since MCF-7 cells are contact inhibited, the ability to grow better in a 3D system than 2D system may definitely be affected by contact inhibition. I do not remember finding an explanation on the mechanism; however, I know the paper widely emphasized on advantages of a 3D culture system, including in vivo-like morphology, better intercellular interactions, and more realistic biology & function.
Note: For some reason, my paper is no longer available to be observed for free on Springer, ScienceDirect, Pub Med, nor any journal website I can find. I will be able to answer more questions in detail once I find a way to retrieve the paper again. If anyone can find it, please let me know. Thank you for your patience.
Hey Audrey,
somehow when I clicked your paper title, I could download your paper.
So I've just sent it to your gmail.
Good luck!!
Thanks, Masae! Actually I found that it's accessible in my lab, but for some reason not from my home. But I appreciate your help! :)
Rina: If you are questioning why the MCF-7 breast cancer cells do not react the same from COX-2 inhibitor treatment in 2D versus 3D growth system, I got from the paper that the MCF-7 cells do react the same with COX-2 inhibitor treatment in that they show cytotoxic effects, presumably by apoptosis, either in a 2D or 3D sytem; however there is an drug insensitivity problem. Cells cultured and studied in monolayer when exposed to COX-2 inhibitors is that the cells are generally insensitive to the drugs compared to standard cancer chemotherapy agents. Other than that, the paper didn't go into detail. Its focus was in showing that the effect of this drug on the cells is more sensitive, with other advangates, in their 3D agarose system compared to the traditional monolayer system.
Leland: You're completely right. Studying the effect of COX-2 inhibitor in normal cells is important on top of its effect on cancerous cells if we want to process this technique for drug delivery. Unfortunately, the paper did not mention etodolac's effects on normal cells. They are only focusing on the addressing the problem of insensitivity of cancer cell lines to the COX inhibitors by seeking an assay system that would resemble the 3D growth patterns found in cancerous tumors. The authors propose that this system can provide valuable information about the mechanism of cell death and growth in the presence of certain kinds of drugs. So it seems like it's more for research applications rather than drug delivery.
Ye: Good questions. I did not previously think about mechanical conditions needed for cells to grow in a 3D system that better resemble in vivo conditions. It looks to me like the agarose culture, using the same method used for human chondrocytes, was a "permissive environment" for MCF-7 cancer cells. The paper did not use any mechanical stimuli, just chemical (the drug). Specifically, what they did was resuspend cells in 1ml of 0.5% low temperature agarose in DMEM, then plated them in 35mm dishes previously coated with 1% high temperature agarose. This was then cooled to allow agarose to gel, followed by adding 2ml media for cell's nutrients. Some cultures were fed with DMEM and others treated with the control fluid and the drug on top of DMEM. From simply this method, cells seem were able to grow and form colonies within the agarose. However, cell death did occur in the control where no drug was added. Seems like mechanical stimili like you mentioned will help this 3D system to hold a more in-vivo like environment for the cells.
GSudario: Actually, that's a great question. So the reason why COX-inhibitors are studied to combat cancer is that the COX-2 is upregulated in a variety of malignancies and favors the growth of malignant cells by stimulating proliferation and angiogenesis. ngiogenesis, the development of new blood vessels from preexisting capillaries, is essential for the development, growth and advancement of solid tumours. Angiogenesis is enhanced by prostaglandins that are synthesised by the catalysis of cyclooxygenases (COX-1 and COX-2) from arachidonic acid. I would think that as long as the cancer, whatever type it is, expresses COX-2 for angiogenesis would be good candidates for study using this technique. MCF-7 line may be chosen for study because there is a wealth of knowledge about the MCF-7 cell line in the literature. I am not sure about other breast cancer cell lines.
The paper predicted that the inhibition of COX-2 by etodolac caused the cells die or at least inhibit cell growth. The cells were visualized under a light microscope to observe cell morphology for necrosis or apoptosis. The authors concluded that etodolac are eliminating polyps or decreasing tumor formation by inducing a form of cell death, presumably apoptosis. Typical signs of apoptotic cell death are indicated by blebbing and cell contraction.
It is actually not completely clear to me exactly how this method aids in the study of the mechanism behind cell death. It may be that if proven that cells dramatically increase in cell death due to COX-2 inhibitors, then it can be assumed that cell death or at least cell growth is affected by this drug. Another perspective is that the authors suggest their method is good for study on the mechanism of cell death due to the ability to observe cell morphology that denotes the type of cell death.
