CNKSR2, a downstream mediator of retinoic acid signaling, modulates the Ras/Raf/MEK pathway to regulate patterning and invagination of the chick forebrain roof plate Free

During the development of the forebrain, the invagination of the roof plate results in the generation of the two cerebral hemispheres from a single telencephalic vesicle. Defects in the invagination of the roof plate often lead to congenital disorders such as middle interhemispheric holoprosencephaly (MIH-HPE) in humans, which causes seizures and intellectual disabilities (Roessler and Muenke, 2010; Weiss et al., 2018). Several genes encoding signaling molecules or transcription factors have been implicated in the etiology of HPE (Dubourg et al., 2007; Dupe et al., 2011; Geng and Oliver, 2009; Ishiguro et al., 2018; Keaton et al., 2010; Krauss, 2007; Nanni et al., 1999; Petryk et al., 2015; Ribeiro et al., 2006). Despite this, there is a lack of comprehensive knowledge about the molecular mechanisms and downstream targets through which these genes regulate the process of separation of the cerebral hemispheres.

We observed that, in the chick forebrain, the roof plate invaginates as a W-shaped structure (Fig. 1J). Previously, we identified RA signaling as one of the regulators of the invagination of the chick forebrain roof plate and showed that the kink or loop of the W expresses Bmp7, while the lateral arms express Wnt7b, Zic2 and Otx2. We have also reported that the inhibition of RA signaling leads to a flattened forebrain roof plate, which is similar to the phenotype observed in HPE (Gupta and Sen, 2015). One major question emerging from this was: how does RA signaling regulate forebrain roof plate invagination in the chick embryo? The answer to this question would shed light on the etiology of MIH-HPE.

In this study, we identify CNKSR2 (connector enhancer kinase suppressor of Ras 2) as being expressed in a similar domain to RA signaling in the forebrain roof plate midline. Therefore, we investigated CNKSR2 as a novel downstream mediator of RA signaling during forebrain roof plate invagination. The expression of CNKSR2 in the developing chick forebrain was first reported in a genome-wide screen examining the expression patterns of metabolism-related genes during embryonic development (Roy et al., 2013). Before this, CNKSR2 was identified as a scaffold protein and a modulator of Ras/Raf/MEK signaling, through a mutagenesis screen in Drosophila melanogaster (Therrien et al., 1998). In humans, mutations in CNKSR2 are associated with seizures, attention, language deficits and X-linked neurodevelopmental disorders (Damiano et al., 2017; Daoqi et al., 2020; Houge et al., 2011; Sun et al., 2018; Vaags et al., 2014), and neurodevelopmental delay (Higa et al., 2021). Recently, loss of CNKSR2 in male mice has been shown to cause behavioral changes, with increased neural activity and alterations in the synaptic proteome (Erata et al., 2021). This underscores the importance of understanding the role of CNKSR2 and the underlying process(s) that are affected by its knockdown during vertebrate nervous system development.

To characterize the function of CNKSR2 in the chick forebrain roof plate, we employed a RNAi-based method to knockdown the transcript. We observed that loss-of-function of CNKSR2 leads to invagination defects, with the appearance of increased cell proliferation and an increase in phosphorylated MEK1/2 (pMEK1/2)-positive cells in the roof plate midline. Furthermore, knockdown of CNKSR2 was observed to affect the patterning of the invaginating forebrain roof plate through the alteration of the expression of roof plate markers such as Bmp7, Wnt7b, Zic2 and Otx2. Thus, our results show that CNKSR2, when acting downstream of RA signaling, is a regulator of both patterning and invagination of the forebrain roof plate through its modulation of Ras/Raf/MEK signaling.

A genome-wide expression screen for metabolism-related genes (MRGs), based on whole-mount RNA in situ hybridization of chick embryos, identified CNKSR2 as one of the genes expressed in the forebrain (Roy et al., 2013). Our attention was specifically drawn to CNKSR2, as its expression domain appeared to coincide with the domain of active RA signaling in the forebrain roof plate midline, which has previously been reported to regulate forebrain roof plate invagination (Gupta and Sen, 2015). Thus, to investigate further, we performed a spatiotemporal expression analysis of CNKSR2 by RNA in situ hybridization. The CNKSR2 transcripts could be first detected at low levels in the midline of the chick forebrain at stages HH16 (Hamburger and Hamilton, 1951) and HH18 (Fig. 1A,B). At stages HH21-HH24, it was robustly expressed in the midline (loop of W) of the invaginating roof plate (Fig. 1C-E). From the spatiotemporal expression profiling, it appeared that the expression of CNKSR2 overlapped the domain of active RA signaling (Gupta and Sen, 2015). To confirm this, we electroporated a reporter construct for RA signaling (pRARE-AP) in the dorsal roof plate at HH10 and analyzed its expression at HH18, HH21 and HH23 (Fig. 1F). Alkaline phosphatase activity, indicating active RA signaling, was detected to be in a domain overlapping the expression of CNKSR2 in the roof plate at these stages (Fig. 1G-Jg-i). Based on this, we hypothesized that the expression of CNKSR2 is likely to be regulated by RA signaling in the context of the invaginating forebrain roof plate of the chick embryo.

