by the induction of clusters where they are thereby inactive
[16, 17]. As many membrane receptors are activated by ligandinduced dimerization, it is also possible to use optogenetics to
mimic this activation [18, 19]. It is also possible to activate
cytosolic kinases with optogenetics. Some of the constructs
that have been designed to date are somehow involved in
EMT, but the design of new constructs using these methods
require an in-depth knowledge of the biochemical and structural properties of the targeted kinases [20, 21]. There are
currently several optogenetic systems available to control the
activity of Rho GTPases: photoactivatable Rac, making use of
the LOV system [8], the CIBN/CRY2 system [9], the Phytochrome B/PIF system [10], the LOV/TULIP system [22],
the iLID system [11]. None of these systems is perfect or
intrinsically better than the others. One has to choose the
best system for each experimental condition, and several parameters should be considered.
2. Spatiotemporal resolution of active Rho GTPase localization is
dictated by the distribution of Rho GTPase activators. It is
determined by the initial recruitment of the cytoplasmic partner to the membrane, the diffusion of the dimer at the plasma
membrane (which increases the spreading of the activation)
and the dissociation time of the dimer. The shorter the lifetime
of the dimer is, the less it diffuses at the plasma membrane
before dissociating. In practice, it is not possible to create
subcellular activation smaller than 4–5 μm in the case of the
first versions of CRY2/CIBN and TULIP systems [13]. So far,
the recent iLID systems have the best spatial resolution
(1–2 μm) as well as the PhyB/PIF (in which dissociation can
be forced with far-red illumination) [10]. However, that comes
with a price: as dissociation happens faster, it requires more
frequent illumination to maintain a constant activation, hence
more phototoxicity and photobleaching.
3. It is possible to achieve optogenetic activation of Cdc42 with
the ITSN-CRY2-mCherry/CIBN-GFP-CAAX dimer through
both transient transfections and stable cell lines. Transient
transfections are faster to set up and will usually give rise to
high expression levels of both components. However, their
reproducibility is difficult to control and can be problematic if
it is necessary to photo-activate a group of cells. Given the
limitations enunciated before with regard to expression levels,
stable cell lines are usually recommended for reproducibility
and ease of use. In this case, it is highly recommended to
FACS-sort the stable cell lines on the basis of their mCherry
and GFP expression levels. However, we noticed that expression levels tend to be lower in stable cell lines, and that some
cell lines tend to spontaneously repress either CIBN or CRY2,
420
Leo Valon and Simon de Beco
[16, 17]. As many membrane receptors are activated by ligandinduced dimerization, it is also possible to use optogenetics to
mimic this activation [18, 19]. It is also possible to activate
cytosolic kinases with optogenetics. Some of the constructs
that have been designed to date are somehow involved in
EMT, but the design of new constructs using these methods
require an in-depth knowledge of the biochemical and structural properties of the targeted kinases [20, 21]. There are
currently several optogenetic systems available to control the
activity of Rho GTPases: photoactivatable Rac, making use of
the LOV system [8], the CIBN/CRY2 system [9], the Phytochrome B/PIF system [10], the LOV/TULIP system [22],
the iLID system [11]. None of these systems is perfect or
intrinsically better than the others. One has to choose the
best system for each experimental condition, and several parameters should be considered.
2. Spatiotemporal resolution of active Rho GTPase localization is
dictated by the distribution of Rho GTPase activators. It is
determined by the initial recruitment of the cytoplasmic partner to the membrane, the diffusion of the dimer at the plasma
membrane (which increases the spreading of the activation)
and the dissociation time of the dimer. The shorter the lifetime
of the dimer is, the less it diffuses at the plasma membrane
before dissociating. In practice, it is not possible to create
subcellular activation smaller than 4–5 μm in the case of the
first versions of CRY2/CIBN and TULIP systems [13]. So far,
the recent iLID systems have the best spatial resolution
(1–2 μm) as well as the PhyB/PIF (in which dissociation can
be forced with far-red illumination) [10]. However, that comes
with a price: as dissociation happens faster, it requires more
frequent illumination to maintain a constant activation, hence
more phototoxicity and photobleaching.
3. It is possible to achieve optogenetic activation of Cdc42 with
the ITSN-CRY2-mCherry/CIBN-GFP-CAAX dimer through
both transient transfections and stable cell lines. Transient
transfections are faster to set up and will usually give rise to
high expression levels of both components. However, their
reproducibility is difficult to control and can be problematic if
it is necessary to photo-activate a group of cells. Given the
limitations enunciated before with regard to expression levels,
stable cell lines are usually recommended for reproducibility
and ease of use. In this case, it is highly recommended to
FACS-sort the stable cell lines on the basis of their mCherry
and GFP expression levels. However, we noticed that expression levels tend to be lower in stable cell lines, and that some
cell lines tend to spontaneously repress either CIBN or CRY2,
420
Leo Valon and Simon de Beco