Hi XiaRong,

To answer your question, we need to know:

1) is the image you showed from a confirmed stable clone?

2) if not, have you done an IF to determine what percentage of cells express your protein?

3) have you done a control transfection/selection with an empty vector to see if you see the same morphology?

I have done many stable transfections myself. I found that the cell morphology changes often do not correlate with whether your target protein is actually expressed. It could be due to where the vector is inserted into the genome or drug selection, because when I transfect with empty vectors and gone through drug selections I often see the same morphology change.

However, if you have confirmed the correlation of this morphology with cbl mutant expression, I agree with some of the previous comments that it appears to be a defect/change in cortical actin network that in turn affects cell-cell adhesion and/or cell-matrix adhesion, as well as cell morphology (lack of lamellipodia). If your mutant is dominant negative, it will not only inhibit normal cbl function but also may affect other cbl-like proteins, such as Hakai (CBL-Like 1) which is know to inhibit E-cadherin. NIH3T3 doesn't express E-cad but does express N-cad and P-cad (Reynolds et al. 1996). Moreover, the paper Nami posted have already shown that cbl mutants affect Rac and induce similar morphology change. It is also well know that Rac affects Rho and their balance affects cadherin and integrin mediated adhesion and cell motility. You may consider to go down this path to examine how your mutant affects cell adhesion, cell motility and RhoGTPases (Rac, Cdc42 and Rho).

Is the different morphology you are seeing just related to different cell densities? The shapes of the cells looks similiar in the photos you sent. Is this just not coming through online?

The mutant cells never reach the complete confluency, in fact they tend to form clonal clusters in the early passages.

Do have any other evidence of transformation for these cells in vitro (ie soft agar) or in vivo in mice? We work with a lot of different neuroblastoma cell lines (fully transformed cells) that form clonal clusters like the ones in your picture. I would put these cells in such assays to see if they truly are "transformed". Hope this helps.

Long time Ago I was working with 3T3 cells transformed by different oncogenes. These one look very similar to those transformed by Src. Did you check it? I also agree with the suggestion that full analysis regarding soft agar and xenograft are interesting

Indeed these cells look transformed, also by the fact that they form clonal clusters as you call them. We are currently working with an conditional transformation system (Src and Ras) in NIH3T3 (also checked/confirmed by soft agar growth), and upon induction those cells look identical with yours. Moreover, there are several reports about the role of Cbl in tumorigenesis, see e.g. very recently: Shiba N, Hasegawa D, Park MJ, Murata C, Sato-Otsubo A, Ogawa C, Manabe A, Arakawa H, Ogawa S, Hayashi Y. CBL mutation in chronic myelomonocytic leukemia

secondary to familial platelet disorder with propensity to develop acute myeloid

leukemia (FPD/AML). Blood. 2012 Mar 15;119(11):2612-4.

Here my impression: the original cells seem to have contact inhibition, which the transduced cells seem to have lost. They seem to be more transformed (more irregular criss-cross growth). To test this I would suggest to repeat in a cell line with robust contact inhibition, such as Rat-1. 3T3 is typically used for transformation assays and foci formation. There is a lot of literature out there. You could also use primary cells. I would expect that they might become immortalized. I have some 293-HEK cells showing such a phenomenon spontaneously when they were cultured for too long. In general, using cell lines for these type of experiments is maybe not the best way. I really would suggest primary cells. Cbl is also a proto-oncogene. Are you using wildtype or a mutant?

The CBL mutant transfected cells acquire a phenotype which resembles that of endothelial cells plated on Matrigel (extracellular matrix). In this condition endothelial cells form tubular-like structure similar to those in the picture. However many cell type may form such reticulates in the presence of the appropriate ECM. It seems a cell-ECM interaction process.

The observation that morphology change took place in all all cells simultaneously indicates that it is not a transformation. Even after a single oncogene overexpression or TSG deletion, the full transformation would still requires additional genetic alterations, which would be rare events and thus only showing up focally. Wouldn't in-vitro transformed 3T3 cells rather look like single foci of piling-up cells, on a background of an monolayer ?

I would rather suggest that your cells experience a down-regulation of focal-adhesion or other adhesion molecules or simply dedifferentiate.

I have the same phenotype with my cells that are GIST 882...they never reach the complete confluency and form clusters like your cells and they are tumoral cells!!

Hi Valeria,

Do you have genetic information of your GIST 882? Does it bear a oncogenic Cbl mutant or c-KIT mutant?

Yes,they have c-kit mutant.

Hi, c-cbl may interacts with cortactin and actin polymerization

See this article

http://jcs.biologists.org/content/113/2/215.full.pdf

I would rather say that it´s migration-inhibition, or contact-inhibition phenotype. did you check apoptosis-markers?

Hi,

I would say that they look very refractile, and grow, as Carlo suggests, like endothelial cells. I would maybe check their ploidy as well, as there are a few cells which look multi-nucleated from you image. I think that it is quite hard to suggest that they are transformed per se. Anchorage independent growth in soft agar and maybe matrigel invasion would be good experiments to do as classical markers of transformation. Did you plate the cells out at the same density? and have you done a simple growth curve because whilst the controls seem to have more cells, the cells do seem to be much larger, and so the clones maybe growing faster. If you do a simple growth curve and keep the experiment going on until well after confluence you could tell if they have differences in contact inhibition.

I assume that your Cbl is a dominant negative for E3 ligase activity, and should therefore expect that this will stimulate signaling via EGFR, and maybe other receptor tyrosine kinases. However, quite often the addition of high dose oncogenes whilst accelerating the cell cycle also promotes apoptosis, so Ulrike's suggestion would also be a good one to follow up.

You say that this is a stable clone, do you have other clones with different expression levels, and does expression correlate with the phenotype you see here?

Hi,

your cells transfected with the mutant resemble endothelial cells grown in matrigel.

Compared to the WT, your mutant cells could also be described to have a more mesenchymal phenotype.

Best,

Clara

The cells look very similar to when HeLa cells are treated with cytochalasin D...maybe affecting the actin ??

Hi XiaRong,

To answer your question, we need to know:

1) is the image you showed from a confirmed stable clone?

2) if not, have you done an IF to determine what percentage of cells express your protein?

3) have you done a control transfection/selection with an empty vector to see if you see the same morphology?

I have done many stable transfections myself. I found that the cell morphology changes often do not correlate with whether your target protein is actually expressed. It could be due to where the vector is inserted into the genome or drug selection, because when I transfect with empty vectors and gone through drug selections I often see the same morphology change.

However, if you have confirmed the correlation of this morphology with cbl mutant expression, I agree with some of the previous comments that it appears to be a defect/change in cortical actin network that in turn affects cell-cell adhesion and/or cell-matrix adhesion, as well as cell morphology (lack of lamellipodia). If your mutant is dominant negative, it will not only inhibit normal cbl function but also may affect other cbl-like proteins, such as Hakai (CBL-Like 1) which is know to inhibit E-cadherin. NIH3T3 doesn't express E-cad but does express N-cad and P-cad (Reynolds et al. 1996). Moreover, the paper Nami posted have already shown that cbl mutants affect Rac and induce similar morphology change. It is also well know that Rac affects Rho and their balance affects cadherin and integrin mediated adhesion and cell motility. You may consider to go down this path to examine how your mutant affects cell adhesion, cell motility and RhoGTPases (Rac, Cdc42 and Rho).

Looks to me like the MT-Cbl cells are simply growing slower. Clearly there is not the same number of cells in the two images. Does the wild-type look like this at a lower cell density? This would also explain why the transgenic cells eventually grow out of the phenotype... Are the exposures identical on the two images?