Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Nov 26.
Published in final edited form as: Mech Dev. 2006 Jul 26;123(10):719–729. doi: 10.1016/j.mod.2006.07.008

Reduction of XNkx2-10 expression leads to anterior defects and malformation of the embryonic heart

Bryan G Allen a, Kristina Allen-Brady b, Daniel L Weeks a,*
PMCID: PMC2094041  NIHMSID: NIHMS32921  PMID: 16949797

Abstract

Normal vertebrate heart development depends upon the expression of homeodomain containing proteins related to the Drosophila gene, tinman. In Xenopus laevis, three such genes have been identified in regions that will eventually give rise to the heart, XNkx2-3, XNkx2-5 and XNkx2-10. Although the expression domains of all three overlap in early development, distinctive differences have been noted. By the time the heart tube forms, there is little XNkx2-10 mRNA detected by in situ analysis in the embryonic heart while both XNkx2-3 and XNkx2-5 are clearly present. In addition, unlike XNkx2-3 and XNkx2-5, injection of XNkx2-10 mRNA does not increase the size of the embryonic heart. We have reexamined the expression and potential role of XNkx2-10 in development via oligonucleotide-mediated reduction of XNkx2-10 protein expression. We find that a decrease in XNkx2-10 leads to a broad spectrum of developmental abnormalities including a reduction in heart size. We conclude that XNkx2-10, like XNkx2-3 and XNkx2-5, is necessary for normal Xenopus heart development.

Keywords: Xenopus heart development, Nkx, tinman, Congenital defects

1. Introduction

The Nk2 genes encode homeobox-containing transcription factors that are involved in a variety of developmental processes. Some of these Nk2 genes are expressed in the pre-cardiac region and are involved in heart development. In Drosophila, there is a single Nk2 gene, tinman, expressed in the pre-cardiac region. Loss of tinman via null mutation prevents cardiac cell specification and differentiation (Kim and Nirenberg, 1989). Vertebrates may have multiple tinman homologues, referred to as Nkx2, expressed in the pre-cardiac region. Mice that are null for a single tinman homologue, Nkx2-5, still form a heart although it is quite defective (Lyons et al., 1995). The Nkx2-5 knockout mice hearts fail to loop, septate, remain undersized and ultimately lead to the demise of the embryo. However, a tubular heart still forms, cardiac muscle is made and a variety of cardiac specific genes are activated. These observations suggest that other Nkx2 proteins or perhaps additional factors perform some of the developmental functions coordinated by a single protein, Tinman, in flies.

Xenopus laevis cardiac development is easily visualized (Kolker et al., 2000) making it an excellent organism to study Nkx2 gene function. Xenopus has three Nkx2 genes expressed in the developing heart field: XNkx2-3, XNkx2-5 and XNkx2-10 (Newman and Krieg, 1998; Newman et al., 2000). The three proteins share four characteristic domains: a 10 amino acid long TN domain, an Nk-2 domain, GIRAW box and a homeodomain that spans approximately 60 amino acids (Newman et al., 2000). These three proteins have similar capacities to bind to DNA (Sparrow et al., 2000). In addition, both Nkx2-5 and Nkx2-10 are better transcriptional activators in the presence of Gata4 (Newman et al., 2000). Expression of XNkx2-3 and XNkx2-5 begins during gastrulation and continues throughout cardiac development into adulthood (Sparrow et al., 2000). Overexpression of either XNkx2-3 or XNkx2-5 increases the size of the embryonic heart through hyperplasia (Cleaver et al., 1996), while dominant negative mutations in XNkx2-3 or XNkx2-5 decrease the size of the developing heart field (Fu et al., 1998; Grow and Krieg, 1998). Alternatively, the overexpression of XNkx2-10 does not produce an obvious effect (Newman et al., 2000). In situ analyses for all three cardiac XNkx2 transcripts have been reported and are summarized in Fig. 1 (Cleaver et al., 1996; Newman and Krieg, 1998; Newman et al., 2000; Sparrow et al., 2000). XNkx2-3 and XNkx2-5 have similar, though not identical, expression patterns throughout development; however, XNkx2-10 mRNA levels fade as heart tube formation is completed leaving it mostly in the pharyngeal endoderm. Embryonic expression of XNkx2-10 was not previously detected beyond stage 42. Because of these data, most studies have concentrated on the role of XNkx2-3 and XNkx2-5.

Fig. 1.

Fig. 1

Open in a new tab

Analysis of XNkx2-10 expression. (A) A cartoon indicating the differences in XNkx2-3, XNkx2-5 and XNkx2-10 expression patterns at developmental stages 24, 28 and 32. Expression regions were estimated from cardiac XNkx2 in situ hybridization analysis (Cleaver et al., 1996; Newman and Krieg, 1998; Newman et al., 2000; Sparrow et al., 2000). The first column delineates XNkx2-10 expression. Strong XNkx2-10 expression is indicated with royal blue while weak XNkx2-10 expression is indicated with baby blue. The second column represents an overlay compilation of XNkx2-3, XNkx2-5 and XNkx2-10 expression regions. XNkx2-3 expression is indicated in yellow while XNkx2-5 expression is indicated in red. Regions where both XNkx2-3 and XNkx2-5 expression overlap is indicated with the color orange. (B) The RT-PCR products of XNkx2-10 and EF1α. expression from stage 10.5 to 12 (lane 1), stage 22 (lane 2), stage 32 (lane 3), stage 46 (lane 4) and adult heart (lane 5). Lane 6 is a negative control (no reverse transcriptase). PCR products were collected at PCR cycle #45. (C) The RT-PCR products of XNkx2-3, XNkx2-5, XNkx2-10 and EF1α. expression from different Xenopus stage 46 and adult tissues. Lane 1 is the RT-PCR product from RNA harvested from Xenopus stage 46 embryo eyes. Lane 2 is from stage 46 embryo hearts. Lane 3 is from stage 46 embryo tails. Lane 4 is from the adult eye. Lane 5 is from an adult heart. Lane 6 is from an adult liver. Lane 7 is from an adult gut. Lane 8 is from adult skin. Lane 9 is from skeletal muscle. Lane 10 is a negative control that does not contain reverse transcriptase. PCR products were collect at PCR cycle #45. (D) Western blots with protein harvested from adult heart (lane 1) and skeletal muscle (lane 2). Lanes 3 and 4 provide evidence that we have generated an antibody that recognizes XNkx2-10. Lane 3 is XNkx2-10 protein generated in vitro. Primary antibodies were against nucleolin (95 kDa) and XNkx2-10 (black arrow) (29.2 kDa). Lane 4 is S-35-labeled XNkx2-10 generated in vitro.

