A combination of motility, resistance to Anoikis, gradients, position and physics (a growing tumor literally pushes) and also randomness. The problem is that we don´t know yet what is the relative contribution of each of these certainly. The different contributions, as expected, will vary by stage and type of tumor. I like the energy concept as well.

Scattering of tumor cells via MET-Signalling might contribute to the phenomenon.

Hi..

Tumor metastasis is a multistep process and its very difficult to approch all the new developments here but i can give you fast and breif overview..

we have long experience in studing tumor metastasis and it seems that some major pathways are responsible for the migratory ability of cells, however still kind of mess of networking signalling. the Rho proteins currently are of major concern and they play a vital role in the motality of cancer cells, besides tumor adhesion molecules can play a crucial role in such process such as VCAM, ICAM, and some proteases are also very important, up todate proteases are significantly considered especially MMP2, and MMP9 which can contribute to the metastasis party through ability of invasion.

There are few genes responsible for the cancer cells to be metastasized. I am working on S100A4 and S100P genes and both are affect the number of focal adhesion which maintains the cells attachment. We found that Benign tumour cells with (high number of focal adhesion) became malignant when transfected with S100A4 or S100P. Results showed that cells with A4 clones reduce the number focal adhesion comparing to un-transfected cells. Means, the transfected cells (lower no. of focal adhesion) have more chance to leave the original tumour to invade the neighbouring tissue and blood vessels. These genes affect cell’s cytoskeleton and assembly of focal adhesion subunits.

The epithelial–mesenchymal transition (EMT) and mesenchymal–epithelial transition (MET), are events that are crucial in process of metastasis of carcinomas. Epithelial cells that acquire fibroblast-like properties and have reduced cell-cell adhesion are the ones that are more motile and most likely to escape from the primary cancer and metastasize.

Tumor mass is like a living person, at a certain point for growing needs more energy and eats more. In a growing mass a competition for nutrients exist. At this point some of them try the way to migrate and to find new useful microenvironments. It is not excluded that hypoxic microenvironment in other parts of the body produce SDF1 and become an actractive microenvironment for cancer cells. I beleive that competition for nutrients inside the tumor microenvironment is a method of selection for cells able to survive elsewhere.

TGF beta signalling may be behind much of the migration itself. Whether that migration is "steered" towards vessels, or towards other targets, or away from the primary tumor by either homing or repulsion or both, or if it is a n example of a "random walk" followed by selection seems unclear.

I should agree with Amer Al-Haideri that proteases like MMP family and TIMP and many other which seems usually over expressed in tumor cells are one of the major events for metastasis, and these results of random mutations during tumor cells divisions as not all the tumors have this ability to metastasize.

Dear Colleagues,

Thanks Robert for this interesting subject.

To my knowledges, I think there is 2 processes that could be studied independently (even if phenomena are linked):

1) cells have to leave an organ or micro-environnement ,

2) then they have to migrate, even for non motile cells and so a step linked to movement that they have to acquire, for most of them..

For step 1

- in solid tumor, and even in "liquid" ones like hemopathies or lymphoma, abnormal cells have to leave the original tissue and thats the first difficulty for them because cells are linked together in an organ by mean of tight junction, that is now known to belong to adhesive protein-cadherin category ie,

-in hemopathie, normal cells are also linked to other cells of the micro-environnement by mean of other cell-cell interaction (selectine, integrine or Ig superfamily proteins).

For this subject that I better know, during normal differentiation of hematopoietic cells from blasts to erythrocytes, myelocytes ... there is a change in surface expression of adhesive proteins (also called CD for immunologists) and when the terminal differentiation step is reached, they can leave their microenvironnement to go in vessels.

During carcinogeneis, changing in gene expression, mutations, trigger an expression by cancer cell of abnormal adhesive protein compared to their normal phenotype and they are no more attached to their neighbour ; the same for hematopoietic cells where blasts expressed membrane proteins which allow them to quit medulla and to circulate in blood.

There is a relation between these membrane proteins and contact inhibition phenomena, that is abolished in cancer cells: the interaction between two cells can regulate gene expression, so division, by the FAK-kinase pathway (ie expression of BCL2 gene)

So, expression of oncogene from proto-oncogernes in cancer leads to proliferation, immortalisation but also changing in adhesive protein expression.

All that can be linked to step 2: acquisition of motility, even for non mobile cells in a solid tumor : some phenotype trigger motility also by the way of interactions like integrin-Fak kinase-cytosqueleton (activation of the actine pathway by proteins like moesine-ezrin).

Also MMPs act, as underlines by Amr Al-Haidari, expressed in cancer cells, but also by normal cells, ie leucocytes. Because when a cell is or become mobile, it has to go through tissues, vessel walls, so digest collagene-elastine-reticuline ....of tissues, but also different basal membranes or matrix; .and it's the role of different MMP (others names ie elastase, collagenase ...).

Finally, and perhaps a 3rd step: more strange is the secondary homing of some cancer cells (bone for breast cancer ie), that could be linked to an ancestral memory of their origin: ie during some hemopathie and medullar proliferation, metastatic cells, blasts, colonialized in general organs like liver or spleen, which have the same embryological origine of medullar tissue.

Regards

Didier

Dear Didier I agree with you and and your beatiful answer but cells need energy for doing all these things. So we come back to my question is the selective action the nutrient competition?

In my mind the most important goal behind the tumor metastasis is different from case to case. Some tumors are highly metastatic from their early stage and before tumor mass become large enough to induce nutrients depletion these tumors can migrate to distant organs through blood stream these tumor may have some mutation related to this property. But it should be pointed out that tumor microenvironment has a key role for metastasis. If lack of nutrients exist, tumor cells certainly try to find the better place for homing, in this condition most of tumor cells secret MMPs enzyme to digest the extra cellular matrix and migrate to other place and some others secret TGF-B for both angiogenesis and immune suppress. Microenvironment even can make force for benign tumors to find more mutations which these mutations can also trigger the pathways of metastasis.

Thanks to all of you, for your contributions! I am wondering why we (usually) do not see something like a benign tumor "metastasize". OK, this seems to contradict itself, since a metastatic tumor needs to be malignant - right? So, for let's say, hypothetically, for a "benign" mass of cells, not growing any fast or destructive for the surrounding tissue, the formation of metastatic cell clumps anywhere would probably called "ectopic cells". I wonder, if the metastatic potential of any stem cell is expected to exist, then I would suggest many more paradoxical tumors, first growing and metastasizing, but finally fully differentiating. But one of the examples of a malignant tumor which under rare circumstances can stop growing, ie. can differentiate, is neuroblastoma.

Or, are there other developmentally "ectopic" findings of cell clusters in mouse and human, which could use the highways to leave their original site and grow at distant sites (without being leukocytes).

Dear Gianfranco,

You are right. I have not well understood your point of view.

