Myc regulates programmed cell death and radial glia dedifferentiation after neural injury in an echinoderm

Electroporation of DsiRNAs reduces Myc expression in the injured radial nerve

In order to determine the functional role of Myc in the early response to the CNS injury in H. glaberrima, we designed two different Dicer-substrate small interfering RNAs (DsiRNAs), designated
as Myc Dsi1 and Myc Dsi2, which targeted distinct (non-overlapping) regions of the
Myc transcript (Fig. 2). The decision to use longer DsiRNAs with a 25-nt sense strand and a 27-nt antisense
strand rather than more traditional shorter 21-nt duplexes was based on the fact that
DsiRNAs can be up to 100-fold more efficient than the classical 21-mers 15]. DsiRNAs were injected and electroporated into the RNC (Additional files 1 and 2). Besides reagent delivery, the injection procedure also served another function.
The diameter of the injection needle was chosen to be greater than the width of the
RNC, so that complete transection is achieved during the injection procedure (Additional
file 2). Two days after surgery, quantitative real-time RT-PCR (qRT-PCR) showed that injection
and electroporation of either Myc Dsi1 or Myc Dsi2 caused a significant ?1.9-fold
decrease in Myc mRNA expression compared with the injection of the control GFP-trageting DsiRNA (Fig.
3a), whereas the animals treated with the control DsiRNA themselves did not show any
significant differences when compared with the animals injected with the vehicle alone.

Fig. 2. Diagram showing the sequences of the two Myc-targeting DsiRNAs (Myc Dsi1 and Myc Dsi2) and their target sites within the open
reading frame (ORF) of the H. glaberrimaMyc transcript. Red and blue letters indicate additional RNA and DNA bases, respectively,
which distinguish DsiRNAs from classical 21-mer duplexes

Fig. 3. RNA interference-mediated Myc knockdown. aMyc expression in the regenerating radial nerve cord on day 2 after injury/DsiRNA injection
as determined by qRT-PCR. Two DsiRNA constructs were used, Myc Dsi1 and Myc Dsi2,
as described in Methods. Expression values are plotted as fold change relative to
a negative control (a GFP-targeting DsiRNA) and expressed in a log2 scale. Error bars
show standard deviation. ** p0.01, *** p0.001. b-c’ Representative in situ hybridization micrographs showing Myc expression in the radial nerve cord on day 2 after transection/DsiRNA injection.
The upper (b and b’) and lower (c and c’) rows of micrographs show longitudinal sections of the RNC of an animal treated
with a control (GFP-targeting) DsiRNA and an animal injected with one of the Myc-targeting DsiRNAs (Myc Dsi1), respectively. The micrographs on the right (b’ and c’) are high-magnification view of the boxed regions in the main micrographs on the
left (b and c, respectively). The red dashed line indicates the position of the plane of injury.
Note the absence of in situ hybridization signal from the cell bodies in the apical
region of the ectoneural neuroepithelium of the RNC in the animal treated with Myc
Dsi1, but not in the animal, which received the control DsiRNA injection

Likewise, in situ hybridization shows that RNAi-mediated gene targeting causes Myc expression to fall below the detection limit in the cells adjacent to the site of
injury/injection. No knockdown occurred when the animals were injected with the control
GFP-targeting Dsi RNA (Fig. 3b-c’).

Forced downregulation of Myc transcripts impairs dedifferentiation of radial glia

Glial dedifferentiation is one of the key cellular events that take place in the injured
sea cucumber RNC shortly after transection 1], 3]. Under normal conditions, fully differentiated radial glial cells show typical palisade
morphology, with their cell bodies mostly localized to the apical region of the neuroepithelium
and the long basal processes extending through the entire thickness of the neural
parenchyma 16], 17]. In response to injury, however, the glial cells loose their basal processes, which
undergo fragmentation and become eventually phagocytosed by adjacent cells. The glial
cells bodies persist in the apical region of the neuroepithelium and retain epithelial
characteristics such as intercellular junctions and apicobasal cell polarity, but
undergo extensive reogranization of the cytoskeleton and chromatin decondensation
1], 3].

