Cdx1 and Gsc distinctly regulate the transcription of BMP4 target gene ventx3.2 by directly binding to the proximal promoter region in Xenopus gastrulae

A comprehensive regulatory network of transcription factors controls the dorsoventral patterning of the body axis in developing vertebrate embryos. Bone morphogenetic protein signaling is essential for activating the Ventx family of homeodomain transcription factors, which regulates embryonic patterning and germ layer identity during Xenopus gastrulation. Although Ventx1.1 and Ventx2.1 of the Xenopus Ventx family have been extensively investigated, Ventx3.2 remains largely understudied. Therefore, this study aimed to investigate the transcriptional regulation of ventx3.2 during the embryonic development of Xenopus. We used goosecoid (Gsc) genome-wide chromatin immunoprecipitation-sequencing data to isolate and replicate the promoter region of ventx3.2. Serial deletion and site-directed mutagenesis were used to identify the cis-acting elements for Gsc and caudal type homeobox 1 (Cdx1) within the ventx3.2 promoter. Cdx1 and Gsc differentially regulated ventx3.2 transcription in this study. Additionally, positive cis-acting and negative response elements were observed for Cdx1 and Gsc, respectively, within the 5′ flanking region of the ventx3.2 promoter. This result was corroborated by mapping the active Cdx1 response element (CRE) and Gsc response element (GRE). Moreover, a point mutation within the CRE and GRE completely abolished the activator and repressive activities of Cdx1 and Gsc, respectively. Furthermore, the chromatin immunoprecipitation-polymerase chain reaction confirmed the direct binding of Cdx1 and Gsc to the CRE and GRE, respectively. Inhibition of Cdx1 and Gsc activities at their respective functional regions, namely, the ventral marginal zone and dorsal marginal zone, reversed their effects on ventx3.2 transcription. These results indicate that Cdx1 and Gsc modulate ventx3.2 transcription in the ventral marginal zone and dorsal marginal zone by directly binding to the promoter region during Xenopus gastrulation.


INTRODUCTION
Bone morphogenetic protein 4 (BMP4), which belongs to the transforming growth factor β family, plays numerous roles in embryonic development.It regulates the initial development of vertebrates by acting as an early morphogen that determines ectoderm and ventral mesoderm differentiation and dorsoventral (DV) patterning (Dosch et al., 1997;Tucker et al., 2008;Wills et al., 2008).This protein directly activates the expression of ventx family members, namely, ventx1.1, ventx1.2, ventx2.1, ventx2.2, ventx3.1, and ventx3.2 (Henningfeld et al., 2002;Lee et al., 2011;Shapira et al., 1999;Kumar et al., 2022).Given its association with changes in the developmental fate of embryos, Ventx1.1 is considered a major downstream target of BMP4.A ventralized Xenopus embryo is produced via the ectopic expression of Ventx1.1 (Hwang et al., 2002b).In contrast, the antimorphic ventx1.1 construct promotes neuralization and expands the expression domain of Spemann organizer-specific genes, such as Gsc and Chordin (Hwang et al., 2002b).Moreover, ventx2.1 regulates neural crest specification and ectomesenchyme formation in the ectoderm (Scerbo and Monsoro-Burq, 2020).In Xenopus embryos, morpholino-based ventx2.1 knockdown results in the development of a partially duplicated secondary axis and enlargement of the Spemann organizer (Onichtchouk et al., 1996(Onichtchouk et al., , 1998)).Our previous investigation on the evolution of the ventx family in Xenopus (Kumar et al., 2022) revealed that Ventx1.1 and Ventx2.1 have a shared ancestor.In contrast, Ventx3.1 and Ventx3.2 are the eISSN: 1016-8478 / © 2024 The Author(s).Published by Elsevier Inc. on behalf of Korean Society for Molecular and Cellular Biology.This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).most unique members of the Ventx cluster (Kumar et al., 2022).While the function of ventx3.1 remains unknown, limited information is available regarding that of ventx3.2.The BMP4/ Smad1 signaling pathway targets ventx3.2, a downstream gene that is specifically expressed in the ventral marginal zones (VMZ) and ectodermal area during the gastrulation of Xenopus embryos (Shapira et al., 1999(Shapira et al., , 2000)).Additionally, the ectopic expression of Ventx3.2 in Xenopus restores the dominant negative Bmp4 receptor-based attenuation of BMP4 signaling.This leads to the dose-dependent ventralization of embryos (Shapira et al., 2000) and recapitulation of ventx1.1 function during gastrulation.This suggests that Ventx3.2 can restore BMP4 function in developing Xenopus embryos, making it a crucial ventx family member.Research utilizing antimorphs has shown that the Ventx3.2antimorph can induce secondary axis formation, whereas partial loss of Ventx3.2 enhances the efficacy of secondary axis formation by the dominant negative Bmp receptor or Smad6 in Xenopus laevis (Shapira et al., 2000).These loss-of-function investigations show that Ventx3.2plays a major role in limiting the Spemann organizer.
In this study, we aimed to investigate the transcriptional regulation of ventx3.2during the embryonic development of Xenopus.We hypothesized that Gsc suppresses ventx3.2expression in the gastrula to preserve dorsal identity, whereas Cdx1 enhances ventx3.2transcription to retain the identity of the ventral mesoderm.Our findings provide novel insights into the complex molecular mechanisms underlying the regulation of ventx3.2transcription in the ventral and dorsal mesoderm during Xenopus gastrulation.

