Zeiss LSM880 Confocal Microscope, remember to use glass-bottom multi wells

Hi Sam

I am not sure if you can successfully define the volume of a spheroid using brightfield imaging, I'd prefer fluorescence microscopy and - depending on the sample size required - with an optical sectioning capability.

For the small sample size I suppose Fahimeh's suggestion then works. For the larger ones were you keep the sample in mulitwells most high content screening setups which have confocal will then be suitable to image with the better imageing result in the lower part of the object towards the objective on the inverted setup (depending on size you might more or less of the total but after 72h you might only get a very small part of the control).

There are also solutions available providing very low phototoxicity based on light sheet microsopy, not sure if you can purchase them yet (e.g. oblique plane microscopy, SCAPE but also commercially available light sheet microscopy e.g. Zeiss) but you might need to establish a different sample mounting method. A commercially available solution for multiplexed 3d light sheet imaging has been developed by Luxendo now Bruker....

There is loads to think about and a lot depends on your assay size, the resolution you require and your financial resources. Don't take my suggestions as complete.

Best

Heiko

Dear all,

Thank you for your answers! I really appreciate your comments and help!

Fahimeh Faqihi I use the ultra-low attachment plates from Corning (Costar ultra-low cluster, 96-well, clear round bottom with lid, polystyrene). Do you think this would be fine for the analysis? Thank you for the microscope recommendation.

Heiko Dussmann exactly, optical sectional capability is actually the right and appropriate thing. Z-stacking would just change the focus, if I got it right?

For the fluorescence microscopy, would you recommend nuclear and some sort of plasma membrane staining, e.g. DAPI + e-cadherin (for epithelial cells, since I am working with HNSCC).

So regarding sample size: I normally cultivate spheroids in 60 wells and have a serial dilution with 5 different concentration of my drug. Spheroids have usually a diameter of around 600 micrometers after 72h of cultivation. After treatment with the highest dose, they obviously loose sphericity, and diameter of the spheroids reduces to 50 to 100 micrometers.

The light sheet approach sounds very interesting, however, as you mention, I would have to mount the spheroids in some way, and therefore this would only applicable as an endpoint measurement, if I have understood you correctly.

Basically, the idea was to do segmentation of the spheroids after having a series of optical sections of my object and extract voxels, so I can design a 3D-model.

Zhi Ren as I mentioned before in my reply to Fahimeh Faqihi : does the plates have to be glass bottom or is it feasible with clear round bottom wells, such as the ones from Corning? Here is a link: https://www.sigmaaldrich.com/AT/en/product/sigma/cls7007

Thank you once again for your help, I am looking forward to read your thoughts on this matter

Sam Augustine Kandathil Hi Sam, Clear bottom wells did not work for me. However, mine was adherent cells, not spheroid; if you got access to that microscope, go and give it a try. and agree with other comments on fluorescent imaging. I suggest IMARIS software; you can get Z stack and reconstruct the image using software and measure volume.

Hi Sam,

you can also use ImageJ/FiJi for 3d reconstruction and quantifications.

For the z-stacks: The optical sectioning strongly depends on the objective NA, but you might not be able to use a high NA objective with the 96well clear round bottom I am afraid. You might consider growing the spheres initially in the round bottom plate and then move them to a plate which is better for imaging for the treatment. If this is a low attachment plate that might work. Dapi and e-cadherin should give you a nice picture of the fixed spheroid at the end of the experiment. Make sure that you staining penetrates well, longer exposure to the antibodies/dyes might be required.

But only a live cell stain (e.g. Hoechst live cell nuclear stain or GFP overexpression) might be useful for your living spheres. there you have to check which concentration is well tolerated, e.g. 50-100ng/ml fro the start of the growth of the spheres in the medium. You can also add PI at a later stage to see the cells with disrupted plasma membrane.

best

Heiko

Hi Sam,

1. I supppose you are using live cell stains for your repetitive imaging of live spheroids. Hoechst works, e-cadherin ?

2. for the plate: you will need to transfer your spheroids to a thin flat bottom plate to perform the 3d measurements, you can coat the plates to achieve low attachement

3. type of microscope: any microscope with 3d imaging capability, motorised stage and focus, High content setups might be more suitable in terms of setting up the experiment, confocal point scanners might be more flexible with the area you want to scan than e.g. spinning disk confocal. But they will take a long time form 60x >100um stacks. You might want to look into options for deconvolution or use structural illumination (something like the apotome) to increase your optical 3d sectioning capability.

(...)

good luck!

Heiko