We performed gain- and loss-of-function studies of RA signaling to determine whether the expression of CNKSR2 is indeed regulated by RA signaling. The constitutively active RA receptor pCIG-VP16RARα-IRES-GFP (Gupta and Sen, 2015) was electroporated in the lateral chick forebrain at HH18 to activate RA signaling ectopically (Fig. 2A). At HH23, GFP⁺ cells indicate the region of electroporation and the cells that expressed the constitutively active RA receptor. RNA in situ hybridization for CNKSR2 revealed that it is now expressed ectopically in the lateral forebrain, in addition to its normal domain of expression in the roof plate midline (Fig. 2C-C″), when compared to control electroporated lateral forebrain (Fig. 2B-B″). This demonstrated that activation of ectopic RA signaling in the chick forebrain is sufficient for inducing the expression of CNKSR2 in the lateral chick forebrain.

To create loss of function of RA signaling in the forebrain roof plate, the dominant-negative RA receptor, pCAGEN-RAR403 (Gupta and Sen, 2015), was electroporated in the forebrain roof plate at HH10 (Fig. 2D). We observed that the loss of RA signaling in the invaginating roof plate leads to the absence of CNKSR2 transcript, as detected by RNA in situ hybridization (Fig. 2G-G″,g, n=5), in comparison to control, pCAG-GFP (Fig. 2F-F″,f, n=5), suggesting that CNKSR2 is regulated by RA signaling.

To investigate whether CNKSR2 acts as the downstream mediator of RA signaling to regulate the invagination of the forebrain roof plate, we knocked down the transcript of CNKSR2 using a RNAi-based approach. Two RNAi constructs were designed to target CNKSR2 RNA and were cloned into the pRmiR vector (Smith et al., 2009). The efficiency of the knockdown was tested through an in vitro sensor assay (Sindhu et al., 2019) (Fig. S1). One of these constructs, CNKSR2-RNAi (1), was the most efficient in knocking down the CNKSR2 transcript. This construct was used for subsequent analyses and is referred to as CNKSR2-RNAi. The knockdown construct was electroporated into the forebrain roof plate at HH10 and harvested at HH23, the stage where invagination defects can be best scored. We found a dramatic decrease in expression of CNKSR2 in the CNKSR2-RNAi electroporated forebrains (Fig. 3C-E‴) compared with the control (control pRmiR-empty) electroporated forebrains (Fig. 3B-B‴), indicating efficient knockdown (Fig. 3A).

Knockdown of CNKSR2 in the forebrain roof plate led to varying degrees of defects in invagination. In control embryos, the forebrain roof plate was observed to invaginate as a W-shaped structure with the expression of CNKSR2 restricted to the middle loop of the W (Fig. 3B-B‴). However, upon knockdown of CNKSR2, depending on the extent of electroporation, the morphology of the invaginating forebrain roof plate was altered. In the embryos with a relatively higher extent of electroporation, we observed a nearly flattened roof plate (Fig. 3C-C‴, n=13/29) which closely resembled the phenotype observed upon inhibition of RA signaling (Gupta and Sen, 2015). Furthermore, in the embryos with a moderate extent of electroporation, we observed the invagination to be U/V-shaped instead of a W-shaped structure (Fig. 3D-E‴, n=16/29). Quantification of the percentages of occurrences of the phenotypes upon knockdown of CNKSR2 showed a higher occurrence of the U/V roof plate than the flattened roof plate (Fig. 3F). However, we observed no change in RA signaling in the forebrains with the knockdown of CNKSR2 (Fig. S2, n=3). In addition, analysis of apoptotic cells by TUNEL assay at HH18 (Fig. S3A-B″,E) and HH23 upon knockdown of CNKSR2 in the roof plate did not show any significant statistical difference (Fig. S3C-D″,F). This observation rules out the role of cell death as a contributor to the invagination defects seen upon knockdown of CNKSR2.

