SEARCH

SEARCH BY CITATION

Keywords:

  • hypoxia;
  • paraquat;
  • nodal signaling;
  • oral ectoderm;
  • redox signaling;
  • ROS

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The sea urchin oral-aboral (OA) axis is established in part by Nodal signaling. The OA axis is also influenced by treatments affecting respiration and Nodal transcription is influenced by redox-dependent transcription factors. This suggests that intracellular redox state plays a role in OA axis specification. Since cellular redox state can be altered by the formation of excess reactive oxygen species (ROS), and hypoxia and paraquat generate ROS in cells, we asked whether these treatments affected specification of the OA axis and Nodal expression. Embryos cultured under conditions that elevate ROS, demonstrate perturbed ectoderm specification, but other territories are not affected. Immunohistochemical and Q-RT-PCR analyses revealed that both oral and aboral ectoderm genes are downregulated. Our results argue that elevating ROS in sea urchin embryos by these treatments blocks early steps in ectoderm differentiation preceding the polarization of the ectoderm into oral and aboral territories. Developmental Dynamics 238:1777–1787, 2009. © 2009 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Nodal signaling is required for the specification of the OA axis in sea urchin embryos (Duboc et al.,2004; Flowers et al.,2004). The events that initiate nodal transcription are not well understood, but several lines of evidence suggest that redox signaling is involved. Initial investigations demonstrated that respiratory asymmetry is an important component of OA axis specification; inhibitors of respiration influence OA axis formation (Pease,1941,1942a,b), and cytochrome oxidase, a mitochondrial enzyme, has higher activity in the presumptive oral ectoderm (Czihak,1963). More recently, it was shown that if embryos are cultured in tight clusters, a redox gradient forms across the embryo and the most oxidative portion of this gradient becomes the oral ectoderm (Coffman and Davidson,2001). It was also recently shown that there is a mitochondrial asymmetry in sea urchin eggs and early embryos, and the region with the highest number of mitochondria, and thus a more oxidative state, tends to become the oral side (Coffman et al.,2004). Since asymmetries in both respiration (Coffman and Davidson,2001) and mitochondrial distribution (Coffman et al.,2004) influence OA specification, this suggests that the intracellular redox state influences OA axis specification and ectoderm polarization.

Studies also suggest that redox signaling might act through Nodal. The redox-sensitive kinase, p38 MAP kinase (p38), functions upstream of Nodal signaling (Bradham and McClay,2006) and p38 is known to be activated by oxidative stress and in turn activates basic-leucine zipper (bZIP) transcription factors (Clerk et al.,1998; Inoue et al.,2005). A bZIP-binding site was identified in the nodal gene regulatory region involved in nodal transcriptional initiation (Nam et al.,2007; Range et al.,2007). Since bZIP transcription factors are redox sensitive (Amoutzias et al.,2006) and respond to p38 (Inoue et al.,2005), this suggests redox signaling might influence OA patterning through nodal.

If redox signaling influences OA axis specification and nodal expression, treatments that alter ROS should influence OA axis specification and nodal transcription. One of the few exogenous treatments known to affect OA axis specification in sea urchin embryos is nickel (Ni). Ni generates oralized embryos (Hardin et al.,1992), which resemble those obtained by overexpression of Nodal (Hardin et al.,1992; Duboc et al.,2004; Flowers et al.,2004). Our previous work demonstrated that treatment with either Ni or cobalt (Co) upregulates nodal transcription (Agca et al.,2009). This is relevant because Ni and Co have been shown to generate ROS (Chandel et al.,1998; Salnikow et al.,2000; Kawanishi et al.,2002; Cavallo et al.,2003; Pourahmad et al.,2003) and to stabilize hypoxia inducible factor (HIF-1α) in other systems (Namiki et al.,1995; Chandel et al.,1998; Salnikow et al.,2000; Maxwell and Salnikow,2004). Hypoxic conditions (O2 concentrations lower than 21%) also stabilize HIF-1α (Wang GL et al.,1995) and generate ROS in a mitochondria-dependent manner (Chandel et al.,1998; Schumacker,2003; Brunelle et al.,2005; Guzy et al.,2005; Mansfield et al.,2005). Hypoxia partially inhibits mitochondrial electron transport, producing redox changes in the electron carriers that increase the generation of ROS (Chandel et al.,1998). Moreover, isolated mitochondria generate increased ROS during hypoxia (Chandel et al.,2000).

