Dear all,

I've been trying for a long time to isolate pure chloroplasts from a few non-model plant species to obtain pure plastidial RNA. For example, I'm working with a fern (Adiantum capillus veneris). While I do get a sufficient amount of intact chloroplasts that can be used for RNA extraction, I always struggle to have them pure enough. In particular, upon checking for contamination via Western Blot, I always get a strong signal for H3 ligase (nuclear marker) in my chloroplast samples. This signal is higher than the signal I get from total protein samples when loading equal amounts of chlorophyll. The same goes for PsbA (chloroplast marker). It thus looks, as if I'm enriching for both chloroplasts and nuclei at the same time.

My extraction protocol involves:

Tissue homogenization with a polytron blender in Chloroplast isolation buffer -> Filtration through a nylon mesh with a pore size of 15 um -> Pre-centrifugation of the filtered homogenate at 200 g for 5 minutes to remove nuclei and cell debris -> Centrifugation of the supernatant at 2'000 g for 5 minutes to pellet the chloroplasts -> Resuspension of the chloroplast pellet and loading onto a Percoll gradient -> Retrieval of the intact chloroplast band -> Washing the chloroplast and centrifuging at 2'000 g for 5 mins -> Final resuspension of the pellet.

I tried using both two-step (40%/80%) and continuous percoll (50%) gradients, with no improvements.

I also tried getting chloroplasts from protoplasts obtained via enzymatic treatment, but with my species of interest, this does not work so well. Also ideally I would isolate import-competent chloroplasts, and thus this method is not so well-suited.

The purity of my chloroplast samples is critical for my experiment, which will involve RNA-seq. If you have any suggestions on how to get rid of this nuclear contamination, it would be deeply appreciated.

Best,

Andrea Fontana