Your responded: "why ?" as I responded "How ?".

"Why ?" is more appropriate to the initial qestion asked "what?".

Your are right because, as cancer cells "work" for themself, have no more "altruism" to maintain a pluricellular organism like our, they had to use other ways to survive (they have no digestive tract to command ie). And it's a strange porcess because a solid tumor is composed of grouped identical cells, they stay first together, they have no contact inhbition,the tumor grows, several tumor cells die, that concur to necrosis of tumoral tissue. The growth also participates in hypoxia that trigger in part angiogenesis, a way for the tumor to have oxygen, but also divert body nutrients to its benefits. It's really a pression of selection, as you said, for the cancer cells to leave the intial tumor when the cancer cells see that they can die in the tumor if not (anthropomorphic ? it's like if humanity having no more food on earth would migrate on another planet). And we observe that in clinic: sometimes the physician discovers a mestastase of a tumor and is not able to visialize the original one (it has perhaps disappear because of necrosis or bad conditions of development).

Finally, as discussed in an another topic, patients suffering from cancer most often die from the metastatic disease, and after a growth of the tumor which surpasses ie 1kg of tumoral mass, by consumption of the reserve of the all organism.

The problem is that the tumor also die with the patient death: the reason for my term "altruism for normal cells is also for their own benefit". It could be compared to virus-bacteria and parasites in evolution: recent virus kill their host (HIV), old ones (influenza) not, as parasites or symbotic bacteria of our digestive tract.

Thanks

Didier

I don't know why tumors "leave the nest" but have a good idea why the go to the tissues they most likely metastasize to. Ocular melanomas originate in the choroid, which is arguably the most vascular tissue in the body: it consists almost exclusively of arteries, veins, and fenestrated capillaries - very much like the liver and spleen. Thus it's no surprised that ALL patients with metastatic ocular melanoma have involvement of their liver and spleen. Which is to say: when cancer cells metastasize, they seek that tissue which most closely resembles their "home base." That's the same reason so many Russians from Siberia who emigrate to America end up in Maine or Vermont!

John Gamel, MD

I met a researcher a few years ago who had very interesting data showing that certain tumor cells can phagocytose macrophages and monocytes and somehow acquire some of the monocyte attributes becoming a highly motile tumor cell. This cell could then respond to all the typical chemoattractants that monocytes and macrophages use. It also could be that the genes he looked for were simply turned on as opposed to being acquired, but non-the-less, it is a very intriguing possibility as the tumor expands and interacts with its environment. Never underestimate the cleverness of a tumor.

Another way of thinking is to use ecological concepts.Why a population decides to migrate and quit its territory? The reasons are multiple : overcrowding , lack of nutrients, lack of energy, fights and competition between subpopulations ( don't forget the very early intraheterogeneity of neoplastic cells and the "hostlity" of some stromal cells like macrophahes M1), etc etc. But , beyond these extrinsic factors, it also necessary to get the intrinsinc ability to migrate and survive during the trip.As always some are pionneers and other followers .It will depend on a complex and overall dynamic interactions between the different cellular and matricial actors.We wiil understand better when we shall get the invivo imaging tools allowing to track each cell of the primary tumor.Still a long way!

There is a multi-step biochemical processes in cells called epithelial to mesenchymal transition (EMT), which plays important role in embryo development, and is also considered the initial steps for a solid tumor cell to leave its mass and become metastatic. In this process, a tumor cell of epithelial profile loses its cell-cell conjuction and cell-memberane adhesion (down of E-cadherin, claudins, collagen, laminin, etc.) , and gains mobility (up of N-cadherin, vimentin, fibronectin, MMPs, etc). It is a complicated process involving many genes and regulating factors, and many stimuli can induce EMT including TGFbeta, Wnt signaling, interleukins, etc. And some reports stated hypoxia and oxidative stress can also induce EMT.

I think if we look at cell migration in normal tissue, we might get insights as to why and how tumoral cells start to migrate out of their microenvironment. For example, epithelial cells of the skin START to migrate towards each other when the integrity of the epithelium is disturbed in a wound. Upon closure of the wound, the cells STOP their movement and stay intact. Most likely, these cells receive the start and the stop signals from their microenvironment or simply they react to loss of contact inhibition.

Now, regarding tumoral cells we can postulate several hypotheses that explains why/how epithelial cells adapt their migratory phenotype: 1) tumoral cells obtain a constant start signal (e.g., activating mutations in EGFR, K-Ras…); 2) tumoral cells lose their ability to sense the stop signal through genetic losses so these cells will start migration blindly and will never stop; or 3) the tumor microenvironment creates a condition similar to wound microenvironment.

I think that different pathological states in which the hypoxia and the decrease of nutrients are present (eg Stroke, Myocardial infarction, wound healing, inflammation) produce the same cytokines (eg VEGF, SDF-1 and so on). These hypoxic situations recruit immature myeloid cells and stem cells etc. What differentiates them is that in the tumor the hypoxic state/ process is never closed and how many years ago Dvorak suggested " that the tumor is a wound that never heals". At this point probably, environmental conditions useful for the formation of metastases are created, with all the features suggested by Didier Jambou and Fariboz Mortazavi.

The communication between the tumor cells and the surrounding cells, the microenvironment; helps drive the process of tumor progression. So going from normal to benign, benign to malignant, malignant to metastatic is driven not just by what's happening inside the tumor cell itself but by what's happening around it.

The microenvironment can exert profound epigenetic effects on cells through cell-derived interactions between cells, or through cell-derived factors deposited into the microenvironment.

Probably this aspect is true Michael

Dear Gianfranco I truly believe that this loss of communication between cells its key for metastasis to occur. I also believe that the root of this problem is the loss of oxidative phosphorylation due to damage to the mitochondria in the cancer cell.

To me its a combination of various factors:

1. Lack of nutrients, hypoxia in the existing tumor induces alterations like HIF-1 stabilization that causes the cells to migrate.

2. Chemokine and cytokines secreted by the tumor cells and the tumor micro-environment leads to EMT.

3. The stem cell theory, not all cells but the stem cells in the tumor mass modulate or retrieve their migrating potential and migrate to a distant location.

4. Loss of p53 a multifunctional protein as the tumorigenesis progresses can also contribute to migratory potential as a few of p53 transcriptional targets are anti-angiogenic and anti-metastatic genes.

As an engineer, I don’t believe in metastasis. From my new observations and deductions, human beings have many capillary gates (sphincters, pericytes, etc.) which are closed when we move, and they are not fully opened when we are not moving, or resting (for unknown reasons, perhaps, to protect us from something). Therefore, if we do not have the right mix of exercises, we’ll have many cells in different locations starving of blood and nutrients (but not cutoff completely) day and night. This situation is more apparent when we observed normal brain developments of children and adolescence in past generations (The new on-line game generation is very different). Even without external causes to damage cells, the starved cells in multiple locations may grow mutated cells simultaneously at different speed, and form multiple cancers (one at a time perhaps) years later. I doubt that we can make them move easily without making a mess because they can be all over the place in small quantities. We may have to wait for them to die, or use external forces to eliminate them if they start to growth day and night when they know how to build their own blood vessels (without capillary gates).

