Dr. Robert Weinberg lecture at the BC Cancer Agency

A LONG long time ago, I wrote a review on Dr. Weinberg’s textbook, The biology of cancer, which actually drew a response from Dr. Weinberg himself. Needless to say, I’ve been a big fan of Dr. Weinberg’s since reading his textbook, and I’ve learned to keep my opinions to myself – mostly.

In any case, I couldn’t resist the opportunity to see him in person, now that he’s visiting the BC Cancer Agency. Although, I arrived at the lecture theater 15 minutes early, and wasn’t at all surprised to discover that the hall is nearly full – there’s a full overflow room as well.

The lecture was organized by GRASSPODS, the Bennet family and several other sources -none of whom, by the way, are sponsoring me, so I don’t feel obliged to list them all.

I’ll revisit this later to add some reasonable formatting.


* Lab: involved in mechanisms leading to tumours. Outline of how primary tumours begin.
* 90% of cancer associated mortality is associated with metastasis, not primary tumours.
* Discussion of tumour progression – can take two tumours, introduce identical mutations. and in one case you get metastasis, in another, no metastasis. Tumourigenicity is also vastly different. Suggest the involvement of stem cells.
The nature of the normal cell of origin is a strong determinant of the phenotype of the tumorigenic cell.

Cascade model of stem cells -> intermediate ->  post-mitotic differentiated cells.  Most cells give up the stem cell properties, and can no longer seed new tumors.

Invasion -metastasis cascade:

Primary Tumor -> invasion -> intravasation -> Trasposition -> Extravations -> micrometastasis -> metastasis.

Most cancers don’t make it all the way to metastasis.  Thus, there are serious challenges for tumour cells to get all the way to the end of the cascade.

So, what changes must occur to metastasize?  several Images: human xenograft, stained.  Human epithelial cells show EMT transformation preferentially on outside of human xenograft, at interface with mouse cells.  Thus, EMT transformation seems probable as a metastasis intermediate.

microenvironment of the primary tumor can contribute importantly to the phenotypic conversion occuring during an EMT…..  (missed the end)

Thus, we probably can’t find out EVERYTHING from just sequencing the tumour – we need to know about the environment of the cells.  eg, tumour structure specific changes.  There is also more to an EMT transformation than just shape – it’s a profound shift in expression, possibly up to 1000 genes.

Much of what we know about EMT comes from transcription factor research.  eg, gene Twist, which reprograms cells involved in EMT.

Cancer cells appropriate early embryogenic programs in order to acquire traits of high-grade malignancy.  (slide heading)

(EMT also occurs in non-cancer as well, such as wound healing. )

Image: metastasis are 85% reduced in one experiment by using siRNA-Twist.  We don’t know if Twist is sufficient for activating EMT, however.

How many of the steps of the invation-metastasis cascade can the twist transcription factors program? (header)

Colonization is unlikely to be activated by twist, but it changes how we see the metastatic process.  It simplifies things however, as turning on twist, you can probably get metastasis by simply turning on twist (or equivalent).

Anecdote: 15 year adenoma -> cancer, 1 year cancer -> metastasis.  Very suggestive, anyhow.

Gupta’s work:  melanocite -> tumor state, (SV40, ER, hTERT and Ras). behaviour was “mild” Put into non melanocite cells caused many other changes.

Some discussion of Slug protein, which enables emigration from priimitive neural crest – also seems to be activated in metastatic cells.

transcription factors that enable neural crest emigration are likely also involved.

member of Weinberg’s lab (Mani?) asked about the nature of cells induced by an EMT: Epithelial cells -> (EMT) -> Fibroblast ??

Tested it, and found that you don’t get a standard fibroblast.  Then what DOES it produce?  Found FOXC2 in cell nuclei of normal human mammary epithelial.  Proves nothing, but suggests: Epithelial cells -> EMT -> Mesenchymal cells -> ??? -> stem cells.

In fact, tried FACS, (CD44, CD24) and showed migration from non-stem cells to putative stem-cell groupings.

