MicroRNA and stem cells II

MicroRNA and stem cells are both pretty hot topics these days. But curiously, there haven't been a whole lot of reports that involve the two together. My last discussion of both, back in March, is here.

However, the combination of microRNA and stem cells just was back in the news, as briefly noted here.

As you recall, microRNA refers to small single-stranded RNA molecules, generally about 21 to 23 nucleotides in length. Each different microRNA (miRNA for short) is transcribed from a DNA gene like any other gene, but the resulting RNA isn't translated into a protein. Instead, the typical miRNA functions by downregulating the expression of another gene that codes for a protein.

An embryonic stem cell (ESC) has the property of pluripotency, which means that it is capable of giving rise to essentially any type of cell in the body of a multicellular organism. Whenever an ESC divides, the resulting daughter cells may also by ESCs (hence pluripotent) or they may be more specialized cells that will eventually give rise to some type of adult body cell.

In any particular ESC, the determination of remaining pluripotent or instead heading down the path to a more specialized cell type depends on what set of genes are expressed. Since miRNAs downregulate gene expression, they can keep an ESC in its pluripotent state, if they are active and suppress a gene that would make an ESC more specialized. But then if such a miRNA is blocked from being expressed, the ESC can start to become more specialized.

On the other hand, as we shall see, some miRNAs may block transcription factors that are needed to maintain a pluripotent state. Such an miRNA needs to be silent in an ESC, so some other protein needs to suppress its expression. (The miRNA called miR-21, discussed later, is an example.)

So the name of the game in studying ESCs, as far as miRNA is concerned, is to figure out what causes a miRNA gene to be expressed or not. Like other genes in an ESC, which genes are expressed is strongly controlled by a few master transcription factors.

There are four such transcription factors which seem to be especially important in ESCs: Oct4, Sox2, Nanog, and Tcf3. As discussed here, the first three of these factors have been found capable of playing a role in turning an ordinary adult cell into a pluripotent stem cell (called an "induced pluripotent stem cell").

Currently there are 336 mature mouse miRNAs known, and 441 mature human miRNAs. It is simple (given the known, complete sequences of mouse and human genomes) to locate the genes for each miRNA. However, in order to determine when a transcription factor regulates the miRNA gene, the promoter for the gene (a separate portion of DNA) must also be located. In order for a gene to be expressed, the right transcription factors have to bind to the gene's promoter.

Finding promoters is a lot harder, but there are techniques that involve searching for methylation of histone proteins that make up the nucleosomes around which cellular DNA is wrapped.

It was known, before the recent research we're discussing, that there were 14,230 sites in the genome where all four of the named master transcription factors could bind simultaneously. Most of those sites were not promoters of some miRNA, but it was straightforward to identify those that were. Of those miRNAs that appeared to be regulated by Oct4, Sox2, Nanog, and Tcf3, it was found that most are in fact preferentially expressed in ESCs. This set of miRNAs would seem to be good candidates for maintaining ESC pluripotency by downregulating other genes.

On the other hand, some of the miRNAs mediated by the transcription factors are silent in ESCs. Subsequent research found that another type of proteins (polycomb proteins) also bind to the miRNA promoters. These proteins were already known to block transcription by binding to gene promoters. But it turns out that some of these silenced miRNAs become active once the ESC loses its pluripotency and begins to differentiate.

The next step will be to figure out what each of these miRNAs regulated by Oct4, Sox2, Nanog, and Tcf3 actually does – either in the ESC or a differentiated cell. That should be very interesting, as the press release suggests:

Putting microRNAs on the stem cell map (8/7/08)

“We now have a list of what microRNAs are important in embryonic stem cells,” says Alex Marson, co-lead author on the paper and an MD/PhD student in the Young lab. “This gives us clues of which microRNAs you might want to target to direct an embryonic stem cell into another type of cell. For example, you might be able to harness a microRNA to help drive an embryonic stem cell to become a neuron, aiding with neurodegenerative disease or spinal cord injury.”