The inhibition of RA signaling led to an increase in cell proliferation in the forebrain roof plate midline (Gupta and Sen, 2015). To determine whether CNKSR2 might be functioning downstream of RA signaling to regulate cell proliferation in the forebrain roof plate, we knocked down the CNKSR2 mRNA, followed by immunostaining for phospho-histone 3 (PH3) (Fig. 4A). We observed that, in the control, pRmiR-empty electroporated forebrains, the PH3-positive cells were higher in the lateral arms of the W and low in the middle loop (Fig. 4B-B″), whereas in the CNKSR2-RNAi electroporated forebrains, the distribution of the PH3-positive cells was uniformly high throughout the roof plate (Fig. 4C-C″). In addition, quantification of the percentage of PH3-positive cells in the roof plate midline of the test and control electroporated forebrains showed a statistically significant difference (n=5 each for test and control) (Fig. 4D).

To further confirm whether CNKSR2 regulates cell proliferation in the forebrain roof plate, we used EdU labeling to detect proliferating cells after knockdown of CNKSR2, according to the strategy described in Fig. 4E. We observed that, although in the control there were very few EdU-positive cells in the middle of the W when compared with the lateral sides (Fig. 4F-G″), in the test (CNKSR2-RNAi) a relatively higher number of EdU-positive cells was observed throughout the roof plate (Fig. 4H-I″). Quantification of the percentage of EdU-positive cells in the test and control electroporated forebrains showed a statistically significant difference between the two (Fig. 4J, n=5 each for test and control). These experiments suggested that cell proliferation in the forebrain roof plate is regulated by the CNKSR2.

Because we observed a change in cell proliferation upon knockdown of CNKSR2, we wanted to determine whether ectopic expression of CNKSR2 would be sufficient for regulating cell proliferation. For this, we expressed mouse CNKSR2 (mCNKSR2) by electroporating pCAG-mCNKSR2 in the lateral region of the chick forebrain at HH18, followed by analysis at HH23 (Fig. S4A). The expression of mCNKSR2 in the electroporated samples was assessed by RNA in situ hybridization with a probe specific for mCNKSR2 (Fig. S4B-C″, n=3). However, ectopic expression of mCNKSR2 in the lateral chick forebrain did not result in a decrease in cell proliferation in this region (Fig. S4D-F, n=4, each for test and control).

We found that the misexpression of mCNKSR2 did not alter cell proliferation in the lateral forebrain; however, changes in cell proliferation were observed in the midline upon knockdown of CNKSR2. Thus, we reasoned that changes in cell proliferation observed in the midline could be an indirect outcome of patterning changes in the forebrain roof plate. We have previously observed that inhibition of RA signaling in the forebrain roof plate leads to patterning defects with changes in expression of the roof plate markers Bmp7, Wnt7b, Zic2 and Otx2. For example, there was a loss of expression of Bmp7 and a shortening of the expression domains of Wnt7b, Zic2 and Otx2 upon inhibition of RA signaling (Gupta and Sen, 2015). Thus, we investigated whether the knockdown of CNKSR2 affects the expression of these markers similarly.

In the control forebrains, some of these markers exhibit unique expression patterns within the roof plate (Fig. 5A). For example, Bmp7 expression was found to be restricted to the middle loop of the W (Fig. 5B-B″), while Wnt7b (Fig. 5D-D″) was expressed at higher levels in the lateral arms of the W (black arrowheads in Fig. 5D″) when compared with the middle loop (red arrowheads in Fig. 5D″). On the other hand, Zic2 and Otx2 exhibited relatively uniform expression across the W-shaped roof plate (Fig. 5F-F″,H-H″). Upon knockdown of CNKSR2, we observed that there was a dramatic decrease in the expression of Bmp7 (Fig. 5C-C″). Similarly, the midline expression domain of Wnt7b was absent, with only expression in the lateral arms of the W remaining (Fig. 5E-E″). On examining the expression of Zic2 (Fig. 5G-G″, the extent of expression marked by the black arrowheads) and Otx2 (Fig. 5I-I″, the extent of expression marked by the black arrowheads), we found a shortening of the domains of expression of these two markers (Fig. 5J). These changes in expression of Bmp7, Wnt7b, Zic2 and Otx2 were remarkably similar, if not identical, to those observed upon inhibition of RA signaling (Gupta and Sen, 2015). Thus, the changes in expression of roof plate markers may be interpreted as altered patterning, as the morphology of the roof plate itself was altered from a W-shaped invagination to either a flattened or U/V-shaped invagination, with the middle loop of the W missing. Furthermore, we observed that ectopic expression of mCNKSR2 in the lateral forebrain (Fig. 5K) was sufficient for inducing the expression of Bmp7 (Fig. 5L-M″, n=5 each for control and test), Zic2 (Fig. S5A-C″), Wnt7b (Fig. S5D-E″) and Otx2 (Fig. S5F-G″, n=3 each for control and test) in this ectopic location.