Ethics Statement
All animal experiments adhered to the regulations established by the Hallym University Institutional Animal Care and Use Committee, as documented in Hallym 2012Hallym -76, 2013Hallym -130, and 2019-79-79.Each member of our research team participated in educational and training workshops on the appropriate handling and utilization of experimental animals.Adult X. laevis were maintained by personnel approved for laboratory animal care in accordance with the guidelines of the Institute of Laboratory Animal Resources of Hallym University.They were housed in suitable containers with a 12-hour light and dark cycle (LD 12:12 h) at 18 °C.

Embryo Manipulation
The X. laevis used in this study were supplied by the Korean Xenopus Resource Center for Research.Embryos were generated using in vitro fertilization (IVF) after stimulating female frogs with 500 units of human chorionic gonadotropin purchased from LG (IVF-C 1000 international unit (IU)).DNA and messenger RNA (mRNA) constructs were injected into the animal poles of embryos at the 1-cell stage.Animal cap (AC) explants were extracted from embryos with or without injection at stage 8.These explants were then grown in L-15 medium until they reached the required stages for subsequent AC assays.
DNA and RNA Preparation pCS4-Myc-Cdx1 and pCS4-3Flag-Gsc mRNAs were created by linearizing the target vectors with the Not1 and Acc65I restriction enzymes, respectively.The linearized vectors (pCS4-Myc-Cdx1 and pCS4-3Flag-Gsc) were used during in vitro transcription tests, which were performed using the mMessage mMachine Kit (Ambion) according to the manufacturer's instructions.The synthesized mRNAs were quantified using Thermo-scientific Nanodrop One and then diluted in diethyl pyrocarbonate-treated water to achieve a final concentration of 1 ng/5 nL.The diluted mRNA was stored at −80 °C for injections.

Embryo Injection and Explant Culture
Oocytes were obtained by administering 500 units of human chorionic gonadotropin to female X. laevis (Sigma-Aldrich).The obtained eggs underwent in vitro fertilization, and embryos at the 1-cell stage were injected with DNA and messenger RNA (mRNA) at the animal pole, as described previously (Kumar et al., 2023).At the blastula stage (stage 8), the AC explants were dissected from the injected and non-injected embryos.Whole embryos (WEs) and dissected ACs were cultivated until stages 11 to 11.5 using 1× L-15 medium (Gibco/Thermo Fisher) and 30% Marc's Modified Ringer's solution, respectively.

Reverse Transcription-PCR
At the 1-cell stage, X. laevis embryos were injected into the animal pole containing 0.5 ng/embryo Gsc mRNA and 1 ng/ embryo Cdx1 mRNA.Subsequently, the ACs were isolated from injected and non-injected embryos and cultivated in 1× L-15 growth medium until they reached stages 11 to 11.5, as described previously (Kumar et al., 2019).Total RNA was extracted from WEs and ACs using the TRIzol reagent (Ambion) according to the manufacturer's instructions.The extracted RNA samples were treated with DNase I to eliminate impurities originating from genomic DNA.Next, reverse transcription-PCR (RT-PCR) was performed using Superscript-IV (Invitrogen) and 1 µg total RNA per reaction according to the manufacturer's guidelines.The thermal cycling protocol was as follows: 30 s at 95 °C, 30 s at 57 °C, 30 s at 72 °C, and 20 to 30 amplification cycles (Table 2).

Quantitative RT-PCR
Samples from all quantitative RT-PCR (RT-qPCR) and ChIP experiments were analyzed using the Biosystems StepOnePlus Real-Time PCR System with Q712-00 Taq Pro Universal SYBR qPCR Master mix (Vazyme).The results were standardized to those of ornithine decarboxylase, a ubiquitous gene in Xenopus embryos that is expressed throughout embryonic development.Graphs were generated using the threshold cycle (Ct) values, delta cycle thresholds, delta-delta cycle thresholds, and the relative fold-change expression method.