Sea urchin embryos cultured under hypoxic conditions have an improperly specified OA axis and reduced nodal transcript expression (Coffman et al.,2004; Coffman and Denegre,2007). Since hypoxia can increase ROS and ROS promotes p38 MAPK activity (Kulisz et al.,2002; Torres and Forman,2003), which is required upstream of Nodal (Bradham and McClay,2006), the reported reduction of nodal transcript levels after hypoxia argues that altering ROS negatively influences nodal expression. We, therefore, undertook a more detailed analysis of the effects of hypoxia using defined O2 levels and treatments that generate ROS by other means.

Intracellular ROS can be generated by treatment with the broad spectrum herbicide paraquat (1,1′-dimethyl-4,4′-bypyridylium dichloride; Bus et al.,1976). Paraquat is reduced in cells by a variety of enzymes to the radical cation, which is reoxidized by molecular O2 to the superoxide anion (Margolis et al.,2000; Tsukamoto et al.,2002). Treatment with paraquat generates ROS that include hydrogen peroxide (H2O2) and the hydroxyl (-OH) radical (Tsukamoto et al.,2002). Paraquat induces oxidative stress and affects redox signaling in both Caenorhabditis elegans (An and Blackwell,2003) and Drosophila melanogaster (Girardot et al.,2004), but its effects have not been investigated in sea urchin embryos.

We investigated the effects of hypoxia and paraquat treatment on early sea urchin embryos using morphological criteria, immunohistochemical analyses with tissue specific antibodies, and analysis of gene expression using quantitative reverse transcriptase PCR (Q-RT-PCR). We found that continuous exposure of sea urchin embryos to hypoxia or paraquat blocked specification of the ectoderm and downregulated genes expressed in the different ectodermal territories, but did not appear to influence specification of other territories. Our results suggest that elevating ROS levels by two different treatments influences the differentiation of the ectoderm in sea urchin embryos but does not specifically activate Nodal transcription.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The Effects of Culturing Embryos in Reduced O2

Embryos were cultured in an atmosphere of O2 ranging from 21% (ambient) to 0% and compared to untreated control embryos at the pluteus stage (72 hr after fertilization). Overt morphological defects were not observed until embryos were grown in an O2 environment of less than 2% O2. When cultured in a 1% O2 (Fig. 1B), embryos are smaller than untreated controls and shortened along their OA axis (Fig. 1A). The reduction in size gradually increased as the concentration of O2 was reduced further to 0.5% O2 and 0.2% O2 (Fig. 1C and D). Embryonic lethality was observed when embryos were cultured in 0% O2 (data not shown).

thumbnail image

Figure 1. Morphological effects of culturing embryos in reduced O2. S. purpuratus embryos were cultured until the pluteus stage in reduced O2 environments (B,C) and compared to control embryos cultured in ambient O2 (A). No distinct morphological effect was observed until embryos were cultured in 1% O2 when they display a reduction in length along the OA axis (B). As the O2 concentration is further reduced to 0.5 % O2 (C), embryos show a more pronounced reduction in length. At 0.2% O2 (D), the lowest viable concentration examined, embryos are round, but still display a normal-appearing archenteron and endoderm.

Download figure to PowerPoint

In addition to a reduction in overall size, spicule formation is inhibited at low O2 concentrations (Fig. 2A). Although skeletogenic mesenchyme cells are present in embryos cultured in the lowest concentration of O2 examined, the orientation of the skeletogenic primary mesenchyme cells (PMCs) was perturbed (arrow; Figs. 2A and 3F). This suggests a potential defect in ectoderm patterning, because PMCs in the blastocoel obtain cues for their proper position from the overlying ectoderm (McClay,1999). Continuous culture in 0.2% O2 resulted in the complete absence of spicules (Fig. 2A). This is similar to the inhibition of spiculogenesis that we observed with Co treatment (Agca et al.,2009), which is thought to induce hypoxic responses in cells (Namiki et al.,1995; Salnikow et al.,2000; Maxwell and Salnikow,2004). The severity of the defect in spiculogenesis is dependent on the length of time in a reduced O2 environment. Spiculogenesis is restored if embryos are returned to an ambient O2 environment before the late gastrula stage (Fig. 2B), but greater recovery is achieved if embryos are cultured for shorter lengths of time (Fig. 2C).