CNKSR2 has been reported to modulate the Ras/Raf/MEK signaling (Sundaram and Han, 1995; Therrien et al., 1998); hence, we decided to investigate whether CNKSR2 functions similarly in the forebrain roof plate. To detect the status of Ras/Raf/MEK signaling, we performed immunostaining for pMEK1/2. We observed that, in the forebrains electroporated with control pRmiR-empty, pMEK1/2-positive cells were low in the midline of the forebrain roof plate but present uniformly in the ventricular zone that flanks the invaginating roof plate (Fig. 6A-A″). However, in the embryos electroporated with CNKSR2-RNAi, we found a statistically significant increase in pMEK 1/2-positive cells in the roof plate midline (Fig. 6B-B″,C, n=5 each for test and control).

As the knockdown of CNKSR2 increased the number of pMEK1/2-positive cells in the roof plate, we wanted to determine whether modulation of the Ras/Raf/MEK signaling could rescue the defects observed in the forebrain roof plate. To do this, we co-electroporated pCIG-MKK1dn (pCIG-MKK1dn-IRES-GFP; Delfini et al., 2005) together with the CNKSR2-RNAi in the HH10 chick forebrain and compared it with the forebrains electroporated with the controls pCAG-GFP, CNKSR2-RNAi and pCIG-MKK1dn, and assessed the rescue of the invagination defect by observing the morphology. For rescue of the patterning defects, we examined the expression of the midline marker Bmp7; for changes in cell proliferation, we performed PH3 immunostaining (Fig. 6D-H‴, n=4 for control and test). We found that in the forebrains electroporated with pCIG-MKK1dn and CNKSR2-RNAi+MKK1dn, the invagination of the roof plate, the reduced cell proliferation in the midline (Fig. 6G-G″,H-H″) and the expression of the midline marker Bmp7 (Fig. 6G‴,H‴) were comparable with that of the control (electroporated with pCAG-GFP) (Fig. 6E-E‴), indicating a rescue. Conversely, the forebrains electroporated with CNKSR2-RNAi showed invagination defects, increased cell proliferation and reduced Bmp7 expression (Fig. 6F-F‴). We assessed for knockdown of CNKSR2 by RNA in situ hybridization in the samples co-electroporated with CNKSR2-RNAi and MKK1-dn, and found there was knockdown and rescue due to expression of MKK1-dn (Fig. S6).

To determine whether the phenotype observed after knockdown of CNKSR2 was not due to off-target effects, we attempted to rescue the knockdown by expressing mCNKSR2 in the forebrain roof plate. Forebrains co-electroporated with CNKSR2-RNAi and mCNKSR2 showed partial rescue in terms of invagination, cell proliferation (n=4) and Bmp7 expression (n=3) (Fig. 6J-J‴) when compared with forebrains electroporated with mCNKSR2 alone (Fig. 6I-I‴, n=4 for PH3 and n=3 for Bmp7).

With respect to cell proliferation, a partial rescue was observed with co-electroporation of CNKSR2-RNAi and mCNKSR2 as this lowered the percentage of PH3-positive cells in the midline when compared with that of the forebrain midline electroporated with CNKSR2-RNAi alone. In fact, the lower percentage of PH3-positive cells in the roof plate midline was similar to that observed in forebrains electroporated with pCAG-GFP, MKK1-dn, CNKSR2-RNAi+MKK1dn and mCNKSR2 (Fig. 6K).

However, we observed that misexpression of MKK1-dn in the lateral chick forebrain did not alter the number of PH3-positive cells compared with control (Fig. S7A-B″), but interestingly was sufficient to induce the expression of roof plate midline marker Bmp7 (Fig. S7D-D″) in this ectopic location compared with control (Fig. S7C-C″). Thus, these experiments indicate that CNKSR2 modulates Ras/Raf/MEK signaling in the forebrain roof plate, which appears to be necessary for its invagination.

Because loss of RA signaling leads to invagination defects (Gupta and Sen, 2015), and in this study we have established that CNKSR2 is a downstream target of RA signaling in this context (Fig. 2), we wanted to investigate whether mCNKSR2 can rescue the forebrain midline invagination defects observed upon inhibition of RA signaling. We observed that simultaneous loss of RA signaling by co-electroporation of pCAGEN-RAR403 (dominant-negative RA receptor) and pCAG-mCNKSR2 did not rescue the invagination defect seen upon loss of RA signaling (Fig. S8D-E′, n=3 for each). The presence of mCNKSR2 was assayed by RNA in situ hybridization (Fig. S8A-C′). These experiments suggest that CNKSR2 may not be the sole downstream mediator of RA signaling in this context, and there may be other as yet unidentified targets that regulate the process of midline invagination in the chick forebrain.