We found it strange that the XNkx2-10 protein would be expressed in the heart field for no apparent reason and began to re-examine its expression. We show by RT-PCR analysis and Western blot analysis that XNkx2-10 expression does not disappear from the embryo as previously thought but it is expressed in several adult tissues, including the heart. Similar to previously reported results (Newman et al., 2000), our XNkx2-10 overexpression studies failed to generate a phenotype. However, reducing the expression of XNkx2-10 using antisense oligonucleotides led to generalized anterior defects and small hearts. This knockdown phenotype was rescued when XNkx2-10 mRNA was supplied to the embryo. Analysis of potential downstream target genes identified reductions in the expression of Cardiac Troponin I and Tropomyosin. There was little change for several other genes known to regulate heart development. These studies suggest that all three tinman homologues, XNkx2-3, XNkx2-5 and XNkx2-10, play a role in Xenopus cardiac development.

2. Results

2.1. XNkx2-10 developmental stage expression

Previous work utilizing in situ hybridization analysis and PCR analysis demonstrated that XNkx2-10 is first transcribed at embryonic stage 14 and then ceases expression at stage 42 (Newman et al., 2000). A summary of the previously reported expression domains (Cleaver et al., 1996; Newman and Krieg, 1998; Newman et al., 2000; Sparrow et al., 2000) for XNkx2-3, XNkx2-5 and XNkx2-10 is shown in Fig. 1A. These in situ results are not identical, yet are similar enough to generally agree that XNkx2-10 is expressed in the early heart field shortly after the appearance of XNkx2-3 and XNkx2-5. Cardiac-specific expression of XNkx2-10 was reported to decrease during the formation of the heart tube but was maintained in the pharyngeal endoderm through stage 42. RNA extracted from adult heart, spleen, kidney, liver, lung, pancreas and skeletal muscle was examined by RT-PCR analysis and reported to lack XNkx2-10 (Newman et al., 2000).

In order to optimize RT-PCR analysis for XNkx2-10, we examined several tissues for XNkx2-10 mRNA. Because previous studies claimed that XNkx2-10 was not expressed in Xenopus tissue beyond stage 42, we included RNA from stage 46 embryos and adult heart as negative controls for this assay. We were surprised to find that both these sources of RNA were positive for XNkx2-10. As demonstrated in Fig. 1B, we found XNkx2-10 mRNA can be found in total embryonic RNA from stages 22 to 46. In order to approximate the relative amount of XNkx2-10 message present throughout this developmental time period, we performed an RT-PCR analysis with 200 ng of RNA from different developmental stages and determined when we could first identify a PCR product on an ethidium-bromide-stained agarose gel. We first detected stage 22 XNkx2-10 PCR product at PCR cycle #32, stage 36 XNkx2-10 PCR product at PCR cycle #36 and stage 46 XNkx2-10 PCR product at PCR cycle #20 (data not shown). Moreover, we determined that XNkx2-10 was present in RNA isolated from the eyes and hearts of stage 46 embryos, but not in their tails (Fig. 1C). Based on these results, we re-evaluated the expression in adult tissues and found that a number of these test positive for XNkx2-10 mRNA including: the eye, heart, liver and gut (Fig. 1C). We did not find XNkx2-10 mRNA in skeletal muscle or the skin. For comparison, XNkx2-3 and XNkx2-5 adult tissue expressions are also demonstrated (Fig. 1C). XNkx2-3 is expressed in the adult heart, liver, gut and skin, while XNkx2-5 is expressed in the heart, liver and gut. To confirm XNkx2-10 cardiac expression, we extracted protein from adult heart and adult skeletal muscle and analyzed them using an antibody raised against an XNkx2-10 peptide. XNkx2-10 protein was detected in the heart but not in skeletal muscle (Fig. 1D). These results indicate that XNkx2-10, like XNkx2-3 and XNkx2-5, is persistently expressed in the adult heart and other adult tissues.

2.2. Reduction of XNkx2-10 using oligonucleotides results in aberrant development

One way to determine if XNkx2-10 plays a role in Xenopus cardiac development is to prevent XNkx2-10 expression via oligonucleotide inhibition. We have approached this using both a morpholino oligonucleotide that can block the translation of XNkx2-10 mRNA and an N,N-diethylethylenediamine (Deed) antisense oligonucleotide that utilizes endogenous RNase H to degrade XNkx2-10 mRNA (Dagle et al., 2003; Hukriede et al., 2003; Veenstra et al., 2000). Prior to this study, the 5′ untranslated region was unknown for XNkx2-10. Using a 5′ RACE approach, we determined approximately 200 bp of the 5′ untranslated region (NCBI GenBank Accession No. 50300784). Utilizing this sequence, both a morpholino oligonucleotide and a Deed antisense oligonucleotide were generated against XNkx2-10 mRNA targeting the translational start site.

Embryos were injected at the one-cell stage with either: nothing (Figs. 2A and a1), 2 ng of XNkx2-10 mRNA (Figs. 2B and b1), 80 ng control morpholino that presumably does not target a Xenopus mRNA (Figs. 2C and c1), 80 ng anti-XNkx2-10 morpholino (Figs. 2D, d1 and d2), 10 pg of anti-XNkx2-10 Deed antisense oligonucleotide (Figs. 2E, e1 and e2), 80 ng anti-XNkx2-10 morpholino plus 2ng of XNkx2-10 mRNA (Figs. 2F and 2f1) or 10pg of anti-XNkx2-10 Deed antisense oligonucleotide and 2 ng of XNkx2-10 mRNA (Figs. 2G and g1). The XNkx2-10 mRNAs used in these rescue experiments have altered sequences to prevent oligonucleotide association. In addition, XNkx2-10 protein translated from injected mRNA carries a T7 epitope tag.

Fig. 2.