The right mix of exercises can be found in my thesis “Taichi healing & Pao’s Law of Exercise”. This is to use traditional taichi (much slower than the modern taichi) to trick the brain that you are not moving and then to elevate heart rate for 30 minutes to send more blood to brain and other organs/tissues efficiently. Of course, one can do other sports to elevate heart rate. Once the HR is up, one has to stop moving, and then wait for stress hormones to go away and CNS to detect all the signals to open up all capillaries. Repeat the whole process (before heart rate returns to resting state) many times to accumulate healing time. It may take 1 to 3 hours a day to complete this start-stop business.

Thesis Taichi Healing & Pao's Law of Exercise:

@ Thamir Ismail

your approach with S100A4 and S100P is really addressing the crucial part of my question. This is fascinating. I will read some literature.

@ Kurt Gehlsen

I remember that since decades scientists try to speculate on "fusion" of tumor cells with leukocytes to explain metastatic behaviour similar to leukocytes, but you now add a slightly different point of phygocytosis by the tumor cells. Experimental fusion of leukocytes with tumor cells (as far as I remember from 20 years ago) could lead to metastasis in animal experiments, but then the chromosomes and genes would be shared by the fused cells. If a tumor cell eats a leukocyte like a monocyte/macro´phage without using the genes, but for example some of the proteins and membranes, that -theoretically- could and would change the migration potential of such a cell. It sounds a little bit like science fiction, but I do not know how to fully exclude such a hypothesis for all types of tumors. But intersting.

Hello Robert, I'm uncertain that tumor cells actually leave their microenvironment; I'm also unsure that the "tools for metastasis" exist and can be gained. Rather, the tumor cells seem to affect other cells in the microenvironment (or take advantage of characteristics these other cells naturally possess) - making these tumor-affected cells "obligate cells" which might perform other functions the tumor requires (even as simple as passing along a "message"). I suppose in a way I agree with Pao - I don't believe in direct metastasis. You are indeed wise to express skepticism about this very complex process. Please see http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2756342/

from the Albert Einstein Institute for further information. You'll also find there a very brief passage from Alice in Wonderland by Lewis Carroll. I thought it very entertaining and you may, as well.

Hi There, Alan, thank you for your comment and the link to the interesting review with the entertaining comparison. In contrast to you and Pao, I have not any doubt that tumor cells can leave their microenvironment - this is in my view clearly documented histologically, genetically and experimentally - and all of my published or unpublished studies support that and the last hundred years of tumorpathology. My question here is, at which stage this first leaving of tumor cells from their original location occurs. I dedicated my scientific interests to cancer research when I started medical school decades ago, but in Germany there was no support for real scientists in medicine at that time but without 'connections' or family in these areas.

In the 1990s I changed from boring molecular genetics of p53 mutations in half of all cancers to work only on my own ideas, trying to find the solution for organ-specific metastasis. It was clear to me that cell adhesion receptors as in leukocytes could direct metastasizing cells - my first slides always used the words: mail-code. Trying to define the metastatic-stem-cell in the late 1990s at Stanford, I thought there could be similar mechanisms for integrin activation within milliseconds after the right chemokine would bind to the right receptor, but I didn't get my grant and Germany didn't pay me another year at the Weissman-Lab, unfortunately. Meanwhile the chemokines appear to play a major role, but not exactly as I expected. It seems to be another mechanism, but still unclear. The other site of the migrating tumor cells is: how do they know to leave... Some of the aspects mentioned in these discussions here were new to me, thanks to all of you!

Any more thoughts? I think it is important to look for tumor stem cells, but when we can not kill them, we may just tell them not to move or migrate. Therefore, any signal which really causes them or influences them may be helpful. And, surprisingly, the opposite may help as well: getting the tumor cells out of their microenvironment via chemokines or anything else into a less friendly environment, perhaps the blood stream, may make them more accessible to chemo- and/or radiotherapy (or less protected by their microenvironmet).

I agree with you , Robert.

The hypothesis of no exitence for a "metastatic" state of tumor is, in theory, possible; but there is some arguments that , in this case, cannot be explained. In general, tumor and that is called its metastasis have rather the same histological origin. Perhaps, the differentation of metastatic cells is often lower than the initial tumor, but it could be due to the changes in envirornment the tumor meets after metatstasis, or simply becau!se a laps of time passed that could increase the number of mutations, increasing by this way "aggressivity" of the cells. If we go back to the precedent hypothesis (why not? certainly original), and if I well understand the fact that tumor and "metastasis" would be only diffrent tumors devolopped at the same time, why would they be the same as cytological origin in different points of the body ?

To my mind, by the most probable hypothesis, metastasis phenomena and new homing of cells is linked to changes, whatever the reasons, in the relation between cancer cells and their environment (loss of contact inhibition , changes in membrane adhesive protein expression).

And about homing to a possible same embryological origin of cancer cells and the new localisation, I could remember to the specialist a very strange phenomena obdserved in some terratoma: cancer developped from seminal cells can lead to the formation of a tumor comprising different tissues like hairs, tooth, skin ... as if this special tumor "wanted" to constitute a hole new organism like during embryogenesis.

Regards

Didier

I once developed a rat tumor model from neuronal precursor cells. The tumor cells showed a striking differentiation potential into astrocytic but also neuronal lineages - and all the typical signs of human medulloblastoma or PNETs (primitive neuroectodermal tumors), like formation of neuroblastic rosettes and a migratory potential. Nowadays specialists of the rather rare medulloblastomas (compared to all cancers in human) invest hundreds of millions in Germany for better understanding the subgroups. But of course it should be no discussion about the facts on metastasis - and I don't want to go into this discussion about questioning metastasis. It is also not surprising, that teratomas can differentiate into all different tissues, this just shows and supports the idea of cancer stem cells. I am only interested in applying real science into cancer research.

Article A model for primitive neuroectodermal tumors in transgenic n...

Stem cells tend to be more fragile... I would lean towards your answer #2. I would also edit #3 or add a #4: What if the tumor is shedding cells all along and the cells that find a new suitable environment are the seedlings for micromets? There is an emergent technology that effectively measures cancer cells in vivo circulation. I know that there are studies underway comparing these counts between patients to discern a connection with metastasis. Going back to Gerard Perret's answer, think of a tumor as a herd and the micromets as their explorers.