Luciferase Assays
Serially deleted and mutated ventx3.2(-1188)lucconstructs were injected with or without Cdx1 and gsc mRNAs.Reporter experiments were then conducted as described previously (Umair et al., 2021).The reporter gene assay involved measuring the relative promoter activity using a luciferase assay kit according to the manufacturer's instructions (Promega).At stages 11 to 11.5, the embryos were divided into 5 groups (n = 3 embryos per group).Then, 10 µL lysis buffer was added to each embryo.Reporter gene activity was assessed through an illuminometer (Berthold Technologies) using 10 µL of each embryo homogenate and 40 µL of luciferase substrate.Each experiment was conducted autonomously and in triplicate (at a minimum).

Site-directed Mutagenesis
Site-directed mutagenesis was conducted using a mutagenesis kit (Muta-Direct, iNtRON Biotechnology) and specific primer oligonucleotides (Table 3) as previously described and according to the manufacturer's instructions (Umair et al., 2021).

ChIP Assay
ChIP was performed as described previously (Kumar et al., 2023).Myc-Cdx1 (1 ng/embryo) and Flag-Gsc (0.5 ng/embryo) mRNAs were introduced at the single-cell stage.Approximately 100 injected embryos per sample were collected and handled using established techniques at stages 11 to 11.5.Chromatin was then subjected to immunoprecipitation using anti-myc (for Cdx1 injected samples) and anti-FLAG (for Gsc-injected samples) polyclonal antibodies (SC-805, Santa Cruz Biotechnology, USA) or normal mouse IgG (SC-2025, Santa Cruz Biotechnology) in the cell lysates.ChIP-PCR was conducted using the immunoprecipitated chromatin; specific primers targeting the Cdx1 response element (CRE) and Gsc response element (GRE) were used (Table 4).

ChIP-Sequencing Analysis
In total, 600 to 700 embryos were injected and collected at stage 11 after the injection of gsc mRNA (0.5 ng/embryo) at the 1-cell stage.The ChIP assay was performed as described previously (Blythe et al., 2009).The immunoprecipitated chromatin was sequenced by Macrogen, and raw data (short reads) were obtained in the Fast Adaptive Shrinkage Threshold Algorithm (FASTA) format.Visualization was performed using the online data analysis tool Galaxy (https://usegalaxy.org).Gsc coverage was plotted together with the display of Model based analysis of ChIP-Seq (MACS) call peak data for the ventx3.2promoter region.

Morpholino Oligos (MOs)
Gene Tools LLC provided the antisense MOs targeting X. laevis Cdx1 and Gsc, with the following sequences: Gsc morpholino (Gsc-MO) was prepared as described previously (Sander et al., 2007).MOs were reconstituted to a concentration of 1 mM in sterile water.Before being microinjected, the MOs were heated at 65 °C and then injected (as indicated) in the ventral and dorsal blastomeres at the 4-cell stage.

Ventx3.2 is a Zygotic Gene Co-expressed With Ventx Family Members
Similar to other Ventx family members, Ventx3.2 had discernible transcript levels near the mid-blastula transition (stages 8-9; Fig. 1A).Ventx3.2 transcript levels increased shortly after stage 9 and peaked during the early to mid-gastrula stages (stages 10-12).Then, they sharply decreased at neurula stage 15, similar to those of other ventx members (Fig. 1A).The expression pattern of ventx3.2 was comparable to that of well-known ventx members, such as ventx1.1 and ventx2.1.This similarity suggests that these genes are likely regulated in a similar manner and have shared functions throughout the early development of Xenopus.As evolutionarily related proteins often share conserved regulatory elements, such as transcription factor (TF)-binding sites and enhancers, we used VENTX to determine the similarities in amino acid sequences.The amino acid sequence of Ventx3.2 was 46.9%, 23.7%, and 27.2% identical to that of Ventx3.1,Ventx1.1, and Ventx2.1,respectively (Fig. 1B).Next, the homeobox domain (HD) of Ventx3.2 was examined to determine its identity.The HD of ventx3.2 was 54%, 55.6%, and 79.4% identical to that of ventx1.1,ventx2.1, and ventx3.1,respectively (Fig. 1C).These findings indicate that ventx3.2 is a divergent ventx family member exhibiting the least resemblance to ventx1.1 and ventx1.2,followed by ventx2.2 and ventx2.1.The slightly reduced divergence with Ventx2 suggests a common evolution for ventx3 and ventx2.Despite its divergence from other Ventx members, Ventx3.2exhibits an early expression pattern and function that resembles the role of BMP4 in limiting the Spemann organizer.

