been systematically quantified [73]. These measurements are valuable to design mRNA imaging reporters with precise photochromatic properties. Furthermore, the development of fluorescent
dyes partly bypasses some of the common problems encountered
using fluorescent proteins, i.e., wide excitation and emission spectrum, sensitivity to photobleaching, tendency to multimerization
since these molecules generally have small size, high brightness and
photostability, and narrow spectrum.
In the following paragraphs we report our tests aimed at optimizing the visualization of single mRNAs in living eukaryotic cells.
We tested several MCP variants in the model organism S. cerevisiae
as well as multiple green FP fused to MCP both in S. cerevisiae and
mammalian cell lines. These comparisons revealed that for efficient
mRNA detection, MCP variants with high affinity for the RNA
reporter remain the best option. In addition, we found that the
green FP Envy shows improved brightness compared to other GFP
variants, both in yeast and mammalian cells.
1.1 mRNA Detection
Using Different MCP
Variants
We previously demonstrated that it is possible to efficiently detect
single mRNAs in S. cerevisiae by using the latest MS2 variant,
MBSV6, in combination with the expression of MCP-NLS2xyeGFP [43, 58] (Fig. 1a–d). To improve the long-term detection
of single mRNAs in living cells (i.e., brightness and photostability)
we generated other MCP constructs that we compared to
MCP-NLS-2xyeGFP for their brightness and propensity to form
aggregates. Even though we find that the MCP-NLS-2xyeGFP
performs better than other constructs tested thus far, here we
report the advantages and disadvantages of other tested reporters.
To improve the brightness of single mRNAs, we generated an
MCP variant fused to 3xyeGFP. This plasmid was transformed in
the yeast strain expressing MDN1 tagged with 24xMBSV6 and
Nup49-tdTomato (Fig. 1e, f). We compared the brightness of
MDN1 mRNAs detected with either MCP-2xyeGFP or
MCP-3xyeGFP (Fig. 1g). As expected, mRNAs labeled with
MCP-3xyeGFP shows a 25% increase in brightness compared to
mRNAs labeled with MCP-2xyeGFP (310,950 Æ 155,349 a.u. and
263,373 Æ 109,890 a.u., respectively). In addition, the number of
MDN1 mRNAs per cell counted with the MCP-3xyeGFP reporter
is similar to the mRNAs counted with MCP-2xyeGFP or by
smFISH (mean Æ S.D. 9.2 Æ 6.1 mRNAs/cell, Fig. 1h). However,
we found that the MCP-3xyeGFP reporter has the tendency to
induce cytoplasmic aggregates, likely due to the propensity of GFP
to multimerize (Fig. 1f, orange arrowheads). These aggregates are
similar to the ones that we previously described [58] and that can
lead to artifactual conclusions about mRNA localization in yeast. It
may still be possible to use the MCP-3xyeGFP reporter for mRNAs
that are less abundant than MDN1, but we recommend always
comparing the live imaging results to smFISH to avoid false
conclusions.
124
Xavier Pichon et al.
dyes partly bypasses some of the common problems encountered
using fluorescent proteins, i.e., wide excitation and emission spectrum, sensitivity to photobleaching, tendency to multimerization
since these molecules generally have small size, high brightness and
photostability, and narrow spectrum.
In the following paragraphs we report our tests aimed at optimizing the visualization of single mRNAs in living eukaryotic cells.
We tested several MCP variants in the model organism S. cerevisiae
as well as multiple green FP fused to MCP both in S. cerevisiae and
mammalian cell lines. These comparisons revealed that for efficient
mRNA detection, MCP variants with high affinity for the RNA
reporter remain the best option. In addition, we found that the
green FP Envy shows improved brightness compared to other GFP
variants, both in yeast and mammalian cells.
1.1 mRNA Detection
Using Different MCP
Variants
We previously demonstrated that it is possible to efficiently detect
single mRNAs in S. cerevisiae by using the latest MS2 variant,
MBSV6, in combination with the expression of MCP-NLS2xyeGFP [43, 58] (Fig. 1a–d). To improve the long-term detection
of single mRNAs in living cells (i.e., brightness and photostability)
we generated other MCP constructs that we compared to
MCP-NLS-2xyeGFP for their brightness and propensity to form
aggregates. Even though we find that the MCP-NLS-2xyeGFP
performs better than other constructs tested thus far, here we
report the advantages and disadvantages of other tested reporters.
To improve the brightness of single mRNAs, we generated an
MCP variant fused to 3xyeGFP. This plasmid was transformed in
the yeast strain expressing MDN1 tagged with 24xMBSV6 and
Nup49-tdTomato (Fig. 1e, f). We compared the brightness of
MDN1 mRNAs detected with either MCP-2xyeGFP or
MCP-3xyeGFP (Fig. 1g). As expected, mRNAs labeled with
MCP-3xyeGFP shows a 25% increase in brightness compared to
mRNAs labeled with MCP-2xyeGFP (310,950 Æ 155,349 a.u. and
263,373 Æ 109,890 a.u., respectively). In addition, the number of
MDN1 mRNAs per cell counted with the MCP-3xyeGFP reporter
is similar to the mRNAs counted with MCP-2xyeGFP or by
smFISH (mean Æ S.D. 9.2 Æ 6.1 mRNAs/cell, Fig. 1h). However,
we found that the MCP-3xyeGFP reporter has the tendency to
induce cytoplasmic aggregates, likely due to the propensity of GFP
to multimerize (Fig. 1f, orange arrowheads). These aggregates are
similar to the ones that we previously described [58] and that can
lead to artifactual conclusions about mRNA localization in yeast. It
may still be possible to use the MCP-3xyeGFP reporter for mRNAs
that are less abundant than MDN1, but we recommend always
comparing the live imaging results to smFISH to avoid false
conclusions.
124
Xavier Pichon et al.