Hi Steingrimur,

To give a better answer, it may be helpful if you would clarify what exactly you mean by "positive markers" and "subjective measurements of upregulation...". However, I will attempt to answer now and we can discuss our thoughts more once I think I truly understand your question.

First, the concept of EMT is not originally from the field of cancer, but rather from the wound healing process particularly cutaneous tissues, but the phenomena appears to much more broadly applicable. In this process, part of the final stages of wound resolution is the migration of healthy adjacent epithelia to recover and or close the wound site. In this setting, it was observed that, largely under the influence of TGF-beta (although this is admittedly highly oversimplified), the healthy epithelial cells undergo a series of changes in which they lose typical epithelial features including growth in a cohesive layer, apical-basal polarity, and polygonal shape to take on the characteristics of mesenchymal cells which allow them to better complete the healing process. These changes are accompanied by loss of expression of typical epithelial markers, notably E-cadherin, and gain in expression of N-cadherin (along with many others including the transcription factors that bring about these changes).

Additionally, when I think about EMT I do not really consider the process of EMT to be differentiation per se. I say this because the process of differentiation is thought to be a unidirectional process (That being said there is some evidence that terminally differentiated colonic epithelium has the ability to de-differentiate to regenerate the entire heterogeneous population of colonic epithelial cells including progenitors under conditions of extreme stress/ injury). Because of this EMT is a transient and reversible shift in phenotype driven by underlying cell signaling transcriptional programing. Thus the process seems more akin to the activation of a T-cell rather than the differentiation of a T-cell from a common lymphoid progenitor.

Because of this distinction, it may be better or more clear to say that cells that have undergone EMT are simply more mesenchymal in appearance and gene expression. That is to say that EMT is not purely binary, but rather a spectrum of epithelial-like to mesenchymal-like. If you consider EMT as a spectrum rather than two binary states, then it is easier to see why its hard to say specifically where a single cell is at on that spectrum if continuous levels of gene expression changes and continuous measure of cell migration are discounted. However, the problem is exacerbated by the fact that many of these studies use large populations of cells rather than single cells. Thus, there is a great deal of heterogeneity across the whole population at many points in the process of cells undergoing EMT (From ability to respond to the given stimulus to the arrangement/ modification of the chromatin containing the genes required to exhibit the mesenchymal phenotype and probably many more than the community is aware of). It is, however, interesting to consider that if EMT studies were carried out at the single cell level (perhaps through high quality flow staining or multifaceted gene reporter systems) then the process may indeed appear more binary allowing the generation of a more concrete definition of EMT at least under those experimental conditions.

As for the role of EMT in metastasis I will point you to a paper by Fisher KR from Vivek Mittal's group at Cornell (PMID: 26560033). In this paper they use a reporter system to longitudinally trace cells that had undergone EMT (at least one flavor of it). Using this set up, they show that cells that had successfully undergone the process of metastasis had for the most part not undergone an EMT. This is interesting work, but overall, I think that EMT has so many possible facets and molecules that can drive a cell further to the right on the epithelial mesenchymal spectrum that many more studies with a similar set up that look at different molecules will be required to demonstrate to what extent EMT is required for metastasis and how far to the right on the spectrum a cell needs to be before it acquires the ability to metastasize. Despite the shortcomings, I think that you may find the paper interesting given your skepticism of EMT and its role in cancer.

Finally, I will pose my own question for you or anyone else that is interested. In the paper mentioned above, they show that EMT confers resistance to chemotherapies. Additionally, it has been shown that cells that have under gone some extent of EMT divide slower. Finally, EMT is classically thought to be a major factor in the invasion metastasis cascade and adhesion independent growth/survival (Labelle M, Begum S, and Hynes RO PMID:22094253 among many more). All of these are key features of both EMT as well as cancer stem cells. My question is to what extent are cells that have undergone EMT and cancer stem cells different populations or are they actually highly overlapping cells populations or even in the most extreme interpretation different names for the same phenomenon (albeit with a different set of markers)? And if the later is true does the fact that the markers are different warrant a different name?

I hope this helps or at least inspires some thoughts and questions. Sorry for the book.

Best,

Andrew

Hi Steingrimur,

To give a better answer, it may be helpful if you would clarify what exactly you mean by "positive markers" and "subjective measurements of upregulation...". However, I will attempt to answer now and we can discuss our thoughts more once I think I truly understand your question.