Moreover, the results give scientists a better platform for analyzing microRNA gene expression in cancer and other diseases. “We and others are finding that the overall gene circuitry for embryonic stem cells and cancer cells is very similar,” notes Marson. “Now that we have connected the circuitry to microRNAs, we can begin to compare microRNAs that are regulated in embryonic stem cells to those in cancer cells.”

Here's a somewhat more detailed description of the research: Stem Cell microRNA, Transcription Factor Interplay Uncovered (8/8/08)

Other research on miRNA and ESCs that has appeared since the previous discussion (here) gives a small taste of what may be learned about the miRNAs silenced in ESCs:

Protein Protects Embryonic Stem Cells' Versatility And Self-renewal (3/23/08)

A protein known as REST blocks the expression of a microRNA that prevents embryonic stem cells from reproducing themselves and causes them to differentiate into specific cell types, scientists at The University of Texas M. D. Anderson Cancer Center report in the journal Nature.

Researchers show RE1-silencing transcription factor (REST) plays a dual role in embryonic stem cells, said senior author Sadhan Majumder, Ph.D., professor in M. D. Anderson's Department of Cancer Genetics. "It maintains self-renewal, or the cell's ability to make more and more cells of its own type, and it maintains pluripotency, meaning that the cells have the potential to become any type of cell in the body."

The details are particularly interesting:

In studies using mouse embryonic stem cells, the researchers found that REST disarms a specific microRNA called microRNA-21 or miR-21. MicroRNAs are tiny pieces of RNA that control gene expression by binding to the gene's messenger RNA.

The team found that MiR-21 suppresses embryonic stem cell self-renewal and is associated with a corresponding loss of expression of critical self-renewal regulators, such as Oct4, Nanog, Sox2 and c-Myc. REST counters this by suppressing miR-21 to preserve the cells' self-renewal and pluripotency.

The researchers discovered the roles of REST and miR-21 in a series of experiments using cultured mouse embryonic stem cells in either a self-renewal state or a differentiating state. They found that REST expression was significantly higher in the self-renewal state. Withdrawing REST reduced the stem cells' ability to reproduce themselves and started differentiation -- even when the cells were grown under conditions conducive to self-renewal. Adding REST to differentiating cells maintained their self-renewal.

These experiments also revealed that REST is bound to the gene chromatin of a set of microRNAs with the potential to target self-renewal genes. REST controls transcription of 11 microRNAs.

Is anything special known about miR-21? Yes, in fact – it is known to play a role in cancers of the colon, liver, and thyroid. (See here.)

Tags: microRNA, stem cells, embryonic stem cells

Cancer stem cells II

Since we've just had a discussion of generalities about cancer stem cells (here), it seems like it would be fun to have a summary of research on CSCs that was presented at the recent meeting of the American Association for Cancer Research in San Diego, or reported since then. So here it is.

As general background, keep in mind that cell surface proteins CD44 and CD24 are considered to be markers of cancer stem cells in various (but not necessarily all) types of cancer. Also, certain cell signaling pathways are thought to be especially important in the activity of cancer stem cells. The list includes Wnt, Sonic hedgehog, Notch, and Bmi1.

Stem Cell-Like Cancer Cells Resistant To Standard Therapy, Responsive To Targeted Therapy (4/29/08)

Previous research had identified a subset of cells in breast tumors that have the ability to form colonies in culture and give rise to tumors in mouse models. Such cells are thought to be cancer stem cells. They express the cell surface glycoprotein CD44, but not CD24, and they appear to be resistant to standard chemotherapeutic agents. However, the drug lapatinib, which inhibits the HER2 pathway, seems to selectively kill these cells.

Getting To The Roots Of Breast Cancer (4/29/08)

This is another report on the research described in the previous item. It provides additional details on the research protocol.