Thanks, Felix, for your comment. I think (Tumor) stem cells can be very resistent to chemo- and radiotherapy, that would explain initial shrinking of a tumor mass After therapy, but recurrence of the tumor. If a stem cell is not dividing often, it may not be that fragile.

Depends on the stem cell pool unfortunately. I should qualify what I meant by stem cells (even cancer stem cells) being more fragile. What I meant is that they may be weaker against normal body processes that would destroy them if they were in circulation or the wrong place. Differentiated cancer cells will display mature markers that immune surveillance may detect as normal. However, the cancer stem cell may not display markers and this may allow those cells a degree of immune-stealth. Clearly a complicated question but a really nice discussion!

Robert...

In addition to malignant cells, solid tumours comprise supporting stromal tissue that consists of Extra Cellular Matrix (ECM), connective tissue cells, inflammatory cells and blood vessels. The stromal compartment and the malignant cells together shape the tumour microenvironment that in turn determines tumour progression and efficacy of anti-tumour treatments. It is now recognized that the host microenvironment undergoes extensive change during the evolution and progression of cancer. This involves the generation of Tumour-Associated Fibroblasts (TAFs), which, through release of growth factors and cytokines, lead to enhanced angiogenesis, increased tumour growth and invasion.

Cancer stem cells. yes probably are source of metastases but may play more role in tumor relapse. To my opinion the only way to kill them is using immunity or modify the immunomicroenvironment of the tumor. Unfortunately a lot of stem cells are present in the blood so we must change the tumor microenvironment or be able to inject innate immune cells (like allogeneic Natural killers) associated to chemotherapy.

Felix, I still do not see why a stem cell should be more fragile. In some patients, for example a breast cancer can come back even decades after the orignal therapy; that supports the idea the tumor (stem) cells were somehow dormant - but the least fragile cells of the cancer.

Coming back to my original question, I guess the two possibilities exist for many tumors: some tumor stem cells already have all the tools to metastasize, some others will gain those tools. So, future pathologists will have to check, if this really is the case for all tumor entities - and how one can identify such postulated subgroups for every (current) tumor entity. This then will help to decide which therapy might be the best in an individualized therapy.

Dear Michele,

To my knowledges, VEGF is secretd physiologically by the pathway of HIF1alpha transcriptional factor, whose principal "sensor" is O2 pressure. In hypoxic conditions, VEGF is increase and it triggers (on a general way) the formation of new vessels. During wound healing , you can observe it but also in pathologies (abnormal vascular proliferation in retina during diabetes, also compensation of artery obstructions by vascular collateral circulation during arteritis).

During metastasis, that's a good way for tumors to increase vascular formation to their benefit, for oxygenation and also nutriment providing, and during hypoxia inside the tumor, this mechanism is stimulated. Mutations occuring during cancer transformation could (or can, because I don't know if it has been demonstrated) transactivate gene of VEGF.

Regards

Didier

Dear Didier and Michele in other disesae states VEGF is overexpressed and probably in hypoxic areas (where there is inflammation VEGF is overexpressed) alaways present in a disease states may be the adapt niche for metastatic cells. Example brain, many patients have TIA (transitory ischemic Attacks) these areas may be favorauble to metastatization, another example is zone of arthosic pain (back pain) these areas too are favourable to bone metastatization. I hope my opinions understandable. Metastasis go where may find nutrition and situation favorauble to their growth.

Evidence to the contrary, vascular normalization does not increase tumor growth rate or increase metastasis.

Heterozygous Deficiency of PHD2 Restores Tumor Oxygenation and Inhibits Metastasis via Endothelial Normalization. Mazzone M., et al., Cell 136, 839–851, March 6, 2009.

Exercise modulation of the host-tumor interaction in an orthotopic model of

murine prostate cancer. Jones L.W., et al., J Appl Physiol 113: 263–272, 2012.

Silencing or Fueling Metastasis with VEGF Inhibitors: Antiangiogenesis Revisited. Longes S., et al., Cancer Cell 15, March 3, 2009.

Antiangiogenic Therapy Elicits Malignant Progression of Tumors to Increased Local Invasion and Distant Metastasis. Paez-Ribes M., et al., Cancer Cell 15, 220–231, March 3, 2009.

When metastasis start, cells inside the organe recive information to go outside of the organe, and cells go to anther organ. As a stem cell, it can arrive and can start to continiou its crzy maraton to obtain more cells to defens the organims or to change cells destroyed yet. But something gnes bad, because the new cell never stop its division. The mecnisms to stop is damage perhaps

I agree with Ken.

It could seem contradictory, but tumor cells used first a natural mechanism fogued by them to allow their growth, but for all cells in general, increase oxygenation is toxic by ROS generation. ROS behing neutralized, if in excess, by different intracllular SODs.

So, we cannot apply to cancer cells physiological "logic" , because these natural mechanisms are "studied" to maintain a "coherent multicellular organism", that's no more the goal of cancer tissue.

Regards

Didier

Dear Ken and Didier I agree with you but I think nature gets/use the simplest and less expensive biologic mechanisms. I think also that at moment we do not understand the process in its totality so it seems complicated and not understandable but I am sure the ability to metastatize or the necessity to leave a place for another one is simply because the initial environment was not sufficient to support the growth of all the cells. Our ability is to discover which of the factors well described by many of these excellent researchers is more important. Saying this in other words is like for tumor immunity, some Keys factors have been understood (e.g. CTL4, PD1)

With your question, I am thinking now. Let see which kind of stupid things I can tell you. First metastases is produced when cells decide to go outside to the organ and go to another site. Why is doing that. Probably something is going something in the blood to attract the cells. Why someone is doing that, because someone need this cells. Why, for example liver attracts cancer cells to it. Probably because liver want to resolve some problem, it want to obtain cells from the body, similar to it to produce new differentiated cells to start or continues its work, detoxification of the body. Which kind of cells are attracting?, cells similar to it, ectodermic, endodermic or mesodermic cells. That is truth because normally the metastasis of a tumour occur in an organ similar in evolution, and in ontogeny, to other. When metastasis cells arrive there. Who kind of things need to do? Need to start to differentiation and start to work with the director of the work tell it, the liver. So the cells do not do it and start proliferation and dividing in stand to be a good worker and only work, work, work. So or the force of the boss is not enough or the worker, cancer cell, do not want to work. So probably both or one of the two systems are not working well. In the evolution, some cells, sexual sex control all other cells to direct cells to a strange situation and new situation, it do not need to divide any more. And as you know, it is really terrible for an organism, only we want to divide for the future, it will be here in the future. So the control of that was been so specials. Now only people with this bad function is death, because if anyone work and all start to be a cell alone the body can not exist any more. If we find and resolve ALL THESE question, we will be with sufficient forces to destroy and to eliminate our older problem. CANCER