Fig. 2

Open in a new tab

Reduction of XNkx2-10 leads to abnormal development. (A–G) Pictures of stage 22 embryos. Scale bar is 3 mm. (A) are non-injected embryos. (B) Embryos injected with 2 ng of XNkx2-10 mRNA (overexpression). (C) Morpholino control injected embryos. (D) Morpholino injected embryos. (E) Deed oligonucleotide injected embryos. (F) Embryos injected with both morpholino and 2 ng of XNkx2-10 mRNA (morpholino rescue). (G) Embryos injected with both Deed oligonucleotide and 2 ng of XNkx2-10 mRNA (Deed rescue), (a1) Stage 42 non-injected embryo. Scale bar is 3 mm. (b1) Stage 42 overexpression embryo, (c1) Stage 42 morpholino control injected embryo, (d1) Morpholino injected embryo. It has an abnormal head shape and delayed eye development (white arrows). (d2) Morpholino injected embryo that is stunted, has an abnormal head shape and delayed eye development, (e1) Deed oligonucleotide injected embryo that is stunted, has an abnormal head shape and delayed eye development. (e2) Deed injected embryo that is stunted, has an abnormal head shape, delayed eye development and delayed gut development, (f1) Stage 42 morpholino rescue embryo, (g1) Stage 42 Deed rescue embryo. (H) Western blot demonstrating that injection of either the morpholino or Deed oligonucleotide leads to reduced XNkx2-10 protein levels in stage 22 embryos (black arrows) and a Western blot demonstrating the translation of rescue mRNA (T7) in stage 22 embryos. (I) Western blot demonstrating XNkx2-10 and nucleolin protein levels at stage 36. (J) Western blot demonstrating XNkx2-10 and nucleolin protein levels at stage 46. Nucleolin serves as a positive control. The expected size of XNkx2-10 is 29.2 kDa while the expected size of nucleolin is 95 kDa. Lane 1 is protein from non-injected embryos. Lane 2 is protein from morpholino control injected embryos. Lane 3 protein from morpholino injected embryos. Lane 4 contains protein from Deed oligonucleotide injected embryos. Lane 5 contains protein from Deed rescue embryos.

Development through stage 22 appeared unaffected (Figs. 2A–G); however, by stage 26, anterior development was delayed. By stage 42, developmental problems are clearly visible. Nearly 100% of both the XNkx2-10 morpholino-injected embryos and the XNkx2-10 deed-injected embryos had some form of developmental delay (Table 1; Figs. 2d1, d2, e1 and e2, white arrows). An embryo was considered developmentally delayed if it had two or more of the following relative to its siblings: stunted growth, delayed gut development, abnormal head shape/ventral edema and a lack of or incomplete development of the eye. It is worth noting that the XNkx2-10 knockdown embryos had a range in developmental delay severity. While many of the XNkx2-10 knockdown embryos had all four of these characteristics, some only had two or three. For example, the XNkx2-10 morpholino-injected embryo in Fig. 2d1 only has a small eye and abnormal head shape/ventral edema. Alternatively, the Deed-injected embryo in Fig. 2e2 has stunted growth, poor gut development, an abnormal head shape and an underdeveloped eye (Fig. 2e2, white arrows). The control and rescue embryos that we determined to be developmentally delayed never had more than two of these delay characteristics. Consistent with previously reported experiments (Newman et al., 2000), injection of XNkx2-10 mRNA had little effect on embryonic development (Fig. 2b1). Finally, co-injection of XNkx2-10 mRNA with either the XNkx2-10 morpholino or the XNkx2-10 Deed oligonucleotide restored normal development (Figs. 2f1 and g1). The rescue result is consistent with the loss of XNkx2-10 initiating events that result in these developmental defects rather than non-specific toxicity due to either XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide injection.

Table 1.

Analysis of treatment groups for developmental delay

Non-injection Overexpression Morpholino control 2.10 Morpholino Morpholino rescue 2.10 Deed Deed rescue
Normal 66 52 54 2 55 5 40
Dev. delay 2 3 0 70 8 44 7
% Delay 2.9 5.5 0 97.2 12.7 89.8 14.9
Open in a new tab

Embryos from three different trials were analyzed when non-injected siblings were approximately stage 42. An embryo was diagnosed as having developmental delay if it met two or more of the following criteria relative to its siblings: stunted growth, underdeveloped or delayed gut development, abnormal shape of the head and the lack of or retarded eye development. Digital photographs were taken of each embryo and coded. Coded pictures were sorted into normal and developmentally delayed groups prior to revealing their identity.

XNkx2-10 morpholino and XNkx2-10 Deed oligonucleotide effectiveness was confirmed by Western blot analysis. Protein was isolated from stages 22, 36 and 46 non-injected (lane 1), control morpholino-injected (lane 2), XNkx2-10 morpholino-injected (lane 3), XNkx2-10 Deed oligonucleotide-injected (lane 4) and XNkx2-10 Deed rescue embryos (lane 5). At stage 22, the XNkx2-10 morpholino and XNkx2-10 Deed oligonucleotide reduced the amount of XNkx2-10 protein to undetectable levels (Fig. 2H, lanes 3 and 4). Injection of the knockdown oligonucleotide with XNkx2-10 rescue message restored the amount of XNkx2-10 protein. This restoration was evident when assaying either with the XNkx2-10 antibody (Fig. 2H, lane 5) or by detecting rescue levels of XNkx2-10 by using an antibody recognizing the T7 epitope (Fig. 2H, T7 in lane 5). However, for every treatment group, the amount of XNkx2-10 protein present at stages 36 and 46 appeared approximately equivalent (Figs. 2I and J) indicating that the oligonucleotides were no longer inhibitory by stage 36. Thus, the observed phenotypes are related to an early perturbation of XNkx2-10 expression.

Based on our Western blot results (Fig. 2H) and the capability of XNkx2-10 to serve as a transcription factor, we investigated a panel of genes known to be expressed during early heart development for alterations in expression levels. This analysis was carried out by RT-PCR using 200 ng of RNA from 10 stage 22 embryos. We analyzed eight successive PCR cycles for product and identified the cycles where product doubling occurred in the non-injected embryo PCR samples (Fig. 3 lanes indicated with an asterisk).

Fig. 3.