First, the concept of EMT is not originally from the field of cancer, but rather from the wound healing process particularly cutaneous tissues, but the phenomena appears to much more broadly applicable. In this process, part of the final stages of wound resolution is the migration of healthy adjacent epithelia to recover and or close the wound site. In this setting, it was observed that, largely under the influence of TGF-beta (although this is admittedly highly oversimplified), the healthy epithelial cells undergo a series of changes in which they lose typical epithelial features including growth in a cohesive layer, apical-basal polarity, and polygonal shape to take on the characteristics of mesenchymal cells which allow them to better complete the healing process. These changes are accompanied by loss of expression of typical epithelial markers, notably E-cadherin, and gain in expression of N-cadherin (along with many others including the transcription factors that bring about these changes).

Additionally, when I think about EMT I do not really consider the process of EMT to be differentiation per se. I say this because the process of differentiation is thought to be a unidirectional process (That being said there is some evidence that terminally differentiated colonic epithelium has the ability to de-differentiate to regenerate the entire heterogeneous population of colonic epithelial cells including progenitors under conditions of extreme stress/ injury). Because of this EMT is a transient and reversible shift in phenotype driven by underlying cell signaling transcriptional programing. Thus the process seems more akin to the activation of a T-cell rather than the differentiation of a T-cell from a common lymphoid progenitor.

Because of this distinction, it may be better or more clear to say that cells that have undergone EMT are simply more mesenchymal in appearance and gene expression. That is to say that EMT is not purely binary, but rather a spectrum of epithelial-like to mesenchymal-like. If you consider EMT as a spectrum rather than two binary states, then it is easier to see why its hard to say specifically where a single cell is at on that spectrum if continuous levels of gene expression changes and continuous measure of cell migration are discounted. However, the problem is exacerbated by the fact that many of these studies use large populations of cells rather than single cells. Thus, there is a great deal of heterogeneity across the whole population at many points in the process of cells undergoing EMT (From ability to respond to the given stimulus to the arrangement/ modification of the chromatin containing the genes required to exhibit the mesenchymal phenotype and probably many more than the community is aware of). It is, however, interesting to consider that if EMT studies were carried out at the single cell level (perhaps through high quality flow staining or multifaceted gene reporter systems) then the process may indeed appear more binary allowing the generation of a more concrete definition of EMT at least under those experimental conditions.

As for the role of EMT in metastasis I will point you to a paper by Fisher KR from Vivek Mittal's group at Cornell (PMID: 26560033). In this paper they use a reporter system to longitudinally trace cells that had undergone EMT (at least one flavor of it). Using this set up, they show that cells that had successfully undergone the process of metastasis had for the most part not undergone an EMT. This is interesting work, but overall, I think that EMT has so many possible facets and molecules that can drive a cell further to the right on the epithelial mesenchymal spectrum that many more studies with a similar set up that look at different molecules will be required to demonstrate to what extent EMT is required for metastasis and how far to the right on the spectrum a cell needs to be before it acquires the ability to metastasize. Despite the shortcomings, I think that you may find the paper interesting given your skepticism of EMT and its role in cancer.

Finally, I will pose my own question for you or anyone else that is interested. In the paper mentioned above, they show that EMT confers resistance to chemotherapies. Additionally, it has been shown that cells that have under gone some extent of EMT divide slower. Finally, EMT is classically thought to be a major factor in the invasion metastasis cascade and adhesion independent growth/survival (Labelle M, Begum S, and Hynes RO PMID:22094253 among many more). All of these are key features of both EMT as well as cancer stem cells. My question is to what extent are cells that have undergone EMT and cancer stem cells different populations or are they actually highly overlapping cells populations or even in the most extreme interpretation different names for the same phenomenon (albeit with a different set of markers)? And if the later is true does the fact that the markers are different warrant a different name?

I hope this helps or at least inspires some thoughts and questions. Sorry for the book.

Best,

Andrew

Yes, it is real specially in some chronic fibrogenic liver diaereses and cancer.

Hi, I do agree that the EMT concept can prove useful as a heuristic model, but I am incredulous that cancer cells can flip flop between differentiated states.

In vivo cancer cells are very heterogeneous and cant be assumed to have identical phenotypes. That could help explain the different staining patterns.

As for resistance to chemotherapies, look at castration resitant prostate cancer. Is it because prostate cancer cells undergo EMT and become resistant to chemotherapy or is it a Darwinian process that favors the growth of resistant cell populations within the tumor(s)?

Article The Evolutionary History of Lethal Metastatic Prostate Cancer

Hi Mónica Suárez Korsnes I think comparing cancer epithelial cells to cancer stem cells is being very generous.

The bottom line is that there is no one universal marker available that says that a cell is definitely EMT and not a screwed up epithelial cell. Cancer epithelial cells are very heterogeneous.