Furthermore, we observed that constitutive activation of Ras/Raf/MEK signaling in the forebrain roof plate through the expression of HRasV12 (electroporation of pcDNA3-HRasV12) leads to invagination defects (Fig. S9B-B″) when compared with the control (pCAG-GFP) electroporated forebrain (Fig. S9A-A″), suggesting a crucial role of this pathway in forebrain roof plate morphogenesis.

In this study, we have compared a comprehensive spatiotemporal expression profiling of CNKSR2 with that of active RA signaling at stages HH18, HH21 and HH23, and found that the two domains overlap to a large extent. In addition, an in silico analysis of a 2 kb upstream region of chicken CNKSR2 showed the presence of the consensus half-sites [PuG(G/T)TCA] for RA signaling (see Fig. S10). This supports our hypothesis that RA signaling regulates the expression of CNKSR2 in this context. Furthermore, we observed an alteration in the expression of CNKSR2 upon manipulation of RA signaling through loss- and gain-of-function experiments. Taken together, this led us to conclude that CNKSR2 is a downstream mediator of RA signaling in the chick forebrain roof plate.

We observed that the knockdown of CNKSR2 in the forebrain roof plate leads to invagination defects. Whenever the electroporation of the RNAi construct spanned a broad domain, an almost flattened roof plate was observed. This prompted us to determine the underlying causes of the invagination defects upon knockdown of CNKSR2. We reasoned that one likely cause for this could be an abolition of the differences in cell proliferation that normally exists in the roof plate. The rationale behind this is as follows: (1) we have previously shown that RA signaling regulates cell proliferation in the forebrain roof plate (Gupta and Sen, 2015); (2) evidence from the literature suggests that RA signaling can regulate the mitogen-activated protein kinase (MAPK) pathway (Crowe et al., 2003; Nakagawa et al., 2003; Wang and Yen, 2008), which in turn is known to regulate cell proliferation; and (3) CNKSR2 is a scaffolding protein and has been shown to modulate the Ras/Raf/MEK pathway (Sundaram and Han, 1995; Therrien et al., 1998). We thus hypothesized that RA signaling regulates cell proliferation through CNKSR2 and its modulation of Ras/Raf/MEK signaling.

When we assessed cell proliferation and the status of Ras/Raf/MEK signaling upon the knockdown of CNKSR2 in the forebrain roof plate, we found an increase in cell proliferation and pMEK1/2-positive cells in the midline. Interestingly, misexpression of mouse CNKSR2 in the lateral forebrain was not sufficient to decrease cell proliferation in the ectopic location. Possible explanations for this observation are: (1) the context in the lateral forebrain is different from that of the roof plate with respect to the regulation of cell proliferation; (2) different levels of CNKSR2 might be required to effect the desired changes (Peyssonnaux and Eychène, 2001); and/or (3) there are other downstream effector(s) of RA signaling that work in combination with CNKSR2 to regulate cell proliferation in the forebrain roof plate (Ito et al., 2021).

It is possible that patterning of the forebrain roof plate results in the differential cell proliferation observed across this region. Thus, the likely cause of the increase in cell proliferation and pMEK1/2-positive cells upon knockdown of CNKSR2 could be a result of altered patterning of the invaginating forebrain roof plate. This was supported by the observed changes in expression of the markers Bmp7, Wnt7b, Zic2 and Otx2 upon knockdown of CNKSR2. We hypothesized that CNKSR2 modulates the levels of the Ras/Raf/MEK signaling and thereby patterns the roof plate (expression of markers), which in turn drives its invagination. In support of this, we observed that co-electroporation of CNKSR2-RNAi with dominant-negative MKK1 not only rescues the invagination defect and the normal expression pattern of the midline marker Bmp7, but also restores the domain of reduced Ras/Raf/ MEK signaling and reduced cell proliferation in the midline. Interestingly, the inhibition of Ras/Raf/MEK signaling in the lateral forebrain was sufficient for the expression of the midline marker Bmp7 but not sufficient for reducing cell proliferation. This suggests that differences exist between the context of the roof plate and the lateral forebrain, and that other factors, besides Ras/Raf/MEK signaling (which functions downstream of RA and/or CNKSR2) are likely to be important for the reduction in cell proliferation observed in the roof plate midline.