Fig. 3

Open in a new tab

RT-PCR analysis of selected genes expressed in the developing heart. Stage 22 RNA was harvested from non-injected (NI), overexpression (OE), XNkx2-10 morpholino (M), XNkx2-10 Deed oligonucleotide (D), morpholino rescue (MR) and Deed rescue (DR) embryos. RT-PCR products were harvested over eight consecutive PCR cycles. Decreases in XTnIc and Tropomyosin expression were observed for the XNkx2-10 knockdown embryos (black arrows). XNkx2-10 expression was reduced in XNkx2-10 Deed injected embryos (red arrow). The number on top of the column of gels indicates the final PCR cycle that product was collected. Lanes marked with an asterisk were determined to have an amount of PCR product that had approximately doubled when compared to the previous lane’s PCR product. Experiments were performed in duplicate. EF1α. served as a positive control.

Previous dominant negative studies of XNkx2-3 and XNkx2-5 observed a reduction in the amount of cardiac troponin I (XTnIc) mRNA (Grow and Krieg, 1998). XtnIc initiates its expression between developmental stages 20 and 27 and is expressed only in the heart developing region (Drysdale et al., 1994). Stage 22 embryos injected with either the XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide did not have a detectable PCR product for XTnIc (Fig. 3, black arrows). β-tropomyosin embryonic (β-TMemb) initiates its expression between stages 22 and 25 and at stage 25 is expressed in the developing somites, developing heart region and tail region (Gaillard et al., 1999). Stage 22 embryos injected with either the XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide had a 2-PCR cycle reduction in the amount of β-TMemb message (Fig. 3, black arrows). This suggests that in the absence of XNkx2-10, activation of these genes is delayed or diminished.

Other cardiac-related transcription factors were not affected including: Tbx5, Gata4, XNkx2-3 and XNkx2-5. Thus, normal levels of these transcripts do not ensure normal activation of XTnIc or β-TMemb. We also examined the levels of the transcription factor PitX2c. PitX2c is associated with laterality control, atrial development and atrio-ventricular chamber separation (Dagle et al., 2003). However, PitX2c expression was also unaffected (Fig. 3).

XNkx2-10 mRNA was reduced to undetectable levels in the Deed oligonucleotide-injected embryos but not in the morpholino-injected embryos (Fig. 3, red arrow). This correlates to their respective modes of action. We also note that by stage 22, embryos injected with XNkx2-10 mRNA have approximately double the amount of messenger RNA that could be used for XNkx2-10 protein synthesis. Elongation factor 1α (EFIα) detection served to control both for RNA recovery and quality.

2.3. Early reduction of XNkx2-10 leads to cardiac abnormalities in later-stage embryos

We analyzed the hearts of stage 46 embryos injected with either the XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide. By this stage, heart looping, chamber formation and valve formation are normally near completion. Embryos were fixed in Dent’s fixative and stained with FITC-labeled phalloidin to allow visualization of the heart. Hearts were assessed for overall size, trabeculation, evidence of leftward bending of the outflow tract and atrial positioning.

Analysis of cardiac size was based upon comparison of the maximum circumference of the ventricle measured in stacks of confocal sections. Eighteen non-injected and morpholino-injected stage 46 hearts were measured, and 16 hearts from all other groups were analyzed (Table 2). The mean circumference of hearts in embryos treated with either the XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide was about half that seen in any other treatment group. We performed a global analysis of all the data using the non-parametric Kruskal–Wallis test, which was highly significant (H = 71.32, p = 2.19 × 10−13). To examine the underlying differences among the various treatment groups, we ran pairwise analyses comparing each treatment group with both the XNkx2-10 morpholino and XNkx2-10 Deed oligonucleotide treatments using the Mann-Whitney test (see Table 2). As expected, the ventricle sizes of the XNkx2-10 morpholino-injected embryos (p value range: 1.62 × 10−6 to 6.75 × 10−7) were significantly smaller than the control embryos (non-injected, morpholino control and overexpression). We noted that ventricular area was not dependent on the choice of oligonucleotide inhibition as there was no significant size difference between the XNkx2-10 morpholino-injected embryos and XNkx2-10 Deed-injected embryos (p = 0.783). Furthermore, rescue of both XNkx2-10 morpholino and XNkx2-10 Deed oligonucleotide injection resulted in ventricular areas that were comparable to control samples (p value range 6.75 × 10−7−4.26 × 10−6). A box plot showing the median, interquartile distance (first and third quartiles) and extending “whiskers” (3/2 times the interquartile distance) of maximum ventricular area for each treatment group is shown in Fig. 4.

Table 2.

Comparison of embryonic stage 46 maximum ventricular areas between treatment groups

N Minimum (μm2) Maximum (μm2) Mean (μm2) Comparison to morpholino p value Comparison to Deed p value
Non-injected 18 40538 94347 64438 5.77 ×10−7* 2.28 × 10−6*
Morpholino control 16 44187 87134 63617 6.75 × 10−7* 1.70 × 10−6*
Overexpression 16 38988 101080 61894 1.62 × 10−6* 5.11 × 10−6*
Morpholino 18 13283 43275 31355 – 7.83 × 10−1
Deed 16 14508 48304 30702 7.83 × 10−1 –
Morpholino rescue 16 40877 81463 54478 9.61 × 10−7* 4.26 × 10−6*
Deed rescue 16 46271 69732 60677 6.75 × 10−7* 1.70 × 10−6*
Open in a new tab

Embryos were fixed in Dents solution and stained with FITC-phalloidin prior to imaging by confocal microscopy. Images demonstrating maximum ventricular areas are shown in Fig. 5.

*

p values were determined using a pairwise comparison with either the morpholino or Deed cohorts.

Fig. 4.

Fig. 4

Open in a new tab

Box plot demonstrating the range of maximum ventricular area values. Sample 1 contains data from non-injected embryos. Sample 2 contains data from morpholino control injected embryos. Sample 3 contains data from XNkx2-10 overexpression embryos. Sample 4 contains data from XNkx2-10 morpholino injected embryos. Sample 5 contains data from XNkx2-10 Deed oligonucleotide injected embryos. Sample 6 contains data from morpholino rescue embryos. Sample 7 contains data from Deed rescue embryos. The morpholino and Deed oligonucleotide significantly reduced the maximum ventricular area of their respective hearts. Alternatively, co-injection of either the morpholino or Deed oligonucleotide with 2 ng of XNkx2-10 mRNA (rescue) restored the size of the maximum ventricular areas. The * symbol indicates a p value of less than 0.00001.