Cells cultured in monolayer are insenstive to the drugs due to cellular attachments to the plastic and less in vivo tumor-like growth patterns.
MK: I really appreciate your comments! Actually, this paper tested on contact inhibition of the MCF-7 cells with the effect of the drug. The authors measured contact inhibition by measuring the number of colonies (2 or more cells clumped together)over time. In their normal control (without drug, only DMEM), the MCF-7 cells had an increase in the number of colonies from about 10 colonies in week 0.5 to about 55 colonies in week 1.5. (From literature, it is known that MCF-7 cells tend to grow in colonies.) But, there was cell death, although it is not statistically different between 0.5 and 1.5 weeks.
Anyhow, it was good working with you! Maybe we can battle this issue if we work together again! ;)
Cindy: From my understanding, they grew the MCF-7 cells on agarose culture not only b/c it is known through literature that MCF-7 cells can grow on soft agar (Citation of paper: Osborne, C.K. Biological differences among MCF-7 human breast cancer cell lines from different laboratories. Breast Cancer Res.Treat., 9, 1987.), but also b/c it is a "permissive environment" that has already been used for human chondrocytes (Citation: Aulthouse, A.L., Expression of the human chondrocyte phenotype in vivo, invitro cell. Dev. Biol., 25, 1989.).
You're right, they compared the effect of the drug on MCF-7 cells from their 3D growth system using agarose with their results from their monolayer system. The paper suggests quite a noticeable difference. They suggest that "it was clear that etodolac, which did not show any cytotoxicity or cell growth inhibitory effects in monolayer cultures at 250ug/ml concentrations, is having a clear effect on the growht of cells grown in agarose cultures where three-dimensional colonies can form. The paper also suggested that this method can be further used for other drug studies, so it is not limited to COX-2 inhibitor drugs.
Hi Suruchi: You, as well as the others who commented, are very right. This method is indeed lacking the study of other factors from this agarose system on the growth of cells. From what I got from the paper, the 3D system would be free of cellular attachment to plastic and would allow in vivo tumor-like growth patterns, compared to a 2D system. They suggested that the 3D model will offer advantages over monolayer for studying the effects of COX-2 selective inhibitors in vitor by showing that the clusters of cells that grew were adversely affected by the drug at concentrations below that necessary to see an effect in monolayer. Even tho they were able to have data proving more sensitive 3D model compared to 2D model, there wasn't much detail on exactly why.
(editted)
Hi Aron: Thank you for your questions. There were 2 controls for the 3D agarose culture experiment: one called "Normal Control" in which cells were cultured with no drug (only DMEM for nutrients) and another called "Vehicle Control" in which cells were cultured with DMEM and ethanol. Vehicle was done because etodolac is a highly lipophilic drug that requires significant quantities of an organic solvent to solubilize and keep the drug dissolved in the media.
The paper did not measure the stress on the cells. It is good that you pointed out the possibility that the cells may have died due to infavorable environment of the agarose rather than the drug effects. The paper grew cells in monolayer for comparison, in which they exposed increasing concentrations of etodolac over the range of 10-250 ug/ml in 2.5% ethanol in DMEM to match the agarose culture. MCF-7 cell growth inhibition assays were conducted and estimated. From this 2D system, there were no difference in cell growth between high and low concentration of drug treatment. Also, there were no effects of the drug seen on these monolayer cultured cells. Specifically, there were no obvious decrease in cell number, size of cell, or typical signs of apoptotic cell death (indicated by blebbing and cell contraction).
Moreover (and this is probably the answer you are looking for), the authors conducted an experiment with low drug concentration (20ug/ml), which had no statistically significant difference between the Vehicle Control (with DMEM and ethanol) and the 20ug/ml treatment group. They compared this group with another having high drug conc (200ug/ml), which has statistically significant increase in cellular death at 2.5 week time point (compard to 20ug/ml, both cultured in agarose).
I believe they used the 20ug/ml treatment group, which had similar results with the Vehicle Group, to show that the MCF-7 cells were able to grow on the 3D agarose gel w/o any obvious increase in cell death. Then, by showing that cells died with higher drug concentration at 200ug/ml, they were able to suggest that much increase in cell death was caused by the drug and not the agarose.
Actually, COX-2 inhibitors induces cell death, not cell growth. This paper worked on varying the concentrations of COX-2 inhibitors to not really control but to compare cancer death (rather than growth) between 20 and 200ug/ml to prove the obvious effects of the drug on MCF-7 cancer cells. There was not mention about the drug's effect on normal cells, but that is certainly a question to be addressed.
Post a Comment