An undiscussed point in all these very stimulating questions and answers is the moment at which metastasis begins.It could shed light on the problem of causality.I think there are two very different situations: 1-The very early metastasis ( even , in some cases at the in situ stage). In this case the number of cells involved is very small, the intravasation is processed cells by cells and the peripheral sites are made only of disseminated tumor cells.2-The metastasis at the invasive stage.The number of cells involved is much higher and intravasation results mainly from collective migration ( with leaders and followers).The peripheral sites ( prepared as preniche) are made of micrometastases.It's only in this case that the driving forces are also ( in addition to intrinsic ones) extrinsic ( overcrowding, lack of nutrients , hypoxia, acidosis,chaotic vasculature ,etc). On contrary, the driving force of the first wave is mainly intrinsic ( genetic/epigenetic acquistion of a mesenchymal/stem phenotype).What is not known is the dynamic collaboration between this two successive waves ( we can also add the possibility of round trips between the primary tumor and the metastases : self seeding hypothesis).This collaboration could affect the colonization process ( improving or destabilizing the niche) and the fate of metastatic deposits ( dormancy or outgrowth ).This late point is crucial to understand for a clinician.The contribution of the two routes of dissemination ( hematogenous vs lymphatic) at the different steps of the metastatic spread ( early vs late ) is also unknown. Lymphatic vessels are not functionnal inside the tumor but are numerous and dilated around the tumor, mainly when the tumor is advanced.

Once again, in vivo imaging, genomic single cell analysis and the use of concepts from ecology and social sciences ( leadership, diaspora for example) are the keys to make progressing our understanting.

Dr. Gerard: You talk of early and late waves of metastases. There are papers saying that gene signatures of metastases of certain solid tumors, suggest that many of these cells were already present in the primary tumor. This is puzzling. Is this evidence for the first (early wave) of mets that you mention? What other evidence suggests the existence of these 2 waves?

There are so many typical and common cancers with known routes of metastasis, but also some rare cancers; at the end there is a lot of heterogeneity. I am not a certified pathologist, but I could imagine (but could be wrong!) that tumor entities with a very early sign of metastatic spread, maybe melanomas, could be closer to the actively migrating stem cell in contrast to maybe less or very late metastasizing tumor types. But if there is not such a typical bahaviour in such tumor types, maybe one could try to find more "typical" behaviour in such tumor types, i.e. fast or slowly metastatic - and with interesting exceptions of those - and use especially the exceptions for a different (hopefully better) treatment.

My question really wants to stress the very first cell of a tumor moving away from the tumor and, at the same time, having the capability to survive and divide, i.e. defining the "metastatic" stem cell I was looking for in the 1990s.

Dear Venil

You ll find a brillant demonstration of this first wave in :

Rhim AD, Mirek ET, Aiello NM, Maitra A, Bailey JM, McAllister F, et al. EMT and Dissemination Precede Pancreatic Tumor Formation. Cell. 2012; 148(1-2): 349-61.

Thanks for the Ref Dr. Gerard.

Very interesting and contentious topic. Cannot hope to give a more comprehensive answer than already given.

However, it should be said that the phenomenon of anoikis is worth considering, before actual migration and seeding. Before moving away from its primary locus, the cell should survive detachment, which is a process tied to many core cellular mechanism, which may already be perverted in cancer.

Thanks, Gerard, for pointing to the Cell paper with a model for early metastasis, much earlier as most would have expected. This supports the view of cancer stem cells with all the tools of migrating away from the primary tumor.

I tend to be in the camp that believes the potential for metastasis is an intrinsic characteristic of tumoral cells although the tendency for metastasis seems to be rather variable among tumors.

For example pancreatic cancer (tumor type in the Cell paper) is an exception in the sense of early metastasis as majority of pancreatic cancer patients will show distant metastasis even after surgical resection. In contrast, other tumor types (e.g., breast and colon cancer) are not as aggressive. If all tumor types would have dispatched metastases very early on, there would be no cure following surgical resection of the primary lesion in any tumor type.

Dear robert and fred

I agree that pancreas might be a special case.For stage 1and 2 breast and colon cancers , the fact that surgical resection is generally curative does not exclude an early metastasis but could be the consequence of a strong local stimulus inducing dormancy. This could mean that if stemness is a prerequisite for dissemination it is not enough for outgrowth in the new site. An other point to consider is the observation that in some instances the surgical resection leads to a flare up of metastatic depots.

Dormancy appears to be a really paradoxical phenomenon. Some breast cancer patients can live several decades after removal of the primary tumor, but then develop distant metastases with same histology. Of course, if we could understand how to trigger a dormancy switch in migrating tumor cells, that would be interesting. It is hard to understand how a migrating tumor cell can be aggressive enough to leave its primary tumor environment and survive for decades somewhere else.

One possible explanation is cell plasticity and adaptation to a new environment.Even agressiveness could be reversible.

I think aggressiveness is dictated in part by proteolytic cascades. For example ADAM and MMP family members regulate many invasive properties of cells. They chew through ,matrix proteins, basement membrane components, etc. But also, they regulate chemokine, cytokine, growth factor, and receptor, release from cells. The proteins also regulate each other. ADAM members cleave other ADAM members from the cell surface. And MMPs cleave ADAMs and ADAMs cleave MMPs. Since ADAM members also contain distegrin domains, they bind to integrins on the cell surface of other proteins, that in turn mediates activation of the cells, and as you can imagine would regulate migratory propery of cells.

Given all this, at least for ADAM members, they are regulated by their micro environment. Even changes to membranes such as high cholesterol, for example, will affect ADAM activity. This phenomenon is interesting because ADAM10 activity is protective for Alzheimer's disease, but cholesterol can affect its activity. And there is a causal link between high cholesterol and Alzheimer's, for example.

So even if the proteinases are dysregulated in cancer cells, and could promote tumor formation, they may be regulated or stay dormant because of their micro environment.

That is why chronic inflammation, promotes cancer. It is all about altering the finely controlled balance of proteolytic activities. In inflammation, when TACE/ADAM17 activity goes up, TACE also cleaves EGF family members, EGF receptors, and chemokines ( membrane bound ones), to name just a few. TACE also cleaves membrane proteinases that regulate their activities. And not only is it the cleavage event, but most of the membrane proteins that are released leave behind c- terminal fragments, that end up going to the nucleus to affect transcriptional regulation.

I forgot to mention about ADAM10. ADAM10 is necessary for NOTCH and WNT signalling. These signaling pathways are necessary for epithelial to mesechymal transitions. Also, TGF beta signalling is affected by ADAM family members such as TACE and ADAM12. So regulation of the ADAM family members is what likely controls cancer cell development and invasiveness.

In terms of the different types of cancer and how they behave, it is at least in part what proteinases and signalling pathways are dysregulated in the cells. An ADAM10 inhibitor works great to inhibit proliferation in breast cancer cell lines, but does virtually nothing in ovarian cancer cell lines.