Other more subjective cardiac differences were also observed for the XNkx2-10 morpholino- and XNkx2-10 Deed-injected embryos. The defects seen were consistent with incomplete execution of heart morphogenesis including decreased trabeculation and failure of the atria to move to a more anterior position. In addition, we routinely noted rightward misplacement of the atrial septum (Fig. 5P). Reduction of XNkx2-10 did not lead to a complete failure to execute laterality changes as the outflow tracts bent to the left as they do normally (Figs. 5M and N and U–X). The extent of anterior defects and pericardial edema make it difficult to ascertain if the cardiac defects are a primary effect or secondary effect of an XNkx2-10 knockdown. However, all these phenotypes were rescued when the XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide was co-injected with 2 ng XNkx2-10 mRNA (Figs. 5Q–T and Y–β) suggesting that the cardiac defects are a result of XNkx2-10 knockdown. As has been reported by others, injection of 2 ng of XNkx2-10 alone did not appear to affect heart development (Newman et al., 2000) (Figs. 5I–L).

Fig. 5.

Fig. 5

Open in a new tab

Confocal microscope generated image comparisons of embryonic stage 46 hearts. Scale bar is 200 μm. The embryos were fixed in Dents fixative and then stained with FITC-phalloidin. Images were captured at 5 nm intervals. Panels (A, E, I, M, Q, U and Y) are a compilation of 20–30 images. Panels (B, F, J, N, R, V and Z) are single images demonstrating the maximum ventricular areas (red outline). Panels (C, G, K, O, S, W and α) are single images demonstrating the spiral valve in the outflow tract (cyan arrows). Panels (D, H, L, P, T, X and β) are single images demonstrating the atrial septum (magenta arrows).

3. Discussion

Previous experiments found little evidence that XNkx2-10 plays a role in heart development (Newman et al., 2000). This study identifies XNkx2-10 as a regulatory factor for both gross anterior development and for the heart development.

Our RNA expression analysis and Western blot analysis indicate that XNkx2-10 expression continues beyond what was previously reported (Newman et al., 2000). Not only was XNkx2-10 present through at least stage 46 of development (analyzed in whole embryos, stage 46 heart and eye tissue), but was also found in many adult tissues, including the heart. This distribution suggests that XNkx2-10 may have multiple roles in development as well as a possible role in tissue maintenance.

Embryos lacking XNkx2-10 are apparently normal through gastrulation and neurulation, but eventually display a number of defects. By stage 46, embryos treated to reduce the levels of XNkx2-10 had defective anterior morphology and hearts. The more widespread effects are consistent with the previous characterization of XNkx2-10 expression in the anterior region of the developing embryo (Newman et al., 2000). Because the later-stage knockdown embryos have multiple defects, assigning specific defects to the direct action of XNkx2-10 must be done cautiously. We do not know if the observed defects are a direct or indirect result of a knockdown of XNkx2-10. However, all the defects could be rescued with the addition of XNkx2-10 mRNA supporting the idea that at least the initiating events leading to the anterior and cardiac defects are due to XNkx2-10 reduction.

3.1. What genes are misregulated when XNkx2-10 levels are lowered?

We have started to sort out the consequences of XNkx2-10 reduction on embryonic levels of genes important for early gross anterior and cardiac development. We present our analysis on stage 22 embryos. At this stage, there is no gross indication that either XNkx2-10 morpholino or XNkx2-10 Deed oligonucleotide targeting XNkx2-10 have affected development. However, both treatments result in undetectable levels of XNkx2-10 protein. In the absence of XNkx2-10 protein, transcript levels of EF1α, Nkx2-3, XNkx2-5, Gata4, Tbx5 and PitX2c appear normal. We do not yet know if the loss of XNkx2-10 causes localized changes in these genes. However, β-TMemb which is expressed in the developing somites and heart and XTnIc which is expressed in the developing heart region are reduced. Thus, at least two genes expressed during early cardiac development are altered even though apparently normal embryonic levels of XNkx2-3, XNkx2-5, Gata4 and Tbx5 are present. A similar analysis using stage 36 embryos also showed reduced levels of XTnIc (data not shown). However, by stage 36, the embryo’s hearts are already smaller, so a reduction of XTnIc mRNA levels are hard to separate from less total heart tissue.

The reduction of XNkx2-10 levels is limited to early stages. By stage 36, all embryos have restored synthesis of XNkx2-10 protein; however, those that lacked XNkx2-10 through stage 22 develop abnormally and have smaller hearts. This restoration of XNkx2-10 later in development does not make up for its loss in early development.

We can contrast these XNkx2-10 knockdown hearts to those generated by the loss of Nkx2-5 in the mouse (Lyons et al., 1995). In the mouse Nkx2-5 null, the heart tube forms, but the laterality and septation program are not executed (Lyons et al., 1995). Because the Nkx2.5 null mouse embryos die approximately 10 days postcoitum, it is not clear, how far heart development might have proceeded.

3.2. How does the loss of XNkx2-10 compare to other treatments used to study heart development in Xenopus embryos?

The alteration of the other two Nk2 family members, XNkx2-3 and XNkx2-5, were carried out using dominant negative constructs (Fu et al., 1998; Grow and Krieg, 1998). The addition of mRNA encoding either XNkx2-3 or XNkx2-5 dominant negative protein leads to reduction in heart size and in the most dramatic examples no heart at all. In particular, studies by Grow and Krieg suggest that the embryos had little morphological consequence besides a failure to form a heart when assayed at stage 30. The reduction of XNkx2-10 reduced the level of XTnIc, but embryos still form a heart. Furthermore, although the embryos looked very similar to controls through stage 22, by stage 26 anterior abnormalities are already apparent.

Studies on Tbx5 and Tbx20 (Brown et al., 2005) using morpholino-based reduction approach reported delays in development starting at neurulation (stage 16) and show stage 38 and 45 embryos with significant anterior defects and edema. The general appearance of later-stage embryos share some features seen in embryos that have reduced levels of XNkx2-10. However, the hearts of embryos with reduced levels of Tbx5 often failed to close the cardiac tube and embryos with reduced levels of either Tbx5 or Tbx20 had small hearts that failed to loop. Although the hearts of the XNkx2-10 embryos are small, they have fused and executed many parts of the normal looping program.