Ventx3.2 Transcription is Upregulated by Cdx1 and Downregulated by Gsc
In this study, the temporal expression patterns of both Cdx1 and Gsc overlapped with those of ventx3.2.(Supplementary Fig. 1A  and B).Thus, it was hypothesized that Cdx1 and Gsc would be present in the ventral and dorsal areas of the embryo, respectively, to precisely regulate the ventral-specific gene ventx3.2.
We investigated the effects of Cdx1 and Gsc on the  endogenous expression of ventx3.2 by administering cdx1 and gsc mRNA through injection (Fig. 2A).The injection of Cdx1 mRNA resulted in the activation of ventral genes, such as ventx3.2,ventx1.1, and ventx2.1, in the AC.The result was confirmed using RT-qPCR and RT-PCR analyses (Fig. 2B and Supplementary Fig. 2A).Conversely, the endogenous expression of ventral genes, such as ventx3.2, in the AC was diminished by Gsc mRNA (Fig. 2C and Supplementary Fig. 2B).These findings imply that Cdx1 and Gsc exert opposite effects on ventx3.2expression in X. laevis.

Gsc and Cdx1 Response Elements are Located Within the Proximal Region of the ventx3.2 Promoter
ChIP-sequencing analysis results showed that Gsc was bound to the upstream promoter region of ventx3.2.Peak calling, followed by coverage plot analysis, helped in the identification and mapping of probable Gsc-binding sites in the promoter region of ventx3.2(Fig. 3A).Examination of the ChIP coverage area of Gsc revealed the presence of 2 putative GREs in the region upstream of the ventx3.2coding sequence.To identify the actively functioning GRE, we created luciferase constructs that included ventx3.2promoter regions with sequential deletions (Fig. 3B).Gsc reduced the promoter activity of ventx3.2(-1188)luc by 0.2-fold (Fig. 3C, bars 1-2).To delineate the functional GRE within the ventx3.2(-1188)promoter, we employed a series of deleted ventx3.2constructs.Coinjecting Gsc mRNAs with ventx3.2-deletedconstructs considerably reduced the promoter activity from -1188 to -197 bp (Fig. 3C, bars 3-10).However, deletion of the constructs at -167 and -47 bp did not cause any significant reduction in promoter activity (Fig. 3C, bars 11-14).These data suggest that the putative active GRE could be found within -197 to -47 bp of the ventx3.2promoter.Next, we investigated the effect of Cdx1 on the ventx3.2(-1188)promoter.Consistent with RT-qPCR results, simultaneous expression of Cdx1 mRNA with the ventx3.2(-1188) promoter markedly increased the activity of the ventx3.2(-1188)to ventx3.2(-197) promoter, reaching up to 35-fold (Fig. 3D).Examination of the deleted promoter constructs of ventx3.2revealed that Cdx1 did not impact the activity of the ventx3.2(-47)bp promoter construct (Fig. 3D, bars 13-14).This result suggests that the CRE is located upstream of the -47 bp region of the ventx3.2promoter.

Site-directed Mutagenesis of GRE and CRE in the ventx3.2 Promoter Uniquely Eliminates Gsc-and Cdx1-Mediated Transcription
Caudal type TFs have selectivity for a conserved DNA sequence with the consensus sequence 5′-ATTT-3′ (Hwang et al., 2002a;Kumar et al., 2019).A thorough examination of the ventx3.2(-1188) promoter sequence revealed a single potential CRE (ATTTG) between positions -70 and -65 bp.We also identified 2 GREs ,GRE1 and GRE2 between positions -174 to -169 (CTTTG) and -466 to -461 (GAAAG), respectively (Fig. 4A  and C).To investigate whether the GREs and CRE function as response elements for Gsc and Cdx1, we introduced a point mutation through site-directed mutagenesis by altering the 2 conserved nucleotides within the response elements (Fig. 4A  and C).The reporter gene assay was conducted using the mutant ventx3.2(-197)mGRE1and ventx3.2(-642)mGRE2and wild-type versions of the promoters, both with and without Gsc, at stage 11.
Both datasets indicate that the proximal region of the ventx3.2promoter contains functional GRE and CRE consensus sequences that regulate its activity.

