Dear Annette,

for the first question:

Fibronectin is NOT a chemoattractant, its a matrix protein to which cells adhere. So when people said to use fibronectin they might have meant coating your trans-well inserts in fibronectin before seeding the cells. and 10% FBS ought to migrate ur cells..it is in fact used as a positive control for trans-well migration assays. So coat your filter inserts ON BOTH SIDES with fibronectin before seeding your cells. Just keep the insert in a fibronectin solution in such a way that the filter is completely immersed in the solution at 37 degrees from 30 min-1 hour (atleast) just before seeding ur cells. after seeding, put the well in the lower chamber containing 10%FBS, and incubate. What happens here essentially is, when u seed your cells on a fibronectin coated filter, they attach to them by integrin and fibronectin interaction. when you put the insert in 10% FBS, the cells start migrating from the upper surface of the filter to the lower, but still stay attached to the fibronectin on the lower surface of the filter. While imaging, you just have to bring the lower surface of the filter into focus thats it.

The concentration of the fibronectin solution and the incubation time have to be standardized for your cell type.

For the second question:

You can try using the fibronectin from BD. I think thats cheaper than sigma..and for the kind, that depends on the origin of the cells you are using. i.e, if you are using human cells, you should use human fibronectin and likewise.

Hope this was helpful.

Dear Annette,

for the first question:

Fibronectin is NOT a chemoattractant, its a matrix protein to which cells adhere. So when people said to use fibronectin they might have meant coating your trans-well inserts in fibronectin before seeding the cells. and 10% FBS ought to migrate ur cells..it is in fact used as a positive control for trans-well migration assays. So coat your filter inserts ON BOTH SIDES with fibronectin before seeding your cells. Just keep the insert in a fibronectin solution in such a way that the filter is completely immersed in the solution at 37 degrees from 30 min-1 hour (atleast) just before seeding ur cells. after seeding, put the well in the lower chamber containing 10%FBS, and incubate. What happens here essentially is, when u seed your cells on a fibronectin coated filter, they attach to them by integrin and fibronectin interaction. when you put the insert in 10% FBS, the cells start migrating from the upper surface of the filter to the lower, but still stay attached to the fibronectin on the lower surface of the filter. While imaging, you just have to bring the lower surface of the filter into focus thats it.

The concentration of the fibronectin solution and the incubation time have to be standardized for your cell type.

For the second question:

You can try using the fibronectin from BD. I think thats cheaper than sigma..and for the kind, that depends on the origin of the cells you are using. i.e, if you are using human cells, you should use human fibronectin and likewise.

Hope this was helpful.

Yes, it was! Thank you!

Using it as a positive control is what they meant, yes. But I didn't register that both sides of the insert needed to be coated. Makes sense. I think I've tried everything else, except I'm also going to re-isolate the cells from a primary tumor. It seems like my cells simply aren't migrating, but it works for other people. Thanks again!

Annette

Although, now that I think about it, I seem to remember thinking that fibronectin had chemoattractive activity in addition to being for adherence. No? Thanks

Well I know fibronectin only as an matrix component..never came across its function as a chemo-attractant.... or maybe I am just confused between the terms chemo-attractant and cytokine...There is an RGD (Arginine, Glycine, Aspartic acid) domain in fibronectin that does attract and interact with the integrins on the cell surface...but I am definitely sure it does NOT act as a cytokine.

There is published work on fibronectin and fibronectin-derived peptides as chemoattractants. Just google "fibronectin as chemoattractant". Fibrinogen and fibrinogen-derived peptides are also chemoattractants. In both cases the work has been done mainly with human fibroblast chemoattraction. No idea if they would work for

mouse melanoma cells. I would survey the literature thoroughly first, rather than asking for casual opinions.

Did you ever thought about using NIH/3T3 conditioned medium (FCS-free!!! supernantant) as chemoattractant for your assay? That's very common.

For how long do you allow your cells to migrate? It can vary from one up to 12 hours. It depends on your cell line.

Always remove the remaining cells in the chamber with a Q-Tip after fixing the cells and again after staining. This will help you to focus and count the right cells later on.

Let me tell you one thing Annette, we had the same problem with B16F10 melanoma cells which is another derivative of B16 cells. In spite of them being very aggressive in vivo, we could not see any migration or invasion of these cells in vitro. This was odd to us and we even used different matrices such as fibrinectin, laminin, vitronectin, etc. as well, but did not make a great difference... The solution for us was (as Janina suggested above) using "conditioned medium (CM)" and this made the B16 cell migrate and invade like crazy!! I definitely suggest use of CM (either from NIH/3T3 cells or even a tumour cell line). This way you may not even need fibronectin although using it doesn't hurt.