Stem Cell Type Supposed To Be Crucial For Angiogenesis And Cancer Growth Does Not Exist? (4/22/08)

This study casts doubt on the existence of a certain type of bone-marrow derived stem cell that has been suspected of circulating in the blood and acting as a precursor to endothelial cells that make up blood vessel walls. Such cells, if they existed, would be an important target for inhibiting angiogenesis in tumors. The researchers showed, using advanced techniques with mouse models, that endothelial differentiation is not a typical function of bone-marrow derived stem cells.

Ovarian Cancer Stem Cells Identified, Characterized (4/17/08)

Researchers have identified, characterized and cloned ovarian cancer stem cells and have shown that these stem cells may be the source of ovarian cancer's recurrence and its resistance to chemotherapy. They isolated cells from samples of either peritoneal fluid or solid tumors. The cancer stem cells that were identified had traditional cancer stem cell markers including CD44 and MyD88 (which interacts with toll-like receptors to activate NF-κB).
The cells also showed a high capacity for repair and self-renewal. Such cells, when isolated, were capable of forming tumors 100 percent of the time. Within those tumors, 10 percent of the cells were CD44 positive, while 90 percent were CD44 negative, indicating that some cancer stem cells had undergone differentiation.

Stem Cells: The Role Of Cancer-initiating Cells In Diagnosis And Treatment (4/15/08)

This press release describes research presented at the AACR meeting related to stem cells and pancreatic, bladder, ovarian, and breast cancer, and glioma.

Research in pancreatic cancer found that in addition to CD44 and CD24, the enzyme aldehyde dehydrogenase was expressed in a small population of tumor cells. Cells expressing aldehyde dehydrogenase had greater growth capacity than those that didn't, and they were also associated with poorer overall survival.

In a study of breast cancer and glioma, surface markers were not found be sufficient as markers of stem cell activity. However, cells with low proteasome activity did have notably greater capacity for self-renewal and tumor production capacity. (Proteasomes are large protein complexes that degrade unneeded or damaged proteins.)

Researchers studying bladder transitional cell carcinomas found, in 40% of cases, CD44+ cells with other stem cell self-renewal patterns. In these cells, 85% had active Gli1, a part of the Hedgehog pathway, originally discovered in human glioblastoma. A relatively small percentage had active Bmi1, Stat3, or β-catenin (part of the Wnt pathway). None had active Oct4 or Nanog (pluripotent stem cells markers).

The research on ovarian cancer (noted above) involving CD44 and MyD88 markers is referenced again here.

Stem Cell Marker Controls Two Key Cancer Pathways (4/14/08)

Research into breast cancer stem cells has identified, for the first time, another gene that may be involved, Msi1. The investigators showed that Msi1 activated Wnt and Notch signaling. Other studies have shown that Msi1 is a marker of human adult stem cells in general because it has been found in human breast, colon, brain, skin, and other cells. Msi1 was found to affects mammary cells to influence whether they develop into muscle, milk duct linings, etc. Further, Msi1 was found to be expressed in particularly aggressive tumors.

Stem Cells And Cancer: Scientists Investigate A Fine Balancing Act (4/11/08)

This is a report of a general talk about how the mechanisms normally involved in balancing different functions of stem cells may also contribute to cancer. For example, research shows that Bmi1 is important for maintaining stocks of stem cells, and without it stocks of stem cells are depleted. But also Bmi1 is overactive in various cancers including brain tumors.

Secrets of cellular signaling shed light on new cancer stem cell therapies (4/10/08)

Researchers are beginning to study inhibition of signaling pathways that seem to be active in tumors fed by cancer stem cells. In this case, inhibition of the Notch pathway is being investigated as part of a treatment, together with chemotherapy, for metastatic breast cancer. An important question is whether cancer stem cells are sufficiently different from normal adult stem cells so that inhibition of Notch signaling is not harmful. Results of testing in mice indicate that Notch signals are not required for the maintenance of blood-forming stem cells in adult mice.