Thank you Marcia for these data confirming that agressiveness is a trade off between intrinsic capacities and adaptation to microenvironment

Cancer cells with damaged mitochondria are very hard to eliminate, and will continue to accrue mutations which make the cancer more invasive, hypermetabolic, and adaptable to metastatic environments.

Non-tumorigenic epithelial cells shed into the vasculature are typically destroyed by shear stress or macrophages and necrose, but circulating tumor cells have an altered cytoskeleton and microtentacles that protect from deformation caused by shear stress. Interestingly to me, both carcinogenic and non carcinogenic circulating breast cells have microtentacles but those in tumorigenic cells are unusually long and motile compared to those in non-tumorigenic cells (Cancer Res. 68, 5678-88, 2008; Cancer Res. 70, 8127-37, 2010). Microtentacles, which are formed by microtubules and also may contain intermediate filaments, are required for efficient attachment and thought to act in the metastatic spread of cancer. Interestingly the common chemotherapeutic paclitaxel, which acts by stabilizing microtubules, strengthens the microtentacles resulting in increased attachment of circulating tumor cells to secondary sites (Breast Cancer Res Treat. 121, 65-78 (2010) so this type of one highly successful therapy for primary tumors may set the stage for a greatly increased risk of metastasis. The observation that ~90% of all cancer deaths are caused by metastatic spread of the primary tumor underscores the importance of this question about metastasis as well as the need to understand circulating tumor cells and how they find favorable attachment sites and what their mechanisms of reattachment may be.

If we believe that cancer process is a defense wound healing process then we can explain the tendency of cancer cells to metastasis. Due to accumulative signals to the healing cell to repair the sick area, the cells may go for mad and acquired new characteristic due to different genetic changes. One of these characteristics is the potential to leave the primary site to go far seeking for a solution to this problem or wrongly to deal other tissues to cure the affected area.

To my knowledge based on a proposed model in physics of cancer, metastasis denotes the final stage for most of cancers and onset of proliferation state, where the unhealthy cell profileration rate becomes much greater than the cell removal rate when the immunity system totally fails to demolish those cells as if it resembles to command suicide (halt alert) allowing the geometrical cell growth without any defense. However, there is at least one exception: Hotchkin disease. There are many evidences of success after chemo-treatment that may immunity system struggle for survival even at final stage.Therefore, I believe due to the inherent nature of cancerous cell may be categorized as aggressive and adaptive cells.

Simply, I think its better to say a evolutionary process in terms of starvation which drive the metastasis (cancer cell) to search the new area for nourishment or to minimize the metabolic load at one place therefore cancer cells shows some extraordinary advance signalling and metastasis which makes them immortal.

Thanks, John Tainer, for the two citations on microtentacles - I will need to read more on this. Many years ago, the direct measurements of some cells made real problems with long tentacles, so the normal capacities of an AFM was not sufficient due to some long tentacles; perhaps those are the same used in metastasis.

I believe that metastatic ability of CSCs arises from properties within the microenvironment of the tumor; essentially CSCs would gain the tools for cancer metastasis.

Potentially chemotaxic factors would cause the tumor to leave its microenvironment; as for the specific factors, not sure.

Mitochondrial oxidative stress could actively promote tumor progression and increase the metastatic potential of cancer cells.

http://breast-cancer-research.com/content/pdf/bcr1988.pdf

http://www.medicinabiomolecular.com.br/biblioteca/pdfs/Cancer/ca-2896.pdf

http://www.diagnosticimaging.com/printpdf/167553

Not only chemokine receptors rather other factors such as calmodulin ie involvement of Ca-ioins, including inflammatory cytokines involvements helps in creating microenvironment to spread or mirgrate to other parts.

John Tainer's comments are very interesting. I wonder what contribution the alter lumen of tumor vessels (Mazzone Cell 136, 839–851, March 6, 2009) has on adherence of CTCs to the endothelium. Additionally, the shear stress in a tumor vessels should be more oscillatory ( Circ Res. 2003;93:1225-1232 and Journal of Internal Medicine 2006; 259: 351–363), which down regulates NO production, contributing to endothelial dysfunction - a vicious positive feedback loop.

I'm agree with most of your opinions, but I strongly suggest the role of immune system in defense against cancer cells as professor Parviz Parvin mentioned. It's that cancer cell that modify itself to the degree that can not be recognized by immune cells, may be total loss of receptors to immune cells especially when cancer cells are of high grade type, explaining the early metastatic potential of high grade cancer cells than the low grade one.

The reason that cancer cells do that is: EVOLUTION!

Believe me, they evolve by themselves, that's the problem.

But Paula, evolution is based on random mutations and selection. Tumor cells do not contribute to germ line transmission. WHAT then would be the advantage for the organism? I only could think of more food for the survivors in a group of ancient humans, that would support early death to allow the genes to spread faster around the world... but I think that has nothing to do with any mechanisms tumor cells may use to leave their original niche.

Dear Robert, that is the thing. Cancer cells are mutating randomly, and the set of mutations respond to "environmental" pressures, such as the immune system that makes the selective force, among other pressures added by exposure to carcinogens, chronic inflammation and such. There is not advantage for the individual but for the cancer cell itself, like a different entity. I'm not talking about an advantageous strategy for humanity, just for the cancer cell by itself. You see, that is my way of explaining how one of my own cells may become against me, only if it starts behaving as something else evolving in me. Like selfish DNA kind of thing. That may be too extravagant an idea, but it makes sense to me... And gives me the chills...

The most dangerous attribute of cancer cells is their ability to metastasize. Throughout the process of metastasis, tumor cells interact with other tumor cells, host cells and extracellular molecules.metastasis suppressors allow growth in orthotopic sites but prohibit growth at ectopic sites highlights their central role in how tumor cells interact with the myriad microenvironments encountered during the metastatic cascade. Indeed, the evidence presented in this minireview describe functions for a subset of metastasis suppressors and how each suppressor impacts interactions between tumor cells, matrices and other cell types. What emerges is a picture in which tumor cell expression of specific genes regulate success or failure at each step of the metastatic cascade.

When tumor cells express molecules on their cell surfaces (e.g., E-cadherin) which promote cell–cell adhesion or molecules that prohibit their motility and invasion (e.g., Nm23, TIMPs, SseCKS), they are more apt to remain at the primary tumor. Likewise, molecules that affect recruitment and structural aspects of angiogenic vessels could impact tumor cell dissemination. Still other suppressors reduce cell survival during transit (e.g., caspase-8, BRMS1, KAI1). We have not even mentioned the multitude of immunogenic molecules that make circulating tumor cells visible to the immune system.

