Morphogenetic effect of Shh signaling interference
To explore whether disturbance of Shh signaling modifies the OT morphogenesis we first analyzed its phenotype after either hedgehog agonist Pur or antagonist Cyc treatments. Figure illustrates the effects of gain and loss of functions experiments by means of pharmacological treatments applied locally in ovo over two different temporal windows (ED1.5–4.5 and ED5–7). Figure A-C show that ED4.5 old OT underwent significant changes in size after treatments with Cyc or Pur applied at ED1.5. Figure D-F illustrate the effect produced by treatments performed at ED5 and analyzed at ED7. Importantly, during both temporal windows Shh antagonist produces significant reductions in the OT, dorsal prosencephalon (future brain hemispheres) and optic vesicle (future retina) while the Shh agonist produces the opposite effects, i.e. larger than usual OT, dorsal prosencephalon and optic vesicles.
Figure 1 Shh signaling is required for OT expansion. Representative examples of the effects of pharmacological loss and gain of functions experiments with the antagonist Cyclopamine (Cyc) and the agonist Purmorphamine (Pur) applied locally in ovo at two developmental (more ...)
In order to specifically analyze the effect of Shh signalling on the developing OT, localized electroporation of exogenous Shh and GliA was performed at the DMB. Figure shows the result of the electroporation procedure; patches of GFP labeling can be seen distributed over the DMB (Figure A). Histological examination of these GFP+ areas is illustrated in next section. To further evaluate the effectiveness of electroporation and to ascertain whether Shh related genes expressions are modified, the expression of ptc1, and Pax7 was analyzed (by in situ hybridization and immunohistochemistry respectively). We also analyzed the expression of Hnf3β, a typical marker of the basal midbrain.
Figure 2 Results of electroporation. A. Left lateral view of the cephalic region of Stg 17 (upper panel) and Stg 21 (bottom panel) embryos electroporated at Stg 11. Patches of GFP+ areas are observed over the DMB surface. Bars: 100 μm. B. In situ hybridization (more ...)
Shh electroporated DMBs at ED4.5 displayed increased expression of ptc1 (Figure B) and downregulation of pax7 (Figure C) in comparison to the control ones confirming that, indeed, cells located at the dorsal midbrain actively respond to Shh electroporation.
Figure B illustrates in situ hybridization for detection of ptc1 expression. In the control embryo ptc1 expression is restricted to the ventral neural tube; in Shh electroporated embryos, ectopic ptc1 expression is observed in the dorsal midbrain. Figure C illustrates double staining for Shh and Pax7. Control embryos display Shh immunoreactivity in the basal midbrain and Pax7 reactivity in the dorsal midbrain. By contrast the dorsal midbrains of Shh electroporated embryos show ectopic overexpression of Shh and downregulation of Pax7.
Figure D shows double staining for Shh and Hnf3β. In control specimens the dorsal midbrain is negative for Shh labeling and the ventral marker Hnf3β, typically stain the basal midbrain. In Shh electroporated midbrains Shh labeling is visualized dorsally and, in most cases, Hnf3β is restricted to the basal plate. This result suggest that in most cases dorsal progenitors could not be re-specified to ventral lineages, rather they are destined to differentiate toward a tectal fate even in the presence of Shh misexpression. In some Shh electroporated specimens the ventral domain of Hnf3β expression expands laterally into a region that normally corresponds to the OT. This lateral expansion of the Hnf3β+ area invading a zone that normally differentiate into OT suggest an evolution towards a basal fate and could correspond to specimens that, under histological examination, reveal a partial ventralization of the lateral region of the alar plate (see Figure D).
Figure 3 Micrographs of D-V sections of ED4.5 midbrains (H-E stained). A. Control. B. GliA electroporated. C. Shh non-ventralized electroporated. D. Shh partially ventralized electroporated, The intertectal sulcus (arrowhead) that separates the left and right (more ...)
Figure A-D show images of D-V sections, located halfway between the cephalic and caudal ends of the MB, obtained from ED4.5 embryos. DMB submitted to GliA electroporation retained its alar character and differentiated into typical OT (Figure B). Shh electroporation produced two types of results: the majority (7/10) of DMBs retained their alar character and differentiated into OT (Figure C) but some of them (3/10) underwent partial ventralization. In this last case, the DMBs had two different regions (Figure D): the dorsalmost region retained the alar phenotype (“tectal region of ventralized DMB)”- while its ventral zone changed its developmental fate and differentiated into basal structures (“ventralized region of the DMB”).