Stem cells and cancer: cancer pathways that also control the adult stem cell population (4/10/08)

Apc (adenomatosis polyposis coli) is a tumor-suppressing protein that controls β-catenin and hence affects Wnt signaling. When intestinal crypts are damaged and need to be regenerated, Wnt signaling directs stem cells to generate replacement cells. Apc normally turns off Wnt signaling of stem cells when it is no longer needed. The research here showed that if Apc is lost or damaged, Wnt signaling may continue and result in tumor formation

Cancer Stem Cells Created With New Technique (4/9/08)

One of the most important unresolved questions about cancer stem cells is how they originate to begin with. For instance, are they mutations of existing stem cells, or instead precancerous cells that have acquired stem-cell-like capabilities? The research here supports the latter scenario. Starting with normal skin cells, the researchers activated three genes associated with embryonic stem cells. The result closely resembled known cancer stem cells. And they also had more resemblance to normal embryonic stem cells than to normal adult stem cells. One of the genes was Myc, which has also been used to create pluripotent stem cells from skin cells. In addition to the scientific significance of this work, it should also facilitate study of cancer stem cells, which are otherwise hard to locate.

Module Map Links Embryonic Stem Cells And Cancer Stem Cells (4/9/08)

The researchers involved in the work described in the preceding report have additional related findings. They systematically compared gene expression patterns between embryonic stem cells and multiple types of human cancer cells. Gene expression patterns in diverse human epithelial cancers were much like patterns in embronic stem cells. Further, presence of these patterns in cancer cells strongly predicted metastasis and death. On the other hand, normal adult tissue stem cells had an opposite pattern, which was repressed in various human cancers compared to normal tissues. The researchers additionally demonstrated that c-Myc, but not other oncogenes, was sufficient to reactivate the ESC-like program in normal and cancer cells.

And here's some earlier research that features the Nanog and Bmi1 proteins:

To Evade Chemotherapy, Some Cancer Cells Mimic Stem Cells (9/19/07)

Anti-cancer treatments often effectively shrink the size of tumors, but some might have an opposite effect, actually expanding the small population of cancer stem cells believed to drive the disease, according to new findings.

"Our experiments suggest that some treatments could be producing more cancer stem cells that then are capable of metastasizing, because these cells are trying to find a way to survive the therapy," said one of the study's investigators, Vasyl Vasko.

When the researchers applied various anti-cancer drugs to experimental cancer cells, they found that surviving cells expressed more Nanog and Bmi1:

They selected a rare form of cancer, mesenchymal chondrosarcoma (MCS), which has not been well described and for which there is no effective treatment. The researchers first determined that Nanog and BMI1 stem cell markers were more highly expressed in metastatic tumors compared to primary tumors. ...

They then applied various therapies - from VEGF inhibitors such as Avastin to the proteasome inhibitor Velcade - in mice implanted with human MSC, and analyzed the effects on tumors. Some of the treatments seemed to work, because they led to a dramatic decrease in the size of the tumors, Dr. Vasko said. But analysis of stem cell expression before and after treatment revealed that even as some anti-cancer treatments shrank tumors, they increased expression of Nanog and BMI1. "These treatments were not enough to completely inhibit tumor growth, and the cancer stem cell markers were still present," Dr. Vasko said.

Tags: cancer, stem cells, cancer stem cells

Induced Pluripotent Stem Cells II

In this article from the April 4 Science, which I mentioned here, several research reports dealing with induced pluripotent stem cells were discussed. One of these I covered in the post I just noted.

Another just as important report apparently has not yet been formally published, but is (at least temporarily) available online since February 14 at Science Express:

Generation of Pluripotent Stem Cells from Adult Mouse Liver and Stomach Cells

Induced pluripotent stem (iPS) cells have been generated from mouse and human fibroblasts by the retroviral transduction of four transcription factors. However, the cell origins and molecular mechanisms of iPS cell induction remain elusive. This report describes the generation of iPS cells from adult mouse hepatocytes and gastric epithelial cells. These iPS cell clones appear to be equivalent to ES cells in gene expression and are competent to generate germ-line chimeras.