Perhaps most importantly, a growing number of metastasis suppressors, highlighted by KISS1, MKK4, p38 and MKK7 in this review, control the last step of the metastatic process, colonization. When tumor cells arrive at various secondary sites, they find themselves in a milieu of growth promoting and growth inhibitory factors that are different from their site of origin. Depending upon how the cells respond to competing signals, they will either die, remain quiescent or proliferate. It is only when they proliferate that they will become a bona fide metastases. We speculate that a subset of metastasis suppressors will eventually determine in which organs metastases develop. Those molecules would be responsible for the predilection for metastasis observed in many tumor types (e.g., bone metastases from breast and prostate carcinomas).

... And what makes the tumor cell leave its original location?

Good question :-)

"What make" in term of mechanism (cellular, molecular, genetic) and also "Why ,", and in this case, it depends if we are determinist or probabilist-stochastic..

"What ?" is perhaps all that we have precedently discussed. The simple change in surface adhesive protein phenotype could explain the leaving of original environment, change due to dysregulation of adhesin gene expression: by hasard or because allowing the tumoral cells to move (when tumor of a solid tissue): in this case first cadehrin of cellular junctions (leading to loose of contact inhibition), then integrins to "stick" to the basal membranes, in particular of the vessels. That could be comprehensive because interaction between adhesins and their ligand trigger intracellular pathways like FAK- Kinase, so link between receptors and FAK kinase, and FAK kinase act in parallel on gene of proliferation and differentiation (ie BCL2). Also, ghey also act on regulation of MMP gen (MMP degradating the matrix) ....

After that"why ?": perhaps local "pression of selection" because there is a decrease of nutriment for tumoral cells when tumor grows, and if we are finalist (but also that's an anthropomorphic reasonning), tumoral cells "want" to move in a more favoorable environment. And we could suppose that they "choose" an organ for metastasis corresponding to a common embryogenic origin (personal advice).

A lot of things that could be supposed and to be demonstrated.

Regards

Didier

the fertilized egg and the developing embryo at some point stay at the correct site. Many cells in the adult body never really migrate. There are so many tumor types differing in their genetic variation and biological behaviour, but usually are capable to leave their original tumor site. This question only focuses on the first moving and migrating of the very first tumor cell(s) in a developing tumor. And, what is the influence of potential repellents (from surrounding and tumor cells)?

Well Robert.

So, in this case the final point of metastasis would be "out of the discussion" this c. But to my mnd, it's also interesting if we look at the ectodermic, mesodermic, zndodermic first 3 tissues in embryo. Ectoderm giving in particulal skin and neurodermic cell ... and so on.

And if we look at the "specificity" of the destination of metastatic cells, we could say first they go in organ well vascularized (liver), but if we look at the example of the brain, it is not directly on the circulation, and some tumor give more brain metastatis than others (so my think because skin and brain are from the same initial tissue) ; and another case is the bone metastasis majoritary issue form breast, lung, prostate, kidney tumors. The reason why I suppose interesting to look at the origin of cells in embryogenesis.

Regards

Didier

Didier

Fascinating thought! So neoplastic cells kill the organism by using the same mechanisms at work to create life. Death and birth are the same by essence!

Ironic ? or really interested :-)

In fact, embryogenesis is a particular state of the body, where a lot of gene are expressed, then repressed in final organism, butin some physiological processes in our body (menstruation, wound healing …), some inducible gene of the same type are temporary exprerssed, but also , you have some part of what happens during some steps of cancerogenesis: inhibition of apoptosis and cell survival; activation of proliferation genes (often kinases) and some cells go out the G0 phase of cell cycle ito go in steps of division; changing in some secretory phenotype like this of adhesive proteins.

The difference is that in wound healing, the phenomena stops, and genes temporary activated to build a new tissue to replace the injured one are repressed again.

So for me, cancer correspond to physiological processes "lead astry" by some "egoist" cells (anthropomorphism)

And finally, yes ! Birth and death are the same by essence, a continuum for the leaving organisms; and a philosopher said ( one who quoted probably a sentence attributed to a "legendary" ? person of history: Hermes trisgemist): "Nothing is lost, nothing is created, all is transforming".

Regards

Didier

Cancer has so many facettes, many of them have been investigated heavily in basic sciences and even in clinical settings. My original question here should refer to what usually is NOT addressed to come up with something totally new... so, what makes the first migrating tumor cell moving...

Well Robert.

We'll try to be "imaginative" :-) , because some important discoveries in the past came from "dreams" of humans. But are really dreams or an intuitive "a priori" comprehension " of universe ?

Regards

Didier

Didier, Robert

Not ironic at all but really interested.

Your reference to Lavoisier is interesting because it points to the big absent from this discussion : the thermodynamics of life. The book of erwin Schrodinger , What's life ? (1948) argue that's life is a temporary exception to the second principle of thermodynamics which settles that the entropy of a close system must increase. Therefore life creates negentropy. From this point of view could we consider that metastasis is the begining of the end of the second principle's violation? Therefore what's make the first migrating tumor cell moving is the return to normal thermodynamics.

I agree Gerard.

A living organism, well organised to survive and reproduce itself (especially pluricellular ones), "fight" during all its life against a natural increase of entropy in a closed system, and perhaps death is the end of this fight which conduct to the "dispersion" of the organim; and the transformation of its compound and restitution in nature of the basic compounds (molecules). So, the sentence "nothing ...".

Why I said that a tumoral cell is an "egoist" one (even if ti's an anthropomorphic and finalistic term) in a pluricellular organims: that's because it "decides" not to continue to work for this fight for life, and try to function for its own profit, while it belongs to a "coordinated" pluricellular organism. But ithat's not a "proof of intelligence" because in this case, the death of the organism , which was became for the tumor only "an host", leads to its own death.

We could compared this phenomena to the "behaviour " of other organisms like parasites, bacteria and virus. and in this case there is a difference between our "symbiotic " bacteria: in mitochondria, digestive tract, and pathogens: parasite are well adapted, to humans, in general don't kill their host; also the case for some "old" bacteria in evolution ; and also "ancient" virus (like virus of cold).

By contrast, the more recently adapted-to-human pathogens (HIV, Ebola virus ie ) kill rapidly their hosts and/or are at the origin of pandemia (criteria of their "virulence").

Regards

Didier

PS To complete your comment, Gerard, Yes "normal thermodynamics" is "against life", since its origin. Perhaps the reason why it took so much time to appear on earth (3 billions years for first unicellular organisms), and we does know exactly the real possibility of life in other parts of universe.

Excuse me gentlemen I kindly disagree about the concept of life being some kind of exception to the second principle of thermodynamics. It would seem true whenever we take the cell or living system just by itself, but if we consider the whole planet, living things generate more entropy to the surroundings and that would compensate the order inside them. Regards, Paula.

Yes Paula.

if we consider that after "creation" of life, "utility" of living organims is to be active, and this activity and interaction wiht other living organisms "generates entropy".