Give it a try, it should work....

Thanks to you all, this has been very helpful!

I see the same thing, these cells are aggressive in vivo but in vitro there are no major changes (I should have mentioned that I'm testing their ability to migrate in the presence of TCDD). Of the 1 x 10^5 cells I seeded in the insert, 263 is the highest number of migrated cells I ever counted. Also, most of the migrated cells were around the edges of the insert, with very few in the middle. Has anyone else had this problem? Subtle vibrations in the incubator, maybe? This is after I let them migrate for 24 hours toward FBS after the cells were serum starved. In the literature using these cells, the count is usually somewhere between 300 and 500 using 8um pore size--that seems pretty typical. Some used fibronectin-coated inserts, others didn't. The strange thing is that when I swab the inserts, the end of the Q-tip is very black because there are so many un-migrated cells. They just aren't moving.

I am definitely going to look into using conditioned medium, as well as fibronectin. thanks for the suggestions!

Dear Annette,

why do you use fibronectin for coating trans-well filters? If you want to test in vitro invasiveness of B16 melanoma cells, you need to coat upper surface of transwell filters with a reconstituted basement membrane such as the growth-factor free-Matrigel (50 micrograms per filter, www.bdbiosciences.com BD-MATRIGEL). Then you can test invasiveness of your cells by adding, in the upper and in the lower compartment of the transwell units, serum-free medium or serum-free medium supplemented with the substance that you want to test.

If you add the substance that you want to test for inducing or inhibiting invasiveness, you need to add the substance both in the upper and in the lower compartment, because if you add it only in the lower compartment you will test its chemoattractant effect and not its effect on invasiveness.

I tried this method with B16 melanoma cells and others neoplastic cell lines and it works!

Hi Annette,

I have never worked with B16 cells for migration assay (I use leukemic cell lines) but a colleague who works involved various ECM compounds and integrins had provided me with their protocols for transwell migration assays in which I modified for my cell system (I use a chemoattractant). They coat the bottom of the well with different ECM compounds (collagen I, IV, fibronectin, etc) overnight at 4 degree on a rocker, typically this coats the bottom of the well and the bottom of the insert and then they do not use a chemoattractant but instead compare differences in migration to the various ECM compounds. They have uncoated wells as their control/background. But most importantly they use a "migration/adhesion buffer" (RPMI 1640 (or whichever media you use), 1 mmMgCl2, 0.2 mm MnCl2, and 0.5% BSA) for their cells to be resuspended in as well as used in the bottom of the well. This buffer activates the integrins if I recall correctly due to the MgCl2 and MnCl2. Then as someone else mentioned, use a q-tip on the inside of the insert before and after you fix/stain. Hope that helps.

-Amanda

as one of the early inventors of the amnion invasion assay which lead the the current transwell systems, i found FBS useless as a chemoattractant and for the cells your using, rat FN or human FN should work just fine and you will see a difference between FBS and FN. The FN will adhere to the underside of the membrane so cells will be in, on and through the membranes. Hope that helps.

Hello, I had similar issues as I am trying to test B16F10's with different chemoattractants. I plated 50,000 cells, 8um, coated with FBS, Collagen 1, or fibronectin, and let the cells run for 6 hours. I found that very few cells chemotaxed but, after swabbing the top, many cells remained within the pores, and that I could clearly distinguish the nucleus by hematoxylin staining. I believe these can be counted as "migrating", though I am quite new to this assay so I am not entirely sure.

I saw 10-12 cells without chemoattractant, and 190 cells with 10% FBS as chemoattractant.

Lastly, I didn't see you mention starving/synchronizing the cells, but it makes a difference whether you starve the B16F10 before migration. If you plate them at perhaps 600,000 cells into a T25, 24 hours later they are about 90% confluent. Then starve them (0-1% FBS, 0.5% is common) for an additional 24 hours to try and synchronize them in the same cell cycle phase, the counts changed and far more cells migrated.

This is a great thread! Best of luck on the assay!

-MP

I using fluorescence bovine fibronectin in my experiment to create fibronectin patterns on substrates using microfluidic technique and studying the fibroblast cell adhesion 3t3 cells. I can see that cells adhered specifically to the patterns. I got a concentration gradient in my support so now I studying the cells migration from the lower to the higher concentration using timelapse microscopy technique. Yes it is possible to use fibronectin as chemoattractant solution.

Okay this may be a silly question. Can I buy NIH3T3 conditioned medium, or is this a situation where I need to get the cell line and make it myself? I've looked around and can't find where I could purchase it. Thanks