In the zebrafish model for HPE, it has been demonstrated that tgif1, one of the genes linked to HPE, is involved in regulating RA signaling during early patterning of the forebrain. Additionally, the loss of function of Cyp26a1, a retinoic acid-degrading enzyme, in the forebrain phenocopies the tgif1 morphants (Gongal and Waskiewicz, 2008). In addition, various studies in quails (Maden et al., 1996), swine (Hale, 1933), mice (Halilagic et al., 2007; Ribes et al., 2006), chick and zebrafish (Begemann et al., 2001; Maves and Kimmel, 2005) show that either lowering or increasing the levels of RA signaling can lead to forebrain defects, which includes HPE-like phenotypes (Gongal et al., 2011).

The forebrains of the RA signaling reporter (RARE-LacZ) mice do not show the presence of any RA signaling in the forebrain roof plate midline (Rossant et al., 1991). Moreover, mice lacking sources of RA (Mic et al., 2004; Molotkova et al., 2007) show normal roof plate invagination, unlike our observations of the chick forebrain (Gupta and Sen, 2015). In addition, there are no reports of CNKSR2 expression in the forebrain roof plate midline in mice. Taken together, this indicates that the RA signaling-CNKSR2 axis might not be involved in regulation of forebrain roof plate invagination in mice. However, it is possible that mice may have diverged during evolution, and the role of RA signaling and CNKSR2 in regulating forebrain roof plate may be conserved in humans. To date, there is no study suggesting a link between the loss of function of CNKSR2 and HPE in humans, although the loss of CNKSR2 is attributed to phenotypes such as microcephaly, cortical atrophy and intellectual disability (Polla et al., 2019). There is a possibility that CNKSR2 could be acting through a non-cell-autonomous manner on the process of midline invagination, probably through the action of other signaling pathways present in this region, such as Bmp signaling and Wnt signaling (Furuta et al., 1997; Lee et al., 2000). However, an in-depth analysis into this aspect is required to confirm this possibility. In total, there is a need for further investigation to establish a link between RA signaling and/or CNKSR2, and HPE.

In conclusion, our results reveal that the phenotypes caused by the loss of function of CNKSR2 in the chick forebrain roof plate are like those observed with the loss of RA signaling. We postulate that RA and/or CNKSR2 regulates roof plate patterning through modulation of Ras/Raf/MEK signaling, which in turn drives the process of roof plate invagination (Fig. 7). Moreover, our study is the first to identify CNKSR2 as a downstream mediator of RA signaling, opening the possibility that this mechanism of regulating morphogenesis may exist in other contexts besides the chick forebrain roof plate.

Fertilized White leghorn (Gallus gallus) eggs were obtained from the following sources: (1) Central Avian Research Institute (CARI), Bareilly, Uttar Pradesh, India; and (2) Ganesh Enterprises, Nankari Village, Kanpur, Uttar Pradesh, India. The eggs were incubated at 38°C in a humidified incubator until the desired stages. The embryos were staged according to Hamburger and Hamilton (Hamburger and Hamilton, 1951).

Heads from unmanipulated and electroporated chicken embryos were harvested at the desired stage and fixed in 4% PFA (paraformaldehyde) overnight. The heads were cryoprotected by passing through a gradient of 15% sucrose and 30% sucrose prepared in PBS (phosphate-buffered saline). The heads were then embedded in Polyfreeze/OCT (Sigma, SHH0026) and 10 µm coronal sections were generated using a cryostat (Leica CM1850).

pCIG-VP16RARα-IRES-nGFP was a gift from Thomas Jessell (Addgene plasmid #16287) (Novitch et al., 2003). pCAGEN-RAR403 and pCAG-GFP were gifts from Prof. Constance Cepko (Harvard Medical School, Boston, MA, USA). cDNA clones for Bmp7 and Wnt7b were gifts from Prof. Clifford Tabin (Harvard Medical School, Boston, MA, USA). pRARE-AP, cDNA for Otx2 and Zic2 (RT-PCR amplified) constructs have been described previously (Gupta and Sen, 2015). The cDNA for CNKSR2 (ChEST 713i22) was obtained from the chicken EST collection from Source Bioscience (previously MRC Geneservice, UK). The top and bottom strands of the RNAi-containing region of CNKSR2 (CNKSR2-RNAi) were synthesized (Macrogen), annealed and cloned into the pRmiR vector after BsaI digestion (Smith et al., 2009): top strand sequence, 5′-TGCTGTGTACAGGCAACAGACTAAAGGTTTTGGCCACTGACTGACCTTTAGTCTTGCCTGTACA-3′; bottom strand sequence, 5′-CCTGTGTACAGGCAAGACTAAAGGTCAGTCAGTGGCCAAAACCTTTAGTCTGTTGCCTGTACAC-3′ (bold indicates the region in CNKSR2 mRNA that is targeted by the designed RNAi construct).