For me "entropy" , described by physicits, is strange to understand. If the principles can only be appied in a "closed" system, as said by Gerard, evidently only each individual organism can be consider as a closed system, but also debatable because in relation with its external environment (nutrition, excretion, reproduction...). So, how can be apply entropy principles at each level of the system: we have to defined different systems being some "closed" ones, but in relation as russian dolls. Can thermodynamic laws be applied in this case at the level of universe ? , becasue we have to suppose that it is a "closed" system, but isn't it ?

But it could be the subject of a new topic, because now far from the question of Robert.

And perhaps, it's difficult to applied entropy to living organisms, because principles have been described first for "non living systems".

Regards

Didier

Paula and didier

The only closed system is universe ( if we agree there is only one and don't believe the chords theory). The equivalence for negentropy is information. Therefore the exchange of information between living systems could also contradict the second principle of thermodynamics by creating an ordered system, for example an ecological niche.These considerations are not so far from rober't's question if we consider that tumorigenesis changed the local system of biological information and that neoplastic cells disseminate in order to find a more adapted environment with more negentropy.Unfortunately by doing so they increase the entropy of the whole body by modifying the physiological messages between the different part of the body.If true the medical fight against metastasis might be holistic ( I recommend you to read the last dougla Hannahan's paper in Lancet (383,p562) on the concept of battlespace in the war on cancer)

I agree Gerard; and your reference is very interesting; and this is also discussed in an another RG topic of focused on a physic (physicists) approach of cancer.

But also, to my mind and about the subject of systems and their entropy, we could consider that cancer and/or metastic cells , leaving the coordinate organisation of the body to live their own life , are at the origin of a new independent system, with an independent " entropy " (in fact, they have not the same rythm of division), and without the same "goal" of life in a coordinate organims, as they work for their own benefit.

Do they establish communications (cancer cell together) for a "common goal" whatever it is ? I don't know. But they conserve also a "memory" of their "normal" functionning (always to my mind) : I am fascinated by some solid tumors, like teratoma, which are able to build in a "closed tumor" a sketch of embryo or organs (you can find skin, hairs …), as if tumoral cells started a program for a new body in the boby.

But perhaps it's because of their cellular origin (germinal cells): even de-differentiated, they conserve their capacity to build a body compared to other solid tumors.

Regards

Didier

Thanks, Gérard, for pointing me to Doug Hanahans paper in Lancet (it is available at Lancet after registration). I once was strongly recommended by Irv Weissman to join Doug Hanahans lab, at that time at UCSF, but unvortunately, I had insisted in trying to finish my "fascinating and publishable results" at a later sabotaging lab environment at a VA facility (not part of Irvs lab). But back to the question here: I do not consider any useful clues coming from thermodynamics or discussed questionable exceptions. My question considers to start a perhaps underestimated, very first step of a tumor cell, cancer stem cell, metastasizing cell or similar. Then one may use the resulting knowledge to include it in Doug Hanahans view of his review on war on cancer and the ten different possible battlefields to fight on cancer in clinical settings.

Please excuse me, all, but we seem to be straying from the question asked. I recently had occasion to repeat the original question Robert poses above, and was told to read the nice synopsis of recent work in cancer cell migration given at http://www.sciencedaily.com/releases/2013/11/131121111845.htm, which is produced regularly by Science Daily - that particular release is dated late 2013. I recommend it to all as a source of possible explanations for cancer cell migration as well as a synopsis of recent and current work in the area.

I guess the answer also depends on the tissue origin of the cancer. Some cancer cells, e.g. melanoma, have a higher migratory capacity based on the normal cells from which they are derived - nicely described in: Nat Genet. 2005 Oct;37(10):1047-54,PMID: 16142232.

@Alan Holden, thank you, this seems to be one aspect of how a single cell at the tip of a cluster can be activated (by FGF) and influence the other cells to follow...

@ Jürgen Becker, thank you for the interesting citation and the model with human melanoma. Melanoma appears to represent some of the early metastasizing tumor types; I understand that neural crest derived melanoma cells may re-activate their mobility program much easier as other tumor types since melanocytes travel a lot during ontogenesis.

This may be out of topic, but I remember some 1950/60s paper in Nature on melanocytes trafficking into the spleen of some rat strain and forming black spots... I think homing to the spleen of these cells is not yet solved...

I used B16 (mouse) melanoma cells since the late 1990s to compare the integrin VLA-4 to explain hematogenic metastasis: VLA4 on such B16 cells strongly binds to VCAM1 on some endothelial monolayers, e.g. under low shear - supporting the idea of a mechanistic homing effect, explaining their rolling and sticking under blood flow conditions (my experiments). This appeared to be in contrast to Irv Weissman's paper in Cell (mid 1990s) where his lab used B16 variants with no (or low) VLA4 expression. Those cells showed an increased metastatic potential which was attributed to the lack of homotypic aggregation between B16 cells - and easier leaving of the primary tumor.

My different experiments support the idea that VLA4 can be used by B16 melanoma cells to bind to some endothelial cells expressing VCAM1... Two sites of the virtual paradoxon.

Yes Robert,

I am always interesting in adhesive proteins and their relation with intracellular signalling. I suppose that they are very important in metastatic process, and all of them. Phenomena called "contact inhibition" for normal cells , lost in cancer cells, can be linked to cadherins, involved in tight junctions between cells in tissues. Others adhesins, like VLA4 you quoted, with their ligand VCAM-ICAM are in general expressed naturally in some mobile cells (leucocytes, lymphocytes), and not in cells in organs. So, expression in cancer cells with its gene dysregulation which could favour cell mobility, and passage through the vessel wall, between endothelial cells. In parrallel, the link between integrins and superfamily of Ig proteins triggers inside the cell some pathways like focal adhesion kinases (FAKs) , controlling cell proliferation and division. But FAKs also activate expression of MMPs genes , whose related proteins can digest basal membranes … and so on.

Finally, all mechanisms, when dysregulated, which can confer to cancer cells some of their characteristics.

Regards

Didier

Post-scriptum to your comment:

"This appeared to be in contrast to Irv Weissman's paper in Cell (mid 1990s) where his lab used B16 variants with no (or low) VLA4 expression. Those cells showed an increased metastatic potential which was attributed to the lack of homotypic aggregation between B16 cells - and easier leaving of the primary tumor".

OK ; not necessary in contradiction: VLA4 linkage to ICAM1 (vessel) or ELAM1 (leucocytes) favours mobility and passage through the vessel, essential for metastasis, and in the other case, abnormal protein mutants having decrease capability to "stick" to others cells, so more mobile , is a second condition for metastasis.

Didier

I think that in tumors cancer cells get under stress due to competition for oxygen and nutrients and they just adapt to become more mobile when fighting for their survival. Normal cells have signals that function well to suppress such behavior.