It's not surprising that this is significant research, as it's from the same team of Shinya Yamanaka that was the first to report successful creation of induced pluripotent stem cells. (See here.)

So what is this research about? Well, the investigators used the same four transcription factors (Oct3/4, Sox2, Klf4, and c-Myc) as employed in the majority of previous iPS studies. However, instead of applying the transcription factors to fibroblast cells, they were applied to two types of epithelial cells instead.

Fibroblasts are part of a body's connective tissue. They are involved in structure and support for other tissues and contain large amounts of the protein collagen. They do not divide for the most part, and so it is especially significant that it was possible to reprogram them into a stemcell-like state at all.

Epithelial cells, on the other hand, line the inner and outer surfaces of various body structures, including skin and the gastrointestinal tract. Such cells divide more frequently. They have to, in order to replace other cells of the same kind that are exposed to hostile environments. Epithelial cells also tend to be more adherent to other cells, because they more highly express an adherence protein called E-cadherin.

In some sense, then, epithelial cells are a little more like stem cells to begin with, so one might expect better results when attempting to reprogram them.

This expectation seems to have been met. One of the key differences the researchers found is that reprogrammed epithelial cells had less tendency to form cancerous tumors in mice into which they were included. Certainly not an inconsiderable advantage. This characteristic may be related to the finding that c-Myc seems to play a less essential role in reprogramming epithelial cells.

Specifically, reprogramming of epithelial cells was almost as efficient when c-Myc was not used as when it was included with the other three transcription factors. Yet it was not possible to accomplish reprogramming if any of the other three factors was omitted. In contrast, the efficiency of reprogramming fibroblasts dropped by 90% when c-Myc was omitted.

Another intriguing difference was that reprogrammed epithelial cells contained higher levels of expression of β-catenin than reprogrammed fibroblasts did. (You may recall – see here – that β-catenin is an important part of the Wnt signaling pathway.) In this regard, the reprogrammed epithelial cells are more like true embryonic stem cells than reprogrammed fibroblasts are. It's probably not a coincidence that expression of Nanog is stimulated by β-catenin, (see here), since Nanog is considered important for maintaining stem cell pluripotency.

A further advantage of the use of epithelial cells is that many fewer retroviral "integration sites" were needed to include the transcription factor genes into the cell genome, in comparison with fibroblasts. This is another way the risk of cancer is reduced.

Further reading:

• Stomach and liver cells reprogrammed – excellent 2/28/08 open access overview of the research from Nature Reports Stem Cells
  
  
• Epithelial cells made pluripotent – 2/14/08 blog post at The Scientist giving a good summary of the research findings
  
  
• Epithelial cells made pluripotent – blog post at Wisconsin Stem Cell Now on the preceding article, but noted mainly because it's by the owner of another interesting stem cell blog: Stem Cell Nation
  

Tags: stem cells, induced pluripotent stem cells

Embryonic stem cells and Klf4

There's now some additional information on one of the transcription factors written about here, which are able to reprogram adult skin cells into embryonic stem cells. To review, one of the teams responsible for this research used Oct3/4, Sox2, c-Myc, and Klf4 for the reprogramming, while another team used Oct3/4, Sox2, Nanog and Lin28.

Of the transcription factors in the first list, all but Klf4 have been well-studied. So it is of some interest to know more about Klf4, and why it seems to be somewhat less essential than the others.

Some of the interesting details are reported on here: Molecular Alliance That Sustains Embryonic Stem Cell State Identified.

Klf4 is normally active in real embryonic stem cells. To investigate the role Klf4 might be playing in the reprogramming of skin cells, the researchers investigated embryonic stem cells that had been artificially depleted of Klf4. To their surprise, the team found that the cells maintained their pluripotency.