As for chemoresitance, is it EMT or is it Darwin?

Hi Steingrimur Stefansson

I read a few lines about EMT, and I have a question which may look like humor and far from feasibility.

is there any chance for nuclei partial transition for this phenomenon?

please take a look at this article (section Results: Epithelial to Mesenchymal Transition Directs Structural Changes in Prostatic Cancer Cell Nuclei Detectable at the Population Level in vivo)

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4270929/

Hi Pejman, I dont quite understand this study. The researchers injected mice with 2 populations of PC3 previously described as being mesenchymal-like and epithelial-like based on morphology and no positive EMT marker.

Then they were able to isolate these different PC3 populations in vivo based on nuclear structure that was not explained in PLoS One. 2013; 8(10): e76773 ?

Dr, Steingrimur Stefansson

Yes, they used the Mann-Whitney U-test on nuclear biometric data, however, they have proved the method before. But their protocol is so different because of PC3-EMT procedure. Through co-culture, they discovered a new generation which has markers of interest. if I'm not mistaken, mostly in EMT-cells studies cells isolated from blood flow (Likely it does work out by co-culture).

Article Transcription Factors OVOL1 and OVOL2 Induce the Mesenchymal...

I'm a bit incredulous of EMT studies that use circular logic and ad hoc definitions.

"We isolated a subpopulation of cells...…. that showed a stable epithelial phenotype in culture and designated them PC3-Epi. These cells were selected based on two criteria: bioluminescent intensity and epithelial cell morphology.

Consistent with the epithelial state, PC3-Epi cells maintained high E-cad, low Vimentin, and undetectable expression of the EMT-activator ZEB1".

If "immotile" epithelial cells suddenly become "motile" there have to be positive marker(s) on the cell surface, like integrins, that are valuable in biopsies.

Also the EMT idea does not explain how epithelial cells become mesenchymal cells, migrate to distal sites and then turn back to epithelial cells.

Metastatic prostate cancer has been shown to be predominantly due to cell populations in pancreatic cancer that are selected for with testosterone antagonist. These cells are termed "castration resistant", develop and metastasize in nearly all chemotherapy for prostate cancer.

(22) (PDF) Transcription Factors OVOL1 and OVOL2 Induce the Mesenchymal to Epithelial Transition in Human Cancer. Available from: https://www.researchgate.net/publication/257757283_Transcription_Factors_OVOL1_and_OVOL2_Induce_the_Mesenchymal_to_Epithelial_Transition_in_Human_Cancer [accessed Jan 26 2019].

Hi Mónica Suárez Korsnes I appreciate the link. This might be semantics, but I think this article has the argument upside-down.

"Cancer cell plasticity is defined as the capacity of a cell to adapt to dynamic changes, triggered by external signals. Cellular plasticity plays a critical role in cancer cell survival, invasion, and metastasis formation as well as in tumor heterogeneity and in the development of therapy resistance."

I beg to differ: Tumor Cell Heterogeneity Is THE Reason For Their Plasticity.

Cancer cells in solid tumor masses are not a clonal population. They are heterogeneous both phenotypically and genotypically. Plus there are normal cells that are coopted to feed the tumor mass.

If these cancer cells were clonal then solid tumor cancers would have been cured decades ago. And there will be no cure forthcoming as long as researchers think that there is a single "magic bullet" that will target all solid tumors.

As for the EMT debate, I am wondering if some researchers have forgotten some tenets of Darwin's Evolution theory.

Populations are heterogenous.

Descendance with modifications.

Survival of the fittest.

This fits perfectly with the emergence of antibiotic resistant bacteria. This also fits perfectly with castration resistant prostrate cancer. Resistant populations find a niche to thrive.

Bacteria are heterogeneous. Solid tumor cells are heterogeneous.

You don't need some theoretical EMT hand waving that can never be substantiated because this EMT has no universally accepted and verifiable positive histological marker(s).

Hi Steingrimur Stefansson

Thank you for your valuable comments.

I read several papers and am with you about vague aspects of EMT and even crossing whole theory. There is no strong backup for EMT only a few evidence like miR-200 or interleukin-6,etc. partially, I think there are some contributors which have been ignored and they work beforehand, just picture something like generating iPS cells.

a few bystander cells(epithelial) turn into advance (for example with help of mesenchymal stem cells by absorbing one of its exosomes ) it is my theory.

sincerely yours,

this article is shown that cell fusion between mancchymal stem cells and cancer cells is not too far from reality.

Article Enhanced metastatic capacity of breast cancer cells after in...