Apart from their differences in size, by ED4.5 the control OT typically exhibits the incipient dorsal-medial intertectal sulcus (between the left and right hemispheres) and the roof plate at the bottom of the medial sulcus. The roof plate is identified as the thin neuroepithelial band at the dorsal midline (Figure A and E). Normally, the tectal roof plate later differentiates into the lamina commissurales (dorsal intertectal commissure). In GliA electroporated OT the intertectal sulcus tends to disappear but the roof plate can still be recognized in 100% of the embryos as a thinner dorsal-medial neuroepithelial band (Figure B and E). In Shh electroporated specimens the intertectal sulcus disappears and the dorsal aspect of both the left and the right hemitectum form a single dorsal dome without any separation between them (Figure C and E). In Shh electroporated (non-ventralized) DMBs the roof plate does not completely disappear; patches of poorly demarcated zones with a roof plate-like organization can be observed along the dorsal midline (Figure E). In partially ventralized DMB the roof plate disappeared and the dorsal midline can only be identified by the position of the dorsal medial blood plexus (Figure D and E). The ventralized region of the DMB changes its typical tectal structure to a histological organization characteristic of the basal region (Figure A and D). It is composed of groups of neurons resembling the basal organization. The neurons of these motor-like nuclei originate axon fascicles that emerge through the ventral-lateral aspects of the MB forming one or more supernumerary motor-like nerves as judged by their position (Figure E and F). The persistence of the sulcus limitans allows distinguishing the original basal region from the ventralized region of the DMB. An additional sulcus limitans appears between the tectal and the ventralized region of the DMB.
Figure 4 Dorsal-ventral sections of a partially ventralized dorsal midbrain. A partially ventralized dorsal midbrain typically displays a dorsal region with an alar structure and a ventralized region with a basal-like organization. A-C. These sections show the (more ...)
Histogenetic and immunocitochemical characterization of the OT developmental stages in Shh and GliA electroporated dorsal midbrain
The ventralized region of the DMB displays an altered histogenetic organization that can not be characterized within the reference provided by table of developmental stages given in Reference [11
]. For that reason, only observations made in non-ventralized DMBs, i.e., DMBs that retain their tectal character, will be described in this section. Figure A–F show that by ED4/4.5, the OT is at the end of the developmental stage 1 (DS1). At the beginning of DS1 (ED2) the OT structure corresponds to a single layered neuroepithelium. Between ED2–ED4/4.5 the OT evolves through DS1: (a) the 1st
neuronal cohort (future Mes5 sensory neurons of the mesencephalic trigeminal nucleus) appears along the dorsal midline between ED2-ED4 and (b) the 2nd
cohort (future large efferent neurons of the SGC and SGP) appears over the entire OT neuroepithelium from ED3.5-ED4 onwards. These latter neurons are born at the ventricular zone (VZ), near the inner limiting membrane (ILM), and then move to the outermost subpial zone forming an incipient premigratory zone (PMZ).
Figure 5 Radial organization and immunocytochemical patterns at the end of DS1. A. Ectopic expression of pShhiEGFP in an electroporated OT (ED4.5); patches of positive NE cell bodies at the ventricular zone (VZ) and of postmitotic neurons at the premigratory zone (more ...)
Figure A show that both NEcs -ventricular zone- as well as postmitotic neurons - premigratory zone- express Shh revealing a successful electroporation procedure (Figure A). During this period the VZ displays the intense Notch reactivity typical of mNE cells (Figure D) and both the hematoxylin-eosin staining (Figure B) and the Phospho-histone H3 (PH3) immunolabeling (Figure C) reveal groups of mNEcs overlying the ILM. Besides, the NeuroD immunolabeling (Figure E) shows that the PMZ is populated by newly-born neurons characterized by intense NeuroD nuclear reactivity. The beta III Tubulin (βIIITub) labeling (Figure F) shows that these NeuroD+ neurons have already begun the early differentiating phase. These newly born neurons correspond to future large efferent neurons of the SGC. Figure B and C show that both the H-E staining and the PH3 immunolabeling permit easy identification of mNEcs and allow reliable recordings of their position along the D-V section (See also “Mitotic NE cell records” in Methods).