Gsc and Cdx1 Occupy GRE1and CRE to Differentially Modulate ventx3.2 Transcription
Promoter regions carrying GRE1 and CRE mutations led to a significant reduction in the baseline activity of the deleted ventx3.2(-197) promoter (Fig. 4).This suggests that the response elements may be occupied by the transcriptional repressor Gsc and activator Cdx1.Hence, we examined the direct interaction between TFs Gsc and Cdx1 with the proximal region of the endogenous ventx3.2 promoter.Flag-Gsc-and Myc-Cdx1-injected embryos at the gastrula stage were subjected to direct-binding ChIP to confirm their interaction.Following Flagand Myc-antibody immunoprecipitation against injected Flag-Gsc and Myc-Cdx1 with total chromatin, ChIP-PCR, and ChIP-qPCR were performed.Gsc and Cdx1 are directly bound to the proximal region of the endogenous ventx3.2 promoter (Fig. 5B and C, lane 2).ChIP-qPCR results indicated that Gsc predominantly occupied the GRE, whereas Cdx1 occupied the CRE (Fig. 5D).These findings suggest that the ventx3.2

Cdx1 Knockdown Selectively Restricts ventx3.2 Expression in the VMZ, Whereas Gsc Depletion Promotes it in the Dorsal Marginal Zone
In this study, we postulated that Gsc limits ventx3.2transcription in the dorsal marginal zone (DMZ), whereas Cdx1 positively regulates it in the VMZ.To verify our hypothesis, we first examined the endogenous expression of ventx3.2 in VMZ and DMZ explants.At the start of gastrulation, the VMZ and DMZ were extracted separately and cultivated until stages 11 to 11.5 for RT-qPCR analysis.Ventx3.2expression was considerably greater in the VMZ explants than in the DMZ explants, which is consistent with our hypothesis (Fig. 6A, bar graph 1).The expression of Gsc and Chordin prevailed in the DMZ; in contrast, that of ventral members, such as ventx1.1,ventx2.1, and Cdx1, was predominant in the VMZ (Fig. 6A).This spatial expression pattern suggests the presence of different regions of dominance that contribute to the overall regulation of embryonic development in Xenopus.
To determine whether the reduction in Gsc and Cdx1 expression in the DMZ and VMZ, respectively, facilitates or impedes ventx3.2transcription, we employed a loss-of-function strategy (Fig. 6B).We designed a Cdx1 morpholino and used a previously reported Gsc-MO targeting cdx1 and gsc in X. laevis (Supplementary Fig. 3A).To observe the impact of the loss of function of each gene, we first adjusted the concentration for each morpholino.Headless phenotypes were produced when 70 ng Gsc-MO was injected into each dorsal blastomere at the 4-cell stage (Supplementary Fig. 3B).The expression of ventx3.2 in the Gsc-depleted DMZ was increased by 6-fold (Fig. 6C, bar graph 1).The expression of ventral genes, such as ventx1.1,ventx2.1 and cdx1 was also significantly elevated upon Gsc inhibition in the DMZ (Fig. 6C, bar graphs 2, to 4), whereas that of the dorsal-specific gene chordin was reduced by 2-fold (Fig. 6C, bar graph 5).These results indicate that Gsc negatively regulates ventral genes expression (Fig. 6C).
To assess the potential for Cdx1-mediated upregulation of ventx3.2,we doubly knocked down both Gsc and Cdx1 in the DMZ (Supplementary Fig. 4A).Double knockdown resulted in an increase in the expression of ventx3.2but compared to Gsc depletion alone it decreased by approximately 6-fold to 2-fold (Fig. 6C and Supplementary Fig. 4B, bar graph 1).This suggests that Cdx1 may play a positive role in the regulation of ventx3.2expression.The expression of other ventral genes, including ventx1.1,ventx2.1, and bmp4, showed an identical pattern (Supplementary Fig. 4).These results indicate that Cdx1 might positively regulate the expression of ventral genes, including ventx3.2.
We investigated the impact of Cdx1 on the regulation of ventx3.2 in the VMZ, as the expression of Bmp and Cdx1 is particularly prominent in this region.We injected Cdx1 MO (38 ng) into each ventral blastomere at the 4-cell stage.The inhibition of Cdx1 expression in the VMZ led to a reduction in the expression of ventx3.2 and other markers associated with the ventral region (Fig. 6D).Moreover, the reduction of Cdx1 expression resulted in the activation of organizer genes (chordin and gsc) in the VMZ (Fig. 6D, bar graphs 5 and 6).The modulation of Cdx1 expression in the VMZ using its morpholino oligo (MO) resulted in the development of embryos with shortened or curved axis and phenotypes characterized by posterior truncation (Supplementary Figs.3C and 4E).
The reduced expression of ventx3.2compared to that of other ventral genes implies that the regulation of the expression of ventx family members in the VMZ requires normal Cdx1 levels.Furthermore, the additional Cdx1 knockdown in the DMZ with Gsc resulted in the induction of ventx3.2expression.Therefore, we investigated whether Gsc knockdown can restore ventx3.2expression in the Cdx-depleted VMZ.The VMZ still showed reduced expression of ventx3.2(Supplementary Fig. 4C).We observed a decrease in the induction of gsc and chordin expression in the VMZ upon comparing the Cdx1/Gsc-depleted VMZ with the Cdx1-depleted VMZ (Supplementary Fig. 4C, bar graphs 6 and 7).These findings suggest that ventx3.2expression particularly requires a minimal amount of Cdx1 in the VMZ.Overall, we can conclude that Gsc from the dorsal region is essential for limiting the expression of ventx3.2.However, Cdx1 is crucial for the positive expression of ventx3.2 in the ventral region.