The question then was how to explain this. What was found is that two closely-related transcription factors – Klf2 and Klf5 – took over the role of Klf4:

"Most important, the data showed that the other Klfs were bound to the target sites when one of them was depleted." said Dr. Ng. "These Krüppel-like factors form a very powerful alliance that work together on regulating common targets. The impact of losing one of them is masked by the other two sibling molecules."

This family of transcription factors, called Kruppel-like factors, gets its name from a homology to the Drosophila Krüppel protein. Members of this family have been studied for their roles in cell proliferation, differentiation and survival, especially in the context of cancer.

Interestingly enough, according to the research press release,

Klfs were found to regulate the Nanog gene and other key genes that must be active for ES cells to be pluripotent, or capable of differentiating into virtually any type of cells. Nanog gene is one of the key pluripotency genes in ES cells.

"We suggest that Nanog and other genes are key effectors for the biological functions of the Klfs in ES cells," Dr. Ng said.

"Together, our study provides new insight into how the core Klf circuitry integrates into the Nanog transcriptional network to specify gene expression unique to ES cells.

Nanog, of course, is one of the transcription factors in the set of transcription factors which was found to be an alternative, for reprogramming adult cells, to the set that contained Klf4.

The Nanog protein, too, is known to be critically important in pluripotent stem cells. It is a homeobox transcription factor that appears to play an essential role in self-renewal of undifferentiated embryonic stem cells. It also appears to be connected with cancer, because (according to Wikipedia) "It has been shown that the tumour suppressor p53 binds to the promoter of NANOG and suppresses its expression after DNA damage in mouse embryonic stem cells. p53 can thus induce differentiation of embronic stem cells into other cell types which undergo efficient p53-dependent cell-cycle arrest and apoptosis."

The connection of Klf proteins with cancer is not only through Nanog. According to Wikipedia, "Klf4 also interacts with the p300/CBP transcription co-activators." The closely-related p300 and CBP "interact with numerous transcription factors and act to increase the expression of their target genes." And they too are involved with cancer:

Mutations in the p300 gene have been identified in several other types of cancer. These mutations are somatic, which means they are acquired during a person's lifetime and are present only in certain cells. Somatic mutations in the p300 gene have been found in a small number of solid tumors, including cancers of the colon and rectum, stomach, breast and pancreas. Studies suggest that p300 mutations may also play a role in the development of some prostate cancers, and could help predict whether these tumors will increase in size or spread to other parts of the body. In cancer cells, p300 mutations prevent the gene from producing any functional protein. Without p300, cells cannot effectively restrain growth and division, which can allow cancerous tumors to form.

Another intriguing connection of p300 is that it can be inhibited by the action of the sirtuin deacetylase Sirt1. (See here.)

P300/CBP themselves are targets of intense research activity. Their physical structure has only very recently been determined. (See here, here, here.)

Finally (for now), it's interesting that p300 plays a role in stem cell signaling through one of our favorite signaling pathways – Wnt (see here). According to this report: Stem Cell Signaling Mystery Solved, a small molecule called IQ-1 interferes with Wnt signaling via p300:

What IQ-1 does, Kahn explains, is to block one arm of a cell-signaling pathway called the Wnt pathway, while enhancing the signal coming from the other arm of the Wnt pathway. The Wnt pathway is known to have dichotomous effects on stem cells i.e. both proliferative and differentiative. More specifically, IQ-1 blocks the coactivator p300 from interacting with the protein ß-catenin; this prevents the stem cells from being 'told' to differentiate into a more specific cell type.

Additional reading:

A core Klf circuitry regulates self-renewal of embryonic stem cells – research abstract published online 2/10/08

Molecular Alliance Identified that Sustains Embryonic Stem-Cell State – another summary of the Klf4 study

Tags: stem cells, cancer