2D representation of mNEc records along the D-V axis
Figure A-D correspond to 2D maps of mNEcs records obtained from control, GliA electoporated and Shh electroporated DMBs (non-ventralized and partially ventralized DMBs). The 2D maps of mNEcs allocations strictly coincide with the ILM contours indicating that the allocation of the entire population of mNEcs reliably reproduces their positions along the D-V axis (Compare with Figure A-D). In each case, the percentage of ILM area occupied by mNEcs is indicated. A statistical comparison shows that GliA and Shh electroporation significantly increased the area of ILM occupied by mNEcs in the OT. The increase in this parameter is even higher in the tectal region of the partially ventralized DMB.
Figure 6 Two dimensional representations of mNEc records. A. Control. B. GliA electroporated. C. Shh electroporated, non-ventralized. D. Shh electroporated, partially ventralized. The position of each circle is specified by the spatial co-ordinates of each mNEc. (more ...)
There were not significant differences amongst the VMB of control, GliA and Shh electroporation without ventralization (Figure A, B and C). However, both the ventralized region of DMB and the VMB of these specimens (Figure D) display percentages of ILM area occupied by mNEcs significantly higher than the other specimens. It can be noted that the ventralized region of the DMB differs significantly from the tectal region of the same DMB but closely coincide with the values of the VMB.
Statistical analyses of signals derived from mNEc records
Figure summarizes the effects of GliA and Shh electroporation on the I-MI length, the mNEc density and the fractal dimension (estimated by the space filling property). This figure compares the values of these parameters measured in different regions of the MB in control specimens, GliaA electroporated and Shh electroporated MBs (non-ventralized and partially ventralized DMBs). These values correspond to global estimations performed over the entire D-V axis of each region.
Figure 7 Summary of the effects of GliA and Shh electroporations A. Control. B. GliA electroporated. C. Shh electroporated (non-ventralized). D. Shh electroporated (partially ventralized). Values of Mean ± standard deviation of inter-mitotic interval length (more ...)
The global mean I-MI length was significantly shorter in GliA and Shh electroporated (non-ventralized) DMB than in the control OT indicating closer positions between neighboring mNEcs (Figure A-C). There were no significant differences amongst the VMB of these three specimens. The effect of Shh was more intense in specimens with partially ventralized DMB (Figure D). In this case, the mean I-MI length in the tectal region was significantly lower than the value corresponding to the OT of the other three specimens. The value corresponding to the ventralized region of the DMB was higher than that of the tectal regions of the same specimens and approximate to that of the VMB of the same specimens. Besides, the VMB of these specimens underwent a significant and remarkable decrease in I-MI length compared with the VMB of the other three specimens.
With regards to the mNEc density in the alar plate, GliA and Shh electroporation results in significant increases of this parameter with respect to the control OT. However, no significant changes were detected amongst the VMBs of the three specimens. In Shh electroporated partially ventralized DMB, the mNEc density in the tectal region was significantly higher than those of the corresponding regions of the OT of the other three speciments. The ventralized region of the DMB underwent a decrease in mNEc with respect to the tectal region of the same specimens and also with respect to the alar plate of Shh electroporated specimens without ventralization. Interestingly, the mNEc density in the ventralized region of the DMB and in the VMB of the same specimens closely coincide. These mNEc densities however, differs significantly from the values observed in the VMB of the other three specimens.
Consistently with the increase in the global mNEc density, the Box Counting method applied to binary signals revealed a significant increase in the fractal dimension (space filling property) in GliA and Shh electroporated (non-ventralized) DMB (Figure A-C). In specimens with a partially ventralized DMB the value of the fractal dimension also significantly differed from the other three specimens.
All these analyses show that the pattern of values describing the proliferative behavior in the ventralized region of the DMB closely coincides with that of the VMB of the same specimens. These results indicate that the ventralization of the DMB does not only involve a change in its histological organization but also a change in the proliferative behavior of the NEc population residing in the ventralized region.
Analyses of local variations in mNEcs density performed in GliaA electroporated and in Shh electroporated without alar plate ventralization, estimated in successive 25 μm length windows along the D-V axis, indicate the existence of graded space-dependent differences in response to electroporation along this axis (Figure ). In fact, the change in mNEc density in response to the GliA and Shh electroporation is maximal near the dorsal midline and from this zone decreases gradually towards the OT-VMB boundary. These analyses also show that the response to Shh electroporation is higher than the response to GliA electroporation.