Induction of EMT by Snail and Twist EMT-inducing TFs also generates CD44hi CD24lo cells. Similar experiment done, and tied with mRNA expression showing epitheial markers drops dramatically, and mesenchymal markers dramatically rises (N_cad, Vim FN1, FOXC2, Slug, SIP1, Twist).  Thus, epithelial cells transformed this way are going through the “Front Door” of the pathway towards stem cell characteristics.

This is also shown by morphology: mamosphere forming abilities.

Kornelia Polyak looked at primary human breast samples, and demonstrated that also seems to happen in cancers. Marked overexpression of mesenchymal markers.

Stem cell program of EMT cells, is the same (we believe) as breast cancer, etc.  EMT is likely highly “dangerous” for cancer patients.  It allows transportation to new locations AND likely self-renewal!

Can one produce more compelling biological profs that EMTs generate epithelial stem cells?

Mouse model was used by Wenjun Guo.  (using cleared mammary stromal fat pads.)  Slug was found to be highly upregulated in stem cell population. [leaving out much detail here about the experimental design, which was very elegant.]

Performed competitive reconstitution assay – used Slug vector.  Used controls without Slug vector.  Before induction, they’re the same:   Slug induction: able to recreate complete ductal tree as though completely stem-cell-like.  Controls had some behaviour, but ductal tree recreated is small, and non-complete. [again leaving out lots of evidence presented]

Thus, it seems that Slug CAN be used to induce a stem-cell behaviour in a cell population.  It’s possible, but not shown, that this could be used for generating complete organs.  [neat]

Slug is necessary AND sufficient for breast cell Stem Cell transformation.

Most currently used chemotherapeutics kill non-CSCs (cancer stem cells) preferentially.  Thus, as long as the CSC populations are present, recurrance is likely, if not probable.  We need to target the CSC population instead, thus, were need new drugs, etc.

Force epithelial cancer cells to undergo an EMT and acquire stem-cell traits… used this to screen for drugs that preferentially kill stem cells, rather than non-stem cells. For two common drugs (doxorubicin and Paclitaxel) CSCs required 2x dose to kill the CSCs relative to non-CSC.

Screened, and came up with two drugs, both of which are anti-paracitics.  Salinomycin (used in chicken feed!)   (Gupta and Onder)

Example of cell populations tested via FACS.  original 4.9% are stem cell -> Paclitaxel takes you to 70% stem cell state (by killing non-CSCs), whereas salinomycin takes you to 0.2% (by killing CSCs).  This compound isn’t a good drug for cancer, but it may lead to new drugs that could be used.

C.Chaffer did an experiment that shows that floating cells (flopc) in media aren’t all dead, and some of them adhere to suface if re-cultured – there appear to find small sub-population of stem-cell-like.  Followed over time, showed that non-stem-cells spontanously go into stem-cell state and vice versa!

There is a rate of conversion, depending on the cell.

Can not be due to rapidly growing populations of stem cells because they grow SLOWER than non-stem-cells.

Thus, there is a spontaneous transformation from the non-stem-cells to stem-like states.

Thus, the cascade figure is wrong!  The arrow must go both ways – FROM and TO the stem-cell state.  Also, must be used to re-think the cancer treatment on CSC based treatments – because you can’t just kill the CSC once, as they’ll regenerate!  Thus, you must hit both populations – both the non-CSC and the CSC at the same time.

Also have to revisit the cascade…  CSCs can go through non-CSC stages!  [kinda mind blowing, actually]  Thus, not all mutations must happen in the stem cells!  Mutations happen more rapidly in the more rapidly growing cells, which are not stem-cells, but that later BECOME stem cells.

[Great Talk!  Tons to think about, but I’ll just post it now and revisit it later for cleanup.]

4 thoughts on “Dr. Robert Weinberg lecture at the BC Cancer Agency

  1. Heh – I knew it would be standing-room only! I passed so more trainees could go. (Also so I could get some work done in some extremely rare peace and quiet while everyone else was in the lecture theatre!)

    • Yeah – by the end, the aisles were full and there were people crammed like sardines in the wings.

      But hey, even if you missed the talk, he still walked by your desk. I’m jealous.

  2. I really hate being hemmed in by crowds of people who are blocking my exits (a wee phobia, you might call it). So my own private desk walk-past is as close as I’m going to get to greatness today!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.