DISCUSSION
In this study, we investigated the regulatory mechanisms underlying ventx3.2transcription during gastrulation in two embryonic regions.
We focused on Ventx3.2, a repressive ventral-specific HD TF that is poorly studied.Ventx3.2transcription began zygotically at blastulation (stages 8/9) and peaked at mid-gastrulation (stage 12) (Fig. 1A).The ability of overexpressed Ventx3.2 to ventralize Xenopus embryos indicates its crucial role in specifying ventral cell fates during development.Its loss-of-function phenotype causing dorsalization further underscores its importance in pathways critical for DV patterning.The complexity of developmental biology is highlighted by the fact that Ventx3.2 has the least sequence identity with ventx family members (Fig. 1B) but still exerts crucial effects.This uniqueness could hint at specialized functions or regulatory mechanisms that distinguish Ventx3.2 from its other ventx counterparts.
Ventx3.2 restores BMP4 signaling and restricts dorsal mesoderm expansion (Shapira et al., 1999(Shapira et al., , 2000)).Moreover, it directly suppresses the expression of organizer and dorsal mesoderm genes, such as Gsc and Otx2, to maintain ventral mesoderm identity (Shapira et al., 2000).Although ventx3.2 is critical for normal development, the molecular mechanism that regulates its expression remains unknown.In the present study, both gain-and loss-of-function approaches were used to investigate the mechanism underlying ventx3.2regulation in Xenopus.Although the effects of ventx3.2 on Gsc and Otx2 in the ventral mesoderm have been identified, how Gsc counteracts it in the dorsal mesoderm remains unknown.Studies on Xenopus have shown that Gsc overexpression inhibits BMP4 signaling and its downstream targets, resulting in the development of dorsalized embryos (Niehrs, 2001;Niehrs et al., 1993Niehrs et al., , 1994)).
The integration of BMP and FGF signaling is crucial for embryonic patterning (Kumar et al., 2021a(Kumar et al., , 2021b)).Cdx1, a transcriptional activator dependent on FGF, exerts a positive regulatory effect on the expression of ventx1.1 and ventx2.1,which are target genes of BMP4 during the embryonic development of Xenopus (Keenan et al., 2006;Kumar et al., 2019;Pillemer et al., 1998).However, whether ventx3.2 is positively regulated by Cdx1 remains unknown.Therefore, it is hypothesized that Gsc and Cdx1 may distinctly modulate the transcriptional regulation of ventx3.2expression.