Figure 8 Space-dependent differences of Shh and GliA effects along the D-V axis. Each bar represents the mean ± standard deviation of the mNEc density measured in 500 μm length spatial windows located at defined positions along the D-V axis. The (more ...)
Non-linear analyses of signals derived from mNEc records
Figure A-C illustrate examples of I-MI signals, mNEc density signals and binary signals derived from mNEc records performed in control OT, GliA electroporated DMB and Shh electroporated (non-ventralized and partially ventralized) DMBs.
Figure 9 Signals derived from the mNEc records. A. I-MI signals. Each value corresponds to the distance between two adjacent mNEcs. B. Mitotic density signals. Each value corresponds to the density of mNEcs (number of mNEcs/100 μm2) in successive spatial (more ...)
Power spectral density (PSD)
Figure A-D illustrate results of classical spectral analyses performed on mNEc density signals corresponding to controls and GliA and Shh electroporated alar plates. In the case of Shh electroporations with partially ventralized DMB, only the alar (non-ventralized) region was included in this analysis. Values of the means and the standard deviations of the scaling index β obtained from the slope of the lines in the log-log plots are indicated. The statistical analysis of the differences amongst the means values of β indicates the existence of significant differences between the control OT and the remaining alar plates (p<0.05, p<0.01 and p<0.001 respectively). The value of β estimated in the tectal region of the partially ventralized DMB was significantly higher that the values obtained in GliA and Shh electroporated OT. These results indicate that both exogenous GliA and Shh modified the mNEcs spatial organization and that the ventralized specimens underwent a more drastic reorganization of mNEcs.
Figure 10 Power spectral density (PSD) and estimation of the scaling index β. Log-log plots of the PSD of mNEc density signals obtained from controls (A) and GliA electroporated (B) and Shh non-ventralized (C) and tectal region of the partially ventralized (more ...)
Fano factor (FF)
Figure A-D illustrate results of FF analyses applied to binary signals corresponding to control, GliA and Shh electroporated alar plates. The values of the mean and the standard deviation of the scaling index α obtained from the slope of the line in the log-log plots are indicated in each case. The statistical comparison amongst the values of α indicates the existence of significant differences between control vs. Shh electroporated specimens (p<0.04). These last values are also significantly higher than the one corresponding to the GliA electroporated alar plates. The values of α obtained from controls OT approximate the value corresponding to standard Poisson-like process while those of the Shh treated OT correspond to stationary correlated stochastic point process. Consistently with the preceding analysis, the value α estimated in the tectal region of the partially ventralized DMB was significantly higher than the value of the OT of the non-ventralized DMB.
Figure 11 Fano Factor (FF) analyses of binary signals. Log-log plot of the FF algorithm applied to binary signals obtained from control (A), GliA electroporated (B) and Shh non-ventralized (C) and tectal region of Shh partially ventralized electroporated dorsal (more ...)
Hierarchical clustering analysis (HCA)
Figure A-D show dendrograms obtained by means of hierarchical clustering analyses of binary signals corresponding to control, GliA electroporated and Shh electroporated (non-ventralized and partially ventralized) DMBs respectively. Dendrograms corresponding to the four different conditions display obvious visual differences. These differences can be quantitatively evaluated by the values of the slope of the lines in the log-log plots of the number of clusters as a function of the inter-cluster interval length.
Figure 12 Hierarchical clustering analyses (HCA) of binary signals. Dendrograms obtained by means of a HCA of the binary signals and log-log plots of the number of clusters as a function of the inter-cluster interval length obtained from controls (A) and GliA electroporated (more ...)
The values of the scaling indexes estimated in GliA and Shh electroporated alar plates were significantly higher that the one corresponding to the control OT. Besides, coinciding with preceding analyses, the scaling index estimated in the tectal region of the partially ventralized DMB was higher than the value observed in GliA and Shh (non-ventralized) eletroporated alar plates. This analysis confirms that GliA and Shh electroporations significantly modify the power law that governs the spatial organization of the mNEc and that this effect is higher in Shh electroporated alar plates, especially in the tectal region of the partially ventralized DMBs.