thumbnail image

Figure 2. Culturing embryos for different lengths of time in 0.2% O2 reveals a progressive effect on spicule formation. A: Spicule formation is inhibited when embryos are cultured continuously until the pluteus stage in 0.2% O2. At this concentration, embryos are reduced in size and the skeletogenic mesenchyme cells form a ring around the blastopore on the vegetal side of the embryo (arrow in A). Embryos treated for shorter lengths of time, until the late gastrula stage (B) or mesenchyme blastula stage (C), recover from the reduced O2 and the effect on spiculogenesis is correspondingly reduced. D: Control embryos. Left panels are differential interference optic (DIC) images and right panels are polarization optic images of the same embryo. Inset in A shows a vegetal view to reveal the misaligned PMCs.

Download figure to PowerPoint

thumbnail image

Figure 3. Immunostaining of embryos cultured in 0.2% O2 with territory specific markers. Embryos were labeled with antibodies that recognize the oral ectoderm (Ecto-V), aboral ectoderm (Spec1), endoderm (Endo1), and skeletogenic mesenchyme (1d5) at the pluteus stage. Treated embryos (B) show an expansion in Ecto-V (red) staining while Spec1 staining (green) is diminished in these embryos (B, D, F) compared to controls (A, C, E). Endoderm specification does not appear to be affected by culture in 0.2% O2 as the extent of Endo1 labeling (red) is comparable in treated (D) and control embryos (C). Although PMCs are specified based on 1d5 labeling (red), their organization is perturbed in treated (F) embryos compared to controls (E). All embryos are counterstained with the nuclear stain Draq5 (blue).

Download figure to PowerPoint

Immunohistochemical Analysis of Territory-Specific Markers in Embryos Cultured in 0.2 % O2

Because culture in an 0.2% O2 environment suggested that ectodermal specification was altered, immunostaining was performed with different territory-specific markers to determine which embryonic territories are perturbed. Embryos were cultured in 0.2% O2 until the pluteus stage, then fixed and immunostained with either the oral ectoderm-specific antibody, Ecto-V (Coffman and McClay,1990), a PMC-specific monoclonal antibody, 1d5, or an endoderm-specific antibody, Endo1 (McClay et al.,1983; Wessel and McClay,1985), and co-labeled with the aboral ectoderm-specific antibody, Spec1 (Wikramanayake et al.,1995).

When embryos were cultured in 0.2% O2 and immunostained at the pluteus stage, labeling with Ecto-V was expanded relative to Spec1 staining (Fig. 3B) when compared to controls (Fig. 3A). In addition, Spec1 staining in 0.2% O2-treated embryos is severely diminished (Fig. 3B, D, and F) compared to Spec1 staining observed in controls (Fig. 3A, C, and E). In contrast, Endo1 staining (Fig. 3D) is comparable to that observed in control embryos (Fig. 3C). Labeling of the PMCs with 1d5 monoclonal antibody (Fig. 3F) shows that despite the absence of spicules, at this concentration of O2 PMCs are present, but they are not aligned along bilaterally arrayed skeletal elements (Fig. 3F) as they are in controls (Fig. 3E).

Analysis of Gene Expression Levels in Embryos Cultured in 0.2% O2

Immunostaining of embryos cultured in 0.2% O2 showed expanded Ecto-V and diminished Spec1 labeling compared to that observed in controls, which is indicative of a perturbation in ectoderm specification (compare Fig. 3A and B). This could mean that oral ectoderm was expanded at the expense of aboral ectoderm as is seen with Ni and Co treatment (Agca et al.,2009), as Ecto-V is confined to oral ectoderm at the late gastrula stage (Coffman and McClay,1990), or that the ectoderm is not specified, as Ecto-V was shown to label unspecified ectoderm in isolated animal caps (Wikramanayake et al.,1995), and is expressed uniformly at early stages of development (Coffman and McClay,1990). To further characterize this alteration in ectoderm, Q-RT-PCR analysis was performed to assess the levels of expression of ectoderm-specific genes. Q-RT-PCR analysis revealed that all ectoderm-specific genes examined were downregulated compared to controls at both 18 and 24 hr of development in embryos cultured in 0.2% O2 (Fig. 4).