Gsc Inhibits ventx3.2 Transcription in the DMZ to Maintain Dorsal Destiny
We used Gsc from the dorsal mesoderm to examine its inhibitory effect on ventx3.2transcription for the following reasons: First, Gsc stimulates the development of the dorsal mesoderm or organizer, which is crucial for the formation of embryonic patterns (Cho et al., 1991;Steinbeisser et al., 1993;Yao and Kessler, 2001).Second, Gsc represses the expression of genes that prevent the differentiation of the organizer or dorsal mesoderm (Christian and Moon, 1993;Steinbeisser et al., 1995).The antagonistic relationship between the Gsc and Ventx genes during embryonic patterning in Xenopus has been demonstrated (Sander et al., 2007, Lee et al., 2011).Gsc can counteract the ventralizing effect by repressing the expression of genes that are exclusive to the ventral region (Sander et al., 2007).However, the precise molecular mechanisms through which Gsc restricts the expression of ventx family members in the DMZ remain unclear.Shapira et al. (1999) established a temporal window of competence between ventx3.2 and Gsc.As Gsc is expressed on the dorsal side of Xenopus early gastrulae, the absence of Ventx3.2 transcription in that area may be attributed to Gsc.Evidence for this phenomenon was observed through the examination of the temporal expression patterns of both genes, as demonstrated by the intersection of Gsc expression with ventx3.2expression (Fig. 1A and Supplementary Fig. 1B).Second, ectopic expression of Gsc mRNA considerably reduced the expression of endogenous ventx1. 1, ventx2.1, and ventx3.2 in the AC explants (Fig. 2B).
Gsc exhibits substantial repressor activity by directly binding to cis-acting response regions of the target genes, thereby inhibiting transcription (Cho et al., 1991;Latinkic and Smith, 1999;Ulmer et al., 2017).Previous research has demonstrated that Gsc significantly reduces ventx1.1 promoter activity while having no effect on ventx2.1 promoter (Lee et al., 2011).In this study, our first objective was to determine whether Gsc affects the promoter activity of ventx3.2.We used a ventx3.2reporter to investigate the potential direct effects of Gsc on ventx3.2.Here, Gsc overexpression reduced the ventx3.2(-1188)lucreporter gene expression in WEs (Fig. 3C).As the Xenopus genome has been successfully sequenced (Session et al., 2016), Gsc-based genome-wide ChIP-sequencing has been performed during Xenopus gastrulation (Umair et al., 2021).Our genome-wide ChIP-sequencing results revealed Gsc-binding sites in the 5′ flanking region of the ventx3.2promoter (from -466 to -169 bps) (Fig. 3A).The generation of several deleted promoter constructs, together with site-directed mutagenesis, revealed the presence of conserved cis-acting elements (GRE; GAAAG) in the promoter region.Point mutations were created to examine the functions of GRE1 and GRE2.We found that Gsc lost its repressor activity in the ventx3.2(-197)mGRE1 reporter gene.The binding site of Gsc on the ventx3.2promoter was further confirmed through ChIP-PCR (Fig. 5B).These findings indicate that Gsc may function independently as a repressor of ventx3.2transcription.GRE1 (-174 to -169 bp) was identified in the ventx3.2promoter in the present study.
Spatial expression patterns of ventx3.2transcription in the VMZ and DMZ aid in determining the spatial window wherein its expression is differentially regulated.Loss of function studies were conducted to demonstrate the inhibitory effect of Gsc in the DMZ system.Gsc knockdown affects head development and axis patterning, in addition to increasing ventx1.1 and ventx2.1 expression (Sander et al., 2007).We reconfirmed Gsc knockdown by generating headless phenotypes (Supplementary Fig. 3B and Fig. 4D).Gsc knockdown resulted in elevated ventx3.2expression, thus validating the negative effect of Gsc on ventx3.2 in the DMZ (Fig. 6C).
The expression of Cdx1 was significantly increased in the DMZ when Gsc expression was blocked, as shown in Figure 6C.Cdx1 can enhance the expression of ventral genes (Keenan et al., 2006;Pillemer et al., 1998).Thus, we used a double knockdown approach to investigate whether this increased Cdx1 expression has an impact on the expression of ventx3.2 in the DMZ (Supplementary Fig. 4A).Cdx1 had a partially positive impact on the upregulation of ventx3.2 in the Gsc-depleted DMZ.Despite the Gsc-induced inhibition of Cdx1 expression, ventx3.2expression remained high in the DMZ (Supplementary Fig. 4B).Loss-of-function experiments confirmed that Gsc acts as a repressor for ventx3.2.Consequently, the removal of Gsc activates ventx3.2transcription in the DMZ.By integrating both gain-and loss-of-function experiments, our study defines the regulatory mechanism through which Gsc suppresses ventx3.2transcription in the DMZ to maintain the dorsal identity.

