Cancer stem cells (CSCs) have been mentioned here before in passing – recently here, for example. But it's now time to direct particular attention to them.
One reason is that a number of new experimental results concerning CSCs were presented last month at the American Association for Cancer Research meeting in San Diego.
Another reason is that CSCs, like other stem cells, happen to utilize a number of the signaling pathways that are very interesting in connection with cancer, embryonic development, and various other cellular processes. The list includes Myc, Nanog, Notch, Sonic hedgehog (Shh), TOR, and Wnt. This is a rapidly developing field of research, and CSCs are right in the thick of things.
There is also some overlap with the recent active work going on with induced pluripotent stem cells (IPSCs). For example, Myc and Nanog turn up in IPSC research.
Finally, CSCs are also controversial (as is to be expected of any field that's in rapid flux) – and controversies are inherently interesting to read about.
One of the controversies concerns whether CSCs really exist and are important to the extent their principal investigators tend to believe. A reason to be skeptical of their importance is that there are some indications that CSCs should not be too rare, and that they should also be largely resistant to standard chemotherapy drugs. Yet in many cancers, chemotherapy can kill at least 99% of tumor cells, which would therefore include most CSCs. But this reasoning, too, is controversial.
But let's go back to first principles and explain just what a CSC is. Even that is tricky, since there are several models hypothesized for CSC behavior. In most general terms, however, a CSC is much like other adult stem cells, in that when a CSC divides, one daughter cell is an essentially similar CSC, while the other daughter is a more ordinary tumor cell, which does not have as much ability to proliferate by repeated division.
This is already somewhat different from the older picture of cancer. In that picture, one "renegade" cell first acquires somehow a mutation of its DNA which defeats one of the numerous cellular safeguards against uncontrolled cell proliferation. As time goes on, descendants of that cell accumulate further mutations that defeat other safeguards. Eventually, a significant population of cells exists which have a similar set of mutations that can evade most of the safeguards, and at that point, a dangerous tumor starts to grow. In effect, most of the cells of the tumor are stem cells, which continually produce more of their kind, without serious inhibition.
But it is gradually became clear that such a naive picture couldn't be generally correct. The main reason is that many cancerous cells stop dividing after awhile, and they are not capable of seeding a new tumor if (for instance) they are introduced into another part of an experimental animal having the tumor, or into a tumor-free animal. Most cells cannot divide an arbitrarily large number of times. There is a limit to the number of cell divisions, called the Hayflick limit. This limit, typically, is about 52 divisions. When the limit is reached, a cell does not necessarily die, but it does enter a new phase of life called the senescence phase, in which it ceases to divide.
The main reason for the Hayflick limit is a structure, called a telomere, found at the ends of all chromosomes. Telomeres consist of repeated short segments of DNA and protect the chromosome from damage during cell division. But a telomere shortens each time a cell divides, and once it becomes too short, the cell becomes senescent. This is because the mechanisms that protect a cell's DNA normally do not permit the cell cycle to function when telomeres are too short. Even if these mechanisms could be evaded, the cell's DNA would be damaged in the division process, and the resulting malfunctions would likely cause the cell to die.
However, stem cells need to be able to divide far beyond the Hayflick limit. Embryonic stem cells need this ability in order for an embryo to grow from a single cell into an organism with trillions of cells. Adult stem cells also need this ability to be able to replace cells in certain tissues that must be frequently renewed, such as in the skin and the lining of the intestines. In order to make this possible, stem cells express an enzyme, called telomerase, which has the specific function of rebuilding shortened telomeres.
Many, but not all, cancer cells also express telomerase. Only those that do are capable of indefinitely dividing. So such cancer cells have the production of telomerase in common with stem cells. That still doesn't mean that they are CSCs, because they may lack other attributes of stem cells. In particular, they may not be able to produce at least one exact copy of themselves when dividing. Instead, both daughter cells may be more differentiated and specialized than the mother cell, and (among other things) lose the ability to produce telomerase. Once that happens, the daughter cells' fate is eventual senescence, at best.
There are several possible ways in which CSCs might arise. One possibility is that all mutations necessary for a growing cancer occur, over a period of time, in a population of actual stem cells. (Once the mutations first occur, they of course are inherited by all descendants.) Another possibility is that only some of the mutations occur in the stem cell phase, while other mutations necessary for cancer occur after the full "stemness" is lost (but while the cells are still capable of repeated division). Perhaps a late stage of cancer development is that cells with enough oncogenic mutations to become cancerous reacquire an ability to divide repeatedly, if that was lost along the way.
In fact, there is evidence that both of these models, and possibly others, actually occur. It all boils down to the time sequence of mutation events in a particular type of cancer, and the sequence is likely to be different in different types of cancer. In any case, one of the main tasks of cancer research now is to figure out not only what mutations occur in a particular type of cancer, but also the order in which they typically occur. The hope is that sufficient knowledge of this process will make it possible to devise ways to stop it.
Here are some other things to read for additional background:
Update, 5/17/08: I've written up some summaries of recent research on cancer stem cells here.
Additional links:
The Cancer Stem Cells Project
Tags: cancer, stem cells, cancer stem cells
One reason is that a number of new experimental results concerning CSCs were presented last month at the American Association for Cancer Research meeting in San Diego.