Sensor-CNKSR2 was generated by first excising the putative target region from EST using the enzymes NotI and HindIII, and then ligating it with the pCAG-mCherry vector (a gift from Prof. Constance Cepko) between NotI and HindIII restriction sites. pCAG-mCNKSR2 was generated by amplifying the MGC clone (ThermoFisher Scientific, USA) for CNKSR2 (IRAV-158g9) with the primers (forward) 5′-CTAGTCTAGAATGGCTCTGATAATGGAAC-3′ with XbaI site and (reverse) 5′-CGAATGCGGCCGCTTACACATGTGTCTCAATG-3′ with the NotI site, and cloned into the pCAG-MCS backbone digested with XbaI and NotI. TA-mCNKSR2-CDS was generated by PCR amplifying the MGC clone (ThermoFisher Scientific) for CNKSR2 (IRAV-158g9) with the primers (forward) 5′-GGGTTCAACCATTGCTGTCT-3′ and (reverse) 5′-TTACACATGTGTCTCAATGTATG-3′, and cloned into the TA vector [ThermoFisher Scientific, K207020 (previously 45-0007LT)] according to the manufacturer's protocol.

pCIG-MKK1dn was a gift from Prof. Olivier Pourquié (Harvard Medical School, Boston, MA, USA). pcDNA3-HRasV12 was a gift from Julian Downward (Addgene plasmid #39504) (Rodriguez-Viciana et al., 1997).

Human embryonic kidney (HEK-293T, ATCC CRL-3216) fibroblast cell lines were cultured in Dulbecco's Modified Eagle's Medium (DMEM) (Sigma, D7777) with 10% fetal bovine serum.

Human embryonic kidney fibroblast 293T cells at 60-70% confluency was transfected using Turbofect (ThermoFisher Scientific, R0531) according to the manufacturer's protocol, with pRmiR vector(s) and Sensor-CNKSR2 in a 6:1 molar ratio. GFP and mCherry fluorescence were observed after 72 h.

After incubating the eggs for 24 h, 2-3 ml of albumin was taken out to lower the developing chick embryo. A window was then created on the top of the egg and DNA constructs (1 µg/µl-2 µg/µl) with 0.1% of Fast Green (Sigma-Aldrich, 44715) were injected into the forebrain vesicle at HH10 and HH18. Platinum hockey stick electrodes (NepaGene, CUY611P3-1) were positioned on this forebrain vesicle such that the positive electrode was placed over the dorsal forebrain, while the negative electrode was placed below the yolk of the egg at HH10 and HH18, and the electrodes were placed close to each other. Five electric pulses of 15 V each for embryos at HH10 and HH18 were applied for the 50 ms duration with a gap of 950 ms, using a square wave pulse electroporator (Nepa Gene, CUY21SC). Sterile PBS containing penicillin and streptomycin was layered over the embryos and the egg was sealed with tape. Eggs were put back into the egg incubator until the desired HH stage of development and then harvested for tissue section preparation.

The slides were fixed for 5 min in 4% PFA, followed by the quenching of endogenous alkaline phosphatase (AP) by incubating the slides in PBS at 65°C for 1 h. The slides were then equilibrated in the NTM buffer [100 mM Tris (pH 9.5), 100 mM NaCl, 50 mM MgCl₂ and 0.1% Tween 20] for 10 min and then incubated in the NTM buffer containing the substrates BCIP and NBT. The reaction was monitored intermittently for the development of colored precipitate and stopped by adding 4% PFA. Image acquisition was performed using a Leica DM500B microscope equipped with a DFC500 camera.

RNA hybridization was performed on coronal sections of embryonic chick forebrain according to established protocols (Trimarchi et al., 2007). The digoxigenin-labeled riboprobes for RNA in situ hybridization were generated from the cDNA clones mentioned in the previous section by in vitro transcription.

We used an established protocol for immunostaining. Briefly, after a 5 min fixation in 4% PFA, three washes with PBS were carried out followed by 1 h of blocking with 5% of heat-inactivated goat serum (HINGS) in PBS. After that, the primary antibody at the required dilution was layered on the slide in the blocking reagent and incubated overnight at 4°C. The primary antibodies were used at the following concentrations: anti-pMEK1/2 antibody, 1:200 (CST, 9121), anti-PH3 antibody, 1:1000 (Sigma-Aldrich, H0412) and anti-GFP 1:300 (ThermoFisher Scientific, A-6455). Slides were then washed three times in PBS for 5 min each and the secondary antibody (1:250) was layered on the slides followed by incubation for 2 h at room temperature. The secondary antibodies used were as follows: Alexa Fluor 488-conjugated anti-rabbit IgG (Jackson ImmunoResearch Laboratories, 111-545-003) and Alexa Fluor 594-conjugated anti-rabbit IgG (Jackson ImmunoResearch Laboratories, 111-585-045).