Cdx1 Activates ventx3.2 Transcription in the Ventral
Mesoderm During Embryonic Patterning Among the 3 caudal genes in Xenopus, CDX1 most obviously displays a dorsoventrally regulated pattern of expression, which indicates its potential function within the DV patterning network (Pillemer et al., 1998).Cdx1 activates BMP4 signaling and can emit a signal that causes ventralization during the early stages of gastrulation (Pillemer et al., 1998).We selected Cdx1 as an activator of ventx3.2based on the following reasons: The postulation that Cdx1 may serve as a ventral gene was first supported by the observation that the ventrolateral pattern of Cdx1 expression is comparable to the pattern of Bmp4 expression along the marginal zone during early gastrulation (Dosch et al., 1997;Fainsod et al., 1994;Pillemer et al., 1998).Second, Cdx1 overexpression leads to the dorsal enlargement of the xvent1 and xvent2/vox expression domains (Pillemer et al., 1998).Additionally, the ectopic expression of Cdx1 mRNA significantly induces the transcription of ventx1.1 and ventx2.1 during gastrulation (Kumar et al., 2019;Pillemer et al., 1998), suggesting the essential regulatory function of Cdx1 in the transcription of ventx family members in Xenopus.
The present study elucidated the mechanisms through which Cdx1 activates ventx3.2 at the transcription level in Xenopus embryos.To this end, we evaluated the expression of the endogenous ventx3.2 gene and the reporter gene.Our findings demonstrated a significant increase in the expression of endogenous ventx3.2 and reporter gene activity in response to Cdx1 expression.Cdx1 efficiently occupies the Ventx1.1 response element to increase the activity of the ventx1.1 promoter (Kumar et al., 2019).Therefore, we examined the CREs present in the ventx3.2(-1188) promoter.Serial deletions and site-directed mutagenesis on the ventx3.2(-1188)constructs suggest that CRE is a functional cis-acting element responsible for the Cdx1-mediated activation of ventx3.2transcription.
Cdx1 can activate ventral genes without relying on Bmp4.However, its precise mechanism has not been elucidated (Pillemer et al., 1998) Consistent with our findings, previous studies have shown that antisense Cdx1 can stimulate the expression of organizer genes (Pillemer et al., 1998).Depletion of Cdx1 in the VMZ increased the expression of gsc and chordin (Fig. 6D) in the present study.Moreover, the normal expression of ventx3.2 in the VMZ could not be restored by double silencing of Cdx1 and Gsc (Supplementary Fig. 4C), indicating that Cdx1 is essential for ventx3.2transcription in VMZ.Our adoption of a gain-and loss-of-function approach in this study demonstrated the positive interaction between Cdx1 and CRE in the ventx3.2promoter, which is crucial for maintaining the specification of the ventral lineage in X. laevis.A considerable decrease in BMP4 expression was observed in both the Gsc/Cdx1-depleted DMZ and Cdx1-depleted VMZ.In contrast, ventx3.2expression increased in the Gsc/Cdx1-depleted DMZ (Supplementary Fig. 4B) and was not eliminated in the Cdx1-depleted VMZ (Fig. 6D), indicating the involvement of an additional signaling pathway in its retention and positive regulation.However, further research is required to confirm the potential involvement of other BMPs or ventral signals in the transcriptional regulation of ventx3.2.
In conclusion, we demonstrate the rigorous regulation of ventx3.2transcription in the DMZ and VMZ during embryonic development.The transcription of ventx3.2 in the VMZ is positively regulated by Cdx1 through CRE.In the DMZ, it is negatively regulated by Gsc through GRE.The proposed molecular mechanism by which Gsc and Cdx1 regulate ventx3.2transcription and preserve ventral and dorsal mesodermal patterning is depicted in Figure 7.This study provides a novel understanding of the molecular process that underlies the dorso-ventral and anterior-posterior patterning of the vertebrate model system.

Fig. 1 .
Fig. 1.Zygotic transcription and protein identity comparison of ventx family members.(A) Temporal expression patterns of ventx genes in whole embryos, as determined using RT-qPCR.(B) Protein sequences downloaded from "Xenbase," compared for identity using Clustal omega, and schematically drawn.(C) Amino acid sequences of conserved homeodomains (HD) selected and compared for identity using the "Clustal omega Pairwise alignment tool" and schematically drawn.Solid color lines depict identity within the same class, whereas dotted lines indicate identity within different classes of the ventx family.

Fig. 5 .
Fig. 5. Gsc and Cdx1 bind to the ventx3.2proximal promoter region.(A) Schematic representation of common ChIP-PCR primers F (forward) and R (reverse) as well as the location of GRE and CRE in the ventx3.2promoter.(B, C) ChIP-PCR results showing the interaction between Flag-Gsc and GRE and between Myc-Cdx1 and CRE.Common ChIP primers (containing both GRE and CRE) were used for amplification, while ventx3.2CDS (exon 3) primers served as the negative control for both.(D) ChIP-qPCR results showing the occupancy of Myc-Cdx1 on CRE and Flag-Gsc on GRE.Fold enrichment was utilized to normalize ChIP-qPCR readings.*P ≤ .1 and **P ≤ .01indicate statistical significance.Gsc, goosecoid; Cdx1, caudal type homeobox 1; CRE, Cdx1 response element; GRE, Gsc response element; ChIP-PCR, chromatin immunoprecipitation-polymerase chain reaction; IgG, immunoglobulin G; IP, immuno precipitated; CDS, coding sequence
. The presence of CRE within the ventx3.2promoter indicates that Cdx1 independently activates ventx3.2transcription.The synexpression patterns of ventx3.2 and Cdx1 in the VMZ are shown in Figure 6A.To assess the positive effect of Cdx1 on ventx3.2transcription in VMZ, Cdx1 was depleted with MO.The depletion of Cdx1 in the VMZ suggested that it might serve as an activator, as it inhibited the normal expression of ventx3.2.

Table 2 .
Primers used for RT-PCR amplification

Table 3 .
Primers used for site-directed mutagenesis

Table 4 .
Primer used in ChIP-PCR