Another reason is that CSCs, like other stem cells, happen to utilize a number of the signaling pathways that are very interesting in connection with cancer, embryonic development, and various other cellular processes. The list includes Myc, Nanog, Notch, Sonic hedgehog (Shh), TOR, and Wnt. This is a rapidly developing field of research, and CSCs are right in the thick of things.
There is also some overlap with the recent active work going on with induced pluripotent stem cells (IPSCs). For example, Myc and Nanog turn up in IPSC research.
Finally, CSCs are also controversial (as is to be expected of any field that's in rapid flux) – and controversies are inherently interesting to read about.
One of the controversies concerns whether CSCs really exist and are important to the extent their principal investigators tend to believe. A reason to be skeptical of their importance is that there are some indications that CSCs should not be too rare, and that they should also be largely resistant to standard chemotherapy drugs. Yet in many cancers, chemotherapy can kill at least 99% of tumor cells, which would therefore include most CSCs. But this reasoning, too, is controversial.
But let's go back to first principles and explain just what a CSC is. Even that is tricky, since there are several models hypothesized for CSC behavior. In most general terms, however, a CSC is much like other adult stem cells, in that when a CSC divides, one daughter cell is an essentially similar CSC, while the other daughter is a more ordinary tumor cell, which does not have as much ability to proliferate by repeated division.
This is already somewhat different from the older picture of cancer. In that picture, one "renegade" cell first acquires somehow a mutation of its DNA which defeats one of the numerous cellular safeguards against uncontrolled cell proliferation. As time goes on, descendants of that cell accumulate further mutations that defeat other safeguards. Eventually, a significant population of cells exists which have a similar set of mutations that can evade most of the safeguards, and at that point, a dangerous tumor starts to grow. In effect, most of the cells of the tumor are stem cells, which continually produce more of their kind, without serious inhibition.
But it is gradually became clear that such a naive picture couldn't be generally correct. The main reason is that many cancerous cells stop dividing after awhile, and they are not capable of seeding a new tumor if (for instance) they are introduced into another part of an experimental animal having the tumor, or into a tumor-free animal. Most cells cannot divide an arbitrarily large number of times. There is a limit to the number of cell divisions, called the Hayflick limit. This limit, typically, is about 52 divisions. When the limit is reached, a cell does not necessarily die, but it does enter a new phase of life called the senescence phase, in which it ceases to divide.
The main reason for the Hayflick limit is a structure, called a telomere, found at the ends of all chromosomes. Telomeres consist of repeated short segments of DNA and protect the chromosome from damage during cell division. But a telomere shortens each time a cell divides, and once it becomes too short, the cell becomes senescent. This is because the mechanisms that protect a cell's DNA normally do not permit the cell cycle to function when telomeres are too short. Even if these mechanisms could be evaded, the cell's DNA would be damaged in the division process, and the resulting malfunctions would likely cause the cell to die.
However, stem cells need to be able to divide far beyond the Hayflick limit. Embryonic stem cells need this ability in order for an embryo to grow from a single cell into an organism with trillions of cells. Adult stem cells also need this ability to be able to replace cells in certain tissues that must be frequently renewed, such as in the skin and the lining of the intestines. In order to make this possible, stem cells express an enzyme, called telomerase, which has the specific function of rebuilding shortened telomeres.
Many, but not all, cancer cells also express telomerase. Only those that do are capable of indefinitely dividing. So such cancer cells have the production of telomerase in common with stem cells. That still doesn't mean that they are CSCs, because they may lack other attributes of stem cells. In particular, they may not be able to produce at least one exact copy of themselves when dividing. Instead, both daughter cells may be more differentiated and specialized than the mother cell, and (among other things) lose the ability to produce telomerase. Once that happens, the daughter cells' fate is eventual senescence, at best.
There are several possible ways in which CSCs might arise. One possibility is that all mutations necessary for a growing cancer occur, over a period of time, in a population of actual stem cells. (Once the mutations first occur, they of course are inherited by all descendants.) Another possibility is that only some of the mutations occur in the stem cell phase, while other mutations necessary for cancer occur after the full "stemness" is lost (but while the cells are still capable of repeated division). Perhaps a late stage of cancer development is that cells with enough oncogenic mutations to become cancerous reacquire an ability to divide repeatedly, if that was lost along the way.
In fact, there is evidence that both of these models, and possibly others, actually occur. It all boils down to the time sequence of mutation events in a particular type of cancer, and the sequence is likely to be different in different types of cancer. In any case, one of the main tasks of cancer research now is to figure out not only what mutations occur in a particular type of cancer, but also the order in which they typically occur. The hope is that sufficient knowledge of this process will make it possible to devise ways to stop it.
Here are some other things to read for additional background:
- Stemming the tumorous tide (4/17/08) – article at Economist.com, with recent information from the American Association for Cancer Research meeting in San Diego
- Scientists Weigh Stem Cells’ Role as Cancer Cause (12/21/07) – New York Times article by Gina Kolata; gives some history and CSC skeptic opinions
- Stem Cells: The Real Culprits in Cancer? (7/1/06) – detailed Scientific American article, one of whose authors, Michael F. Clarke, has been involved in significant CSC research
- Stem Cells May Be Key to Cancer (2/21/06) – New York Times article by Nicholas Wade
Update, 5/17/08: I've written up some summaries of recent research on cancer stem cells here.
Additional links:
The Cancer Stem Cells Project
Tags: cancer, stem cells, cancer stem cells