In order to determine if there is any effect of RNAi-induced Myc knock-down on post-traumatic dedifferentiation of radial glial cells, we measured
the relative area of the RNC occupied by glial cells, which still retained their basal
processes (i.e., remained in the differentiated state) on day 2 post-injury. The measurements
were performed on sagittal sections and covered the area spanning 1 mm from the site
of the injury/injection. Myc knock-down had a highly significant effect on the extent of glial dedifferention
(one-way ANOVA, F(3,13)=18.99,p=4.94×10^?5). Electroporation of either of the two DsiRNAs, Myc Dsi1 or Myc Dsi2, resulted in
a ?50–60-fold increase in the size of the relative area of the radial nerve cord occupied
by differentiated radial glia (with basal processes), whereas electroporation of the
control (GFP-targeting) DsiRNA resulted in exactly the same phenotype as injection
of the vehicle alone, i.e. extensive dedifferentiation (loss of the basal processes)
in the radial glia (Figs. 4 and 5a). Therefore, Myc is required for induction of dedifferentiation of radial glial cells after CNS injury.

Fig. 4. Representative micrographs showing the effect of Myc knockdown on glial dedifferentiation on day 2 post-injury/DsiRNA injection. The radial
glial cells are visualized by immunostaining with the ERG1 monoclonal antibody (red)
17]; the nuclei (in a, b, d, e) were stained with Hoechst (blue). All micrographs are longitudinal sections with
the plane of the injury (dashed line) to the right. a and b Control injections of the vehicle (a) and an irrelevant (GFP-targeting) DsiRNA (b). c Higher magnification of the radial nerve cord in a control (vehicle-injected) animal.
d and e Injection of Myc-targeting DsiRNAs, Myc Dsi1 (d) and Myc Dsi2 (e). f Higher magnification of the radial nerve cord in a Myc Dsi2-injected animal. Note
that in the control animals (a-c) the glial cells loose their long basal processes, while the cell bodies (arrow in c) remain in the apical region of the ectoneural neuroepithelium (en). In contrast, many of the radial glial cells in the animals injected with Myc-targeting DsiRNAs (d-f) retained their basal processes, which extended through the underlying neural parenchyma
(asterisk)

Fig. 5. Effect of RNAi-mediated Myc silencing on glial dedifferentiation and programmed cell death. a Relative area occupied by fully differentiated radial glial cells within 1 mm from
the wound in control animals injected with the vehicle or an irrelevant (GFP-targeting)
DsiRNA (GFP Dsi) and in animals injected with Myc-targeting DsiRNAs (Myc Dsi1 and Myc Dsi2). Day 2 post-injury/DsiRNA injection. b Relative abundance of TUNEL-positive cells in the radial nerve cord within 1mm from
the wound on day 2 post injury/DsiRNA injection. The control animals were injected
either with the vehicle or with an irrelevant DsiRNA (GFP Dsi). The Myc-targeting DsiRNAs are designated as Myc Dsi1 and Myc Dsi2. The data are plotted as
mean ± standard error. * p0.05, ** p0.01, *** p0.001

Myc knock-down decreases the extent of programmed cell death in the injured radial
nerve cord

Besides glial dedifferentiation, extensive programmed cell death is another characteristic
feature of the early post-injury phase in the RNC 1], 3]. By day 2, the number of apoptotic cells in the vicinity of the injury increases
20-fold in comparison with uninjured animals and then starts to gradually return to
the normal levels as regeneration progresses 3].

Given that Myc, depending on the context, is known either to trigger or suppress apoptosis
in mammalian models 12], 18], we employed TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP end labeling)
assay to determine if Myc is involved in regulation of the programmed cell death in the injured radial nerve
cord of H. glaberrima. There was a significant effect (one-way ANOVA, F(3,15)=9.4,p=9.7×10^?4) of RNAi-mediated Myc down-regulation on relative abundance of TUNEL-positive cells in the vicinity (1
mm) of the wound on day 2 post-injury. Myc-targeting DsiRNAs caused a ?2–4-fold decrease in the number of the cells undergoing
programmed cell death, whereas the control GFP-targeting DsiRNA did not result in
any changes in comparison with vehicle-injected animals (Figs. 5b and 6).

Fig. 6. Representative micrographs showing the effect of Myc on programmed cell death on day 2 post-injury/DsiRNA injection. The cells undergoing
programmed cell death were visualized with TUNEL assay (green). All micrographs are
sagittal sections with the plane of the injury (dashed line) to the right. Dotted line indicates the outline of the radial nerve cord. a and b Control injections of the vehicle (a) and irrelevant control GFP-targeting DsiRNA (b). c and d Injection of Myc-targeting DsiRNAs, Myc Dsi1 (c) and Myc Dsi2 (d)