Inhibition of retinoic acid signaling also has been studied with respect to heart development in Xenopus (Collop et al., 2006). Blocking retinoic acid leads to multiple developmental defects, but among the problems is a failure to fuse the lateral heart fields at the midline. In the same study, treatment with retinoic acid reduced the levels of XNkx2-5, but not Gata-4, while treatment with a retinoic acid antagonist seemed give rise to an early increase in XNkx2-5 expression and slight reduction of Gata-4. In contrast, reductions of XNkx2-10 allow the embryonic heart to fuse and we noted no change in the expression of either XNkx2-5 or Gata-4. The studies on Tbx-5, Tbx-20 and the retinoic acid pathways, along with the studies presented here, indicated that even in a background where there are other developmental abnormalities, the effect on the heart can still be differentially affected by the different experimental treatments.

3.3. What other Nk2 genes might also affect heart development?

A mutation in the homeodomain of human Nkx2-6 is implicated in forming a common arterial trunk (Heathcote et al., 2005). In mice, Nkx2-6 is expressed in the pharyngeal pouches and the developing heart (Tanaka et al., 2000). Interestingly, in Nkx2-6 knockout mice, Nkx2-5 pharyngeal and outflow tract expression increases suggesting that it is compensating for the reduction of Nkx2-6 (Tanaka et al., 2000). Based on sequence comparison, it is unlikely that XNkx2-10 is the amphibian homologue of Nkx2-6. However, the action of multiple Nk2 proteins on proper heart development seems shared.

While XNkx2-10 knockdown embryos demonstrated reduced expression of some known cardiac genes, others did not appear to be affected, including XNkx2-3 and XNkx2-5. Nevertheless, abnormal cardiac development resulted in these embryos, suggesting the presence of developmental pathways requiring XNkx2-10.

Experiments to delineate the cardiac pathways specifically regulated by individual XNkx2 proteins are underway. These studies will allow a better understanding of how the XNkx2 proteins are coordinately regulated to form a normal heart.

4. Experimental

4.1. Embryo and cardiac tissue RNA isolation and RT-PCR analysis

Total embryonic RNA from stage 10.5 to 12, 22, 32, 46, stage 46 eye, heart and tail and adult tissue was isolated according to the manufactures instructions (Ambion RNAqueous small-scale phenol-free total RNA isolation kit). RT-PCR was performed according to manufacture’s instructions (Qiagen OneStep RT-PCR kit). Approximately 200 ng of total RNA was used for each RT-PCR reaction. The primer pair for XNkx2-10 developmental-stage expression analysis was: atgctccccagccccgct and tcaccaggccctgatccc. For XNkx2-10 potential downstream target analysis, we used stage 22 mRNA and the following primer pairs: XNkx2-3 -gcctgcagctgtcctttttg and acgcacgtttttgttttggg (Ma, 2004), XNkx2-5 - ccagagagacaccgctaagg and ctaggccgcagtctaccaag, XNkx2-10 - cgctacctctacccccttct and tgatcagctgtggaggactc, Gata4 - ttctcatcccccagtgtctc and ttcctctcc-catggagtttg, Pitx2c - cagcccagacactgcagata and agggaa-gggttgctgagatt, XTnIc - ccagatggagaaaaggtgga and tgatc- cgaaactcgattcct, Tropomyosin - tggagatggcggagaagaag and gcagcaagtggcagtcacga (Charbonnier et al., 2002), TBX5 - gttactggcctgaacccaaa and atccggtgtagctccatgtc and finally EF1α. cagattggtgctggatatgc and actgccttgatgactcctag. For primer design, we used the online program Primer3 (Rozen and Skaletsky, 2000). For quantitative RT-PCR analysis, 1 μl of RT-PCR product was collected over eight consecutive PCR cycles (Fig. 3). We determined the relative amounts of PCR product using digitally photographed images of ethidium bromide gels with the online program ImageJ.

4.2. XNkx2-10 oligonucleotide-mediated knockdown and rescue

The XNkx2-10 5′ untranslated region (GenBank deposit Accession No. AY654625) was determined by a 5′ rapid amplification of cDNA ends (RACE) approach (Ambion First choice RLM RACE kit). An XNkx2-10 morpholino was generated against the XNkx2-10 5′ start site and untranslated region having the sequence GGAGCATTCCCATCTGCTGTCCTGT (Gene Tools). An XNkx2-10 Deed oligonucleotide was also generated against the XNkx2-10 5′ start site and untranslated region having the sequence T+C+C+C+A+TCTGCTGT+C+C+T+G+T where “+” represents a Deed-modified phosphodiester bond. The Deed oligonucleotide was prepared and purified using H-phosphonate chemistry as previously described (Dagle and Weeks, 1996, 2001). The XNkx2-10 coding region was synthesized from stage 22 embryo RNA by RT-PCR (Qiagen OneStep RT-PCR kit) using the primer pair atgctccccagccccgct and tcaccaggccctgatccc. The PCR product was sequenced to ensure fidelity to XNkx2-10. The expression region was then cloned into the plasmid GFP.RN3 (Zernicka-Goetz et al., 1996) replacing GFP with the coding region of XNkx2-10. Utilizing the XNkx2-10 expression plasmid, mRNA used for rescue experiments was generated with T3 mMessage Machine (Ambion). This rescue mRNA lacks the morpholino and Deed oligonucleotide target sites but does encode for an N-terminal T7 tag.

An antibody to detect XNkx2-10 was made in chickens against the XNkx2-10 peptide CAQQGQLQATQQGIR (Sigma Genosys). The IgY antibodies were purified from egg yolks using the Eggcellent Chicken IgY Purification Kit (Pierce) following manufacture’s instructions. The XNkx2-10 antibody was affinity-purified using the XNkx2-10 peptide in conjunction with the SulfoLink Kit (Pierce) following manufacture’s instructions.

Xenopus laevis males and females were obtained either from Xenopus I or Nasco. Females were induced approximately 12 h before the time to lay with an injection of human chorionic gonadotropin (Sigma). Eggs were obtained and fertilized in vitro using crushed testes isolated from X. laevis males. Fertilized eggs were dejellied for approximately 2 min in 2% cysteine, rinsed with 0.3×MMR and placed in 0.3×MMR 3% Ficoll 400 (Sigma) prior to injection. Single-cell embryos were injected with either: nothing, 2 ng of XNkx2-10 mRNA, 80 ng of a control morpholino (Gene Tools), 80 ng of XNkx2-10 morpholino, 10 pg of Deed oligonucleotide, 80 ng of XNkx2-10 morpholino and 2 ng of XNkx2-10 mRNA, or 10 pg of Deed oligonucleotide and 2 ng of XNkx2-10 mRNA in a total volume of 10 nl. Embryos were allowed to develop at 18 °C. Embryonic developmental staging used criteria described in the “Normal Table of X. laevis (Daudin)” (Nieuwkoop et al., 1967).