For EdU detection, fertilized chicken eggs were electroporated with either pRmiR-empty (control) or CNKSR2-RNAi (treatment) in the forebrain vesicle at HH10, incubated for the desired time and then overlaid with 25 μg in 200 μl PBS (0.5 mM of EdU) of the thymidine analog EdU (ThermoFisher Scientific, C10338) and harvested at HH23. Brains were harvested from embryos 2 h after EdU overlay, fixed in 4% PFA, dehydrated in 15% and 30% sucrose in PBS, and embedded in OCT. 10 µm frozen sections from these brains were processed for EdU detection using the CLICK-IT EdU-labeling kit (ThermoFisher Scientific, C10338) according to the manufacturer's protocol.

TUNEL assay was performed on either pRmiR-empty or CNKSR2-RNAi electroporated forebrains at HH18 or HH23, according to the manufacturer's protocol (Roche, 12156792910).

A Leica stereomicroscope (DM500B) equipped with a DFC500 camera was used to capture RNA in situ hybridization images and fluorescent images.

All control electroporated embryos were considered normal as the roof plate invaginated in a W-shape. DAPI stained CNKSR2-RNAi electroporated forebrain sections were analyzed for either a flat or a U/V-shaped roof plate phenotype, solely based on morphology (n=29 embryos) using Photoshop CS6. Out of these 29 embryos, 16 embryos showed U/V-shaped invagination (55.17%), while the remaining 13 embryos (44.83%) showed a flat roof plate phenotype. Both the values were then plotted as a pie chart using GraphPad 8.0.2 (v.263) software.

A rectangle of 60×75 µm² was drawn in the 20× magnification of the image, and the number of double-positive cells (red, immunostaining; green, GFP) was counted manually in Photoshop CS6. The number of DAPI cells was counted using ImageJ software and the percentage was calculated in three sections each of five embryos and averaged (n=5 each for test and control). As we were assessing the differences between two groups, control and test, we performed statistical analyses using an unpaired t-test and then plotted the graphs using GraphPad 8.0.2 (v.263) software in Figs 4D,H and 6C. For Fig. 6K, statistical analyses were performed using one-way ANOVA with Sidak's multiple comparison test using GraphPad 8.0.2 (v.263) software to compare means between pCAG-GFP versus CNKSR2-RNAi, MKK1-dn-IRES-GFP versus CNKSR2-RNAi+MKK1-dn, mCNKSR2 versus CNKSR2-RNAi+mCNKSR2,CNKSR2-RNAi versus CNKSR2-RNAi+MKK1-dn, and CNKSR2-RNAi versus CNKSR2-RNAi+mCNKSR2. The values are represented as bar graphs, plotted using GraphPad 8.0.2 (v.263) software.

A rectangle of 13×44 µm² was drawn in the 20× magnification of the image and the number of double-positive (red, immunostaining; green, GFP) cells was counted manually in Photoshop CS6. The number of DAPI cells was counted using ImageJ software and the percentages of TUNEL- and PH3-positive cells were calculated in two or three sections each of three embryos and averaged. As we were assessing the differences between two groups, control and test, we performed statistical analyses using an unpaired t-test and then plotted the graphs using GraphPad 8.0.2 (v.263) software.

The expression domains of Zic2 and Otx2 were quantified by ImageJ on sections of four embryos electroporated with either control or CNKSR2-RNAi. The values were averaged and plotted as bar graphs using GraphPad 8.0.2 (v.263) software. Statistical analysis was performed using an unpaired t-test and significance was obtained.

The authors acknowledge Prof. C. Cepko (Harvard Medical School, Boston, MA, USA) for constructs related to RA signaling. The authors also acknowledge Prof. Olivier Pourquié (Harvard Medical School, Boston, MA, USA) for the pCIG-MKK1dn construct. The authors acknowledge the technical assistance of Ms Shruti Jain, Mr Kaustuv Ghosh and Ms Archita Mishra with this study. The authors thank Prof. Amitabha Bandyopadhyay, Dr Sandeep Gupta, Dr Suvimal Kumar Sindhu and Dr Tathagata Biswas for their critical comments on this manuscript. The authors thank Mr Naresh Gupta for his technical support with this study.

The peer review history is available online at https://journals.biologists.com/dev/lookup/doi/10.1242/dev.200857.reviewer-comments.pdf