Immediately following, it is often best to wipe the non-cell side
of the coverslip dry, to prevent bubbles and ensure clarity
during imaging.
14. To determine the degree of ER association, one should stain
cells where the digitonin extraction step is skipped (Subheading 3.1, steps 7 and 8). This can serve as a baseline control for
any given experimental set. The membrane of these cells will be
extracted during the incubation with ice-cold methanol (see
Subheading 3.1, step 13).
15. To ensure that the observed smFISH mRNAs are truly
anchored to the ER, it is also worthwhile to extend the length
of extraction to ensure that all cells are permeabilized and all
free mRNAs have diffused away, as we have done previously
[11]. This can be tricky as the general integrity of the ER
begins to deteriorate after very long extraction times. The
degree of extraction can be assessed by phase or DIC microscopy. Extracted cells have a less dense cytoplasm, a more
defined nuclear rim, and an increase in the number of round
organelles—we believe this is due to mitochondria losing their
extended shape during extraction.
16. To further validate that observed smFISH mRNA foci are
anchored to the ER, smFISH signals can be co-imaged with
an ER marker, such as GFP-KDEL (Fig. 4), or with an ER
protein that is stained by immunofluorescence (see Note 23).
One should be aware that certain overexpressed markers can
disrupt mRNA-ER association. For example, we have observed
that overexpression of GFP-Sec61β disrupts the ER association
of all other mRNAs [11, 18]. This is likely due to the GFPSec61β mRNA effectively outcompeting all other mRNAs from
ER-binding sites. Indeed, to further validate that observed
smFISH mRNA foci are truly anchored to the ER, one can
overexpress GFP-Sec61β, and assess whether these foci disappear after digitonin extraction [11, 18].
17. To test for if smFISH RNA foci are associating with the ER in a
translation-dependent manner, cells can be treated with compounds that dissociate ribosomes from the mRNA, such as
puromycin [16].
18. To ensure that the cytosol has been extracted, one can stain the
unextracted and extracted cells for various components. For
cytosolic proteins, it is best to immunostain those that are
present in the cytosol and nucleus, such as adenosine kinase
[16], so that in the extracted cells the immunofluorescence
signal is present only in the nuclear compartment. For
mRNAs, those that encode cytosolic or nuclear proteins can
be stained by smFISH; however digitonin extraction only
removes about 90% of these from the cytosol, with the
46
Jingze J. Wu and Alexander F. Palazzo
of the coverslip dry, to prevent bubbles and ensure clarity
during imaging.
14. To determine the degree of ER association, one should stain
cells where the digitonin extraction step is skipped (Subheading 3.1, steps 7 and 8). This can serve as a baseline control for
any given experimental set. The membrane of these cells will be
extracted during the incubation with ice-cold methanol (see
Subheading 3.1, step 13).
15. To ensure that the observed smFISH mRNAs are truly
anchored to the ER, it is also worthwhile to extend the length
of extraction to ensure that all cells are permeabilized and all
free mRNAs have diffused away, as we have done previously
[11]. This can be tricky as the general integrity of the ER
begins to deteriorate after very long extraction times. The
degree of extraction can be assessed by phase or DIC microscopy. Extracted cells have a less dense cytoplasm, a more
defined nuclear rim, and an increase in the number of round
organelles—we believe this is due to mitochondria losing their
extended shape during extraction.
16. To further validate that observed smFISH mRNA foci are
anchored to the ER, smFISH signals can be co-imaged with
an ER marker, such as GFP-KDEL (Fig. 4), or with an ER
protein that is stained by immunofluorescence (see Note 23).
One should be aware that certain overexpressed markers can
disrupt mRNA-ER association. For example, we have observed
that overexpression of GFP-Sec61β disrupts the ER association
of all other mRNAs [11, 18]. This is likely due to the GFPSec61β mRNA effectively outcompeting all other mRNAs from
ER-binding sites. Indeed, to further validate that observed
smFISH mRNA foci are truly anchored to the ER, one can
overexpress GFP-Sec61β, and assess whether these foci disappear after digitonin extraction [11, 18].
17. To test for if smFISH RNA foci are associating with the ER in a
translation-dependent manner, cells can be treated with compounds that dissociate ribosomes from the mRNA, such as
puromycin [16].
18. To ensure that the cytosol has been extracted, one can stain the
unextracted and extracted cells for various components. For
cytosolic proteins, it is best to immunostain those that are
present in the cytosol and nucleus, such as adenosine kinase
[16], so that in the extracted cells the immunofluorescence
signal is present only in the nuclear compartment. For
mRNAs, those that encode cytosolic or nuclear proteins can
be stained by smFISH; however digitonin extraction only
removes about 90% of these from the cytosol, with the
46
Jingze J. Wu and Alexander F. Palazzo