All procedures using animals were reviewed and approved by the University of Iowa Institutional Animal Care and Use Committee.

4.3. Embryo analysis

Live embryos were analyzed through embryonic stage 46 with a Zeiss dissecting microscope (Stemi SV 11). Bright-field pictures of whole embryos were taken with a Diagnostic Instrument Spot Camera.

Stage 46 embryos were fixed in Dents solution (80% methanol, 20% DMSO) and stored at −20 °C until needed. For cardiac analysis, embryos were equilibrated to 100% methanol and then stained with FITC-labeled phalloidin (Sigma) [1 part phalloidin (26.4 μM):500 parts methanol] overnight at room temperature. Stained embryos were then washed with methanol and their tissues cleared with benzyl alcohol: benzyl benzoate as previously described (Kolker et al., 2000). Cardiac size and structure were analyzed with a Zeiss Axioplan 2 microscope and a Bio-Rad confocal microscope. The Apotome feature on the Axioplan was used to optically section hearts illuminated with an FITC-GFP filter with an emission filter limit of 535 nm. Measurement from the Zeiss system used the Axioplan software, and sections obtained from the Bio-Rad confocal microscope were analyzed using ImageJ. Figures presented in this paper were oriented and processed with Adobe Photoshop 7.0 and ImageJ.

4.4. Protein extraction and Western blotting

Five stage 22, 36 or 46 embryos were isolated, suspended in 400 μl protein extraction buffer (0.1 M NaCl, 1% Triton X-100, 20 mM Tris pH 7.6, 1 mM PMSF) and sonicated. Approximately 0.5 g of adult heart tissue or skeletal muscle was suspended in 600 μl of protein extraction buffer and sonicated. The solution was spun at 12,000 rpm in a micro-centrifuge (HERMLE Z 200 M/H) for 10 min and the aqueous layer saved. Three microliters of embryo protein extract (~1 embryo equivalent), adult heart protein (at a concentration of 34.81 mg/ml) or adult skeletal muscle (at a concentration of 47.41 mg/ml) was combined with 6 μl of 6×SDS loading buffer, heated at 95 °C for 5 min and loaded on to a 4–15% Tris–HCl precast polyacrylamide gel (Bio-Rad). Protein was transferred from the gel to a piece of 0.45 μm nitrocellulose membrane. The membrane was divided into two pieces. One piece was probed with the nucleolin primary antibody (DSHB #b6-6E7) (Dreyer et al., 1985) and the other piece probed with either the XNkx2-10 antibody or T7 tag antibody. The membranes were blocked overnight in TBSTM [1×TBS, 7.5% non-fat dry milk (Carnation) and 0.1% Tween-20 (Fisher Biotech)] at 4°C. The XNkx2-10 antibody was diluted 1:25 (starting concentration of 0.6 mg/ml) while the T7 tag and nucleolin antibodies were diluted 1:2000 (nucleolin starting concentration of 1.15 mg/ml) in TBSTM. The secondary antibody for XNkx2-10 was an HRP-conjugated goat antichicken antibody (abcam) diluted 1:2000 (starting concentration of 2.82 mg/ml) in TBSTM, while the secondary antibody for both the T7 tag and nucleolin was an HRP-conjugated rabbit antimouse (Sigma) antibody diluted 1:2000 (starting concentration of 9.23 mg/ml) in TBSTM. The labeled proteins were visualized via an ECL reaction (SuperSignal West Pico Chemiluminescent Substrate – Pierce).

S-35-labeled XNkx2-10 was generated using the XNkx2-10 expression plasmid according to manufacture’s instructions (T3 TnT Coupled Reticulocyte Lysate Systems – Promega).

Acknowledgments

The authors recognize the University of Iowa Central Microscopy Research Facility for assistance with the confocal microscope used in this study. The antibody #b6-6E7 developed by Dryer et al. was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the Department of Biological Sciences, University of Iowa. We acknowledge J. Dagle, H. Bartlett, S. Kolker, R. Escalera, C. Fett, E. Li, A. Struck-Marcell, C Welp, E. Hornick and M. Johnson for their helpful discussions and technical support. We also thank P. Krieg (University of Arizona) for initial discussion on XNkx2-10. Funding to D.L.W. from the National Institutes of Health (GM069944) supported this work. B.G.A. is a student in the Medical Scientist Training Program at the Roy J. and Lucille A. Carver College of Medicine, University of Iowa.

