Hi,
I'm working with thyroid follicles and tries to culture them in collagen sandwich cultures (Bovine collagen type I).

It works pretty well every third time or so and if I make for example 24 cultures (using a 96 multiwell plate) only approximately 9-12 are good for further experiments, if I'm lucky...

The problem is that the follicles don´t keep its 3D shape but start to grew as a monolayer instead.

I have tried different types of collagen concentration and I really tries to be as careful as possible when mixing but it seem like in the gels not working,
I get bubbles and it seems like the monolayer are spreading from follicle fragments close to the bubbles...

Any tip of how to avoid bubbles in the gels?

I also want to encurage other to give other tips, how do you work? 96 wells or bigger well? what volumes? Sandwich or single gel method`?
Everything is of intrest since I am pretty new on this field

Hi! The variability in 3D cell cultures is sadly a fact of life now for most of us who have done it.
I have never worked with or even heard about thyroid follicle 3D cultures, but can give you some quick general tips:
a) ensure that the collagen solution (or solutions if you are using different concentrations for your sandwich), on the plate are gently chilled as it sets
b) 96/48 well plates are the easiest to handle, DO NOT use the outer wells (fill them with sterile water/PBS instead) - however depending upon your culture, you may have to use larger wells
c) have you tried using any of the new synthetic matrices (Invitrogen, BioTek, etc) - they do not work on all cell types, but where they do work, you'll get greatly improved consistency.
d) OPTIMIZE cell numbers - that will also help you minimize the mixing and potential for forming bubbles. Many researchers do use a single gel method, embedding cells in the gel - again, its largely trial and error.
All the best
-sam

Thank you for the rapid answer!

Ok, well I guess it feels a little better to know that there are not only lack of skills that makes it fail...

I'm not totaly sure how you mean with the d), how would an optimized cell number help to minimize the mixing?

Yeah, I've been aware of that the single gel method seems to be used frequently, unfortunately I haven't been able to culture my cells in a single gel, since I use follicles/follicle fragment that are allowed to reform follicles they seems to be to dense and only sediment through the gel and form monolayers on the bottom of the well...

But I'd really like to do a single gel, since it seems to me that the monolayer are grew in the border betweens the layers of collagen...

Well, some more optimization work to do I suppose