References

  1. Brown DD, Martz SN, Binder O, Goetz SC, Price BM, Smith JC, Conlon FL. Tbx5 and Tbx20 act synergistically to control vertebrate heart morphogenesis. Development. 2005;132:553–563. doi: 10.1242/dev.01596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Charbonnier F, Gaspera BD, Armand AS, Van der Laarse WJ, Launay T, Becker C, Gallien CL, Chanoine C. Two myogenin-related genes are differentially expressed in Xenopus laevis myogenesis and differ in their ability to transactivate muscle structural genes. J Biol Chem. 2002;277:1139–1147. doi: 10.1074/jbc.M107018200. [DOI] [PubMed] [Google Scholar]
  3. Cleaver OB, Patterson KD, Krieg PA. Overexpression of the tinman-related genes XNkx-2.5 and XNkx-2.3 in Xenopus embryos results in myocardial hyperplasia. Development. 1996;122:3549–3556. doi: 10.1242/dev.122.11.3549. [DOI] [PubMed] [Google Scholar]
  4. Collop AH, Broomfield JA, Chandraratna RA, Yong Z, Deimling SJ, Kolker SJ, Weeks DL, Drysdale TA. Retinoic acid signaling is essential for formation of the heart tube in Xenopus. Dev Biol. 2006;291:96–109. doi: 10.1016/j.ydbio.2005.12.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dagle JM, Sabel JL, Littig JL, Sutherland LB, Kolker SJ, Weeks DL. Pitx2c attenuation results in cardiac defects and abnormalities of intestinal orientation in developing Xenopus laevis. Dev Biol. 2003;262:268–281. doi: 10.1016/s0012-1606(03)00389-0. [DOI] [PubMed] [Google Scholar]
  6. Dagle JM, Weeks DL. Positively charged oligonucleotides overcome potassium-mediated inhibition of triplex DNA formation. Nucleic Acids Res. 1996;24:2143–2149. doi: 10.1093/nar/24.11.2143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dagle JM, Weeks DL. Oligonucleotide-based strategies to reduce gene expression. Differentiation. 2001;69:75–82. doi: 10.1046/j.1432-0436.2001.690201.x. [DOI] [PubMed] [Google Scholar]
  8. Dreyer C, Wang YH, Hausen P. Immunological relationship between oocyte nuclear proteins of Xenopus laevis and X. borealis. Dev Biol. 1985;108:210–219. doi: 10.1016/0012-1606(85)90024-7. [DOI] [PubMed] [Google Scholar]
  9. Drysdale TA, Tonissen KF, Patterson KD, Crawford MJ, Krieg PA. Cardiac troponin I is a heart-specific marker in the Xenopus embryo: expression during abnormal heart morphogenesis. Dev Biol. 1994;165:432–441. doi: 10.1006/dbio.1994.1265. [DOI] [PubMed] [Google Scholar]
  10. Fu Y, Yan W, Mohun TJ, Evans SM. Vertebrate tinman homologues XNkx2-3 and XNkx2-5 are required for heart formation in a functionally redundant manner. Development. 1998;125:4439–449. doi: 10.1242/dev.125.22.4439. [DOI] [PubMed] [Google Scholar]
  11. Gaillard C, Lerivray H, Theze N, Cooper B, Lepetit D, Mohun T, Thiebaud P. Differential expression of two skeletal muscle betatropomyosin mRNAs during Xenopus laevis development. Int J Dev Biol. 1999;43:175–178. [PubMed] [Google Scholar]
  12. Grow MW, Krieg PA. Tinman function is essential for vertebrate heart development: elimination of cardiac differentiation by dominant inhibitory mutants of the tinman-related genes, XNkx2-3 and XNkx2-5. Dev Biol. 1998;204:187–196. doi: 10.1006/dbio.1998.9080. [DOI] [PubMed] [Google Scholar]
  13. Heathcote K, Braybrook C, Abushaban L, Guy M, Khetyar ME, Patton MA, Carter ND, Scambler PJ, Syrris P. Common arterial trunk associated with a homeodomain mutation of NKX2.6. Hum Mol Genet. 2005;14:585–593. doi: 10.1093/hmg/ddi055. [DOI] [PubMed] [Google Scholar]
  14. Hukriede NA, Tsang TE, Habas R, Khoo PL, Steiner K, Weeks DL, Tarn PP, Dawid IB. Conserved requirement of Lim1 function for cell movements during gastrulation. Dev Cell. 2003;4:83–94. doi: 10.1016/s1534-5807(02)00398-2. [DOI] [PubMed] [Google Scholar]
  15. Kim Y, Nirenberg M. Drosophila NK-homeobox genes. Proc Natl Acad Sci USA. 1989;86:7716–7720. doi: 10.1073/pnas.86.20.7716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kolker SJ, Tajchman U, Weeks DL. Confocal imaging of early heart development in Xenopus laevis. Dev Biol. 2000;218:64–73. doi: 10.1006/dbio.1999.9558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lyons L, Parsons LM, Hartley L, Li R, Andrews JE, Robb L, Harvey RP. Myogenic and morphogenetic defects in the heart tubes of murine embryos lacking the homeobox gene Nkx2-5. Genes Dev. 1995;9:1654–1666. doi: 10.1101/gad.9.13.1654. [DOI] [PubMed] [Google Scholar]
  18. Ma CM. Spatial and temporal expression patterns of Xenopus Nkx-2.3 gene in skin epidermis during metamorphosis. Gene Expr Patterns. 2004;5:129–134. doi: 10.1016/j.modgep.2004.05.003. [DOI] [PubMed] [Google Scholar]
  19. Newman CS, Krieg PA. Tinman-related genes expressed during heart development in Xenopus. Dev Genet. 1998;22:230–238. doi: 10.1002/(SICI)1520-6408(1998)22:3<230::AID-DVG5>3.0.CO;2-7. [DOI] [PubMed] [Google Scholar]
  20. Newman CS, Reecy J, Grow MW, Ni K, Boettger T, Kessel M, Schwartz RJ, Krieg PA. Transient cardiac expression of the tinman-family homeobox gene, XNkx2-10. Mech Dev. 2000;91:369–373. doi: 10.1016/s0925-4773(99)00291-9. [DOI] [PubMed] [Google Scholar]
  21. Nieuwkoop PD, Faber J Hubrecht-Laboratorium (Embryologisch Instituut) Normal Table of Xenopus laevis (Daudin): A Systematical and Chronological Survey of the Development from the Fertilized Egg Till the End of Metamorphosis. North-Holland: Amsterdam; 1967. [Google Scholar]
  22. Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Meth Mol Biol. 2000;132:365–386. doi: 10.1385/1-59259-192-2:365. [DOI] [PubMed] [Google Scholar]
  23. Sparrow DB, Cai C, Kotecha S, Latinkic B, Cooper B, Towers N, Evans SM, Mohun TJ. Regulation of the tinman homologues in Xenopus embryos. Dev Biol. 2000;227:65–79. doi: 10.1006/dbio.2000.9891. [DOI] [PubMed] [Google Scholar]
  24. Tanaka M, Yamasaki N, Izumo S. Phenotypic characterization of the murine Nkx2.6 homeobox gene by gene targeting. Mol Cell Biol. 2000;20:2874–2879. doi: 10.1128/mcb.20.8.2874-2879.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Veenstra GJ, Weeks DL, Wolffe AP. Distinct roles for TBP and TBP-like factor in early embryonic gene transcription in Xenopus. Science. 2000;290:2312–2315. doi: 10.1126/science.290.5500.2312. [DOI] [PubMed] [Google Scholar]
  26. Zernicka-Goetz M, Pines J, Ryan K, Siemering KR, Haseloff J, Evans MJ, Gurdon JB. An indelible lineage marker for Xenopus using a mutated green fluorescent protein. Development. 1996;122:3719–3724. doi: 10.1242/dev.122.12.3719. [DOI] [PubMed] [Google Scholar]

RESOURCES