thumbnail image

Figure 4. The expression levels of genes known to influence ectodermal specification are reduced in embryos cultured in 0.2 % O2. Genes expressed in different ectodermal territories were examined in embryos after 18 and 24 hr of culture in 0.2% O2. Transcript levels for nodal, as well as the downstream targets of nodal, BMP2/4, and lefty, are moderately reduced in 0.2% O2-cultured embryos. In contrast, gsc transcript levels are severely downregulated at 24 hr and the aboral ectoderm–specific genes, cyIIIa, tbx2/3, and spec1 are reduced at both 18 and 24 hr. When 50 or 500 μM Ni are combined with culture in 0.2% O2, all oral ectoderm genes analyzed showed higher values compared to 0.2% O2 treatment alone. Yellow bars indicate oral ectoderm genes and green bars aboral ectoderm genes. The order of genes in each treatment is represented in the graph in the top left corner. The gray transparent region represents control levels; anything above or below that region indicates upregulation or downregulation, respectively.

Download figure to PowerPoint

The oral ectoderm genes examined include the oral ectoderm–specific gene, nodal (Duboc et al.,2004; Flowers et al.,2004), and the downstream targets of nodal, BMP2/4 (Duboc et al.,2004), lefty (Duboc et al.,2005), and goosecoid (gsc) (Angerer et al.,2001). Aboral ectoderm–specific genes were also downregulated in these embryos. Those examined include: tbx2/3 (Gross et al.,2003), cyIIIa (Kirchhamer et al.,1996), and spec1 (Hardin et al.,1985). These data suggest that the expression of all ectoderm-specific genes are reduced by 0.2% O2 treatment, and that low O2 conditions do not oralize embryos in the same way as Ni and Co (Agca et al.,2009). This contrasts with previous reports on the effects of reduced O2 on aboral ectoderm gene expression, but confirms its effects on oral gene expression (Coffman et al.,2004). The results indicate that sea urchin embryos cultured in a low O2 environment form ectoderm, but the expression of genes involved in further specification and differentiation is repressed.

The Effects of Culture in 0.2% O2 Combined With Ni Treatment

We previously reported that Ni treatment upregulates the expression of several oral genes (Agca et al.,2009). In other systems, Ni has been shown to act as a hypoxia mimetic by inducing the HIF-1α response (Namiki et al.,1995; Salnikow et al.,2000; Maxwell and Salnikow,2004). We were, therefore, interested in determining whether Ni treatment would overcome the effects of culturing embryos under reduced O2 conditions or if it would augment the response. We determined the level of expression of ectodermal genes in embryos cultured in 0.2% O2 and compared the levels of expression with those of embryos cultured in 0.2% O2 with 50 or 500 μM Ni by Q-RT-PCR. The levels of oral gene expression were consistently higher in embryos cultured in 0.2% O2 combined with either 50 or 500 μM Ni than those in 0.2% O2 alone. However, the levels were not as high as those observed in embryos after treatment with Ni alone (Fig. 4). This suggests that Ni can overcome the negative effects of a low O2 environment and it does not appear to act as a hypoxia mimetic in sea urchin embryos.

Increasing ROS in Embryos Using Paraquat

Immunostaining (Fig. 3) and Q-RT-PCR (Fig. 4) suggest that overall ectodermal development is suppressed under low O2 conditions. However, the generation of ROS is only one of the potential consequences of a low O2 environment. Because hypoxic exposure can elicit a range of physiological and metabolic responses (Reiber,1995; Semenza,1999; Harper and Reiber,2006), we also treated embryos with paraquat to determine whether this method of influencing redox signaling by generating ROS would yield similar results to those observed with embryos cultured in 0.2% O2.

To investigate the morphological effects of paraquat on early sea urchin development, we exposed embryos to increasing concentrations of paraquat chloride (Fig. 5A–D). Embryos were cultured continuously from fertilization until the pluteus stage in different concentrations of paraquat and compared to untreated controls. Alterations in morphology were first observed at 5 mM paraquat, where embryos become shortened along their OA axis (Fig. 5B) compared to untreated controls (Fig. 5A). As the concentration of paraquat increased, the length along the OA axis of embryos was reduced in a graded manner (Fig. 5B–D). Although embryos retain their bilateral symmetry at 25 mM paraquat, the size of their skeletons was reduced (Fig. 5C). At 50 mM paraquat (Fig. 5D), embryos were round and their spicules were short and disoriented. At all concentrations examined an invaginated gut and mesenchyme cells were observed in the blastocoel (Fig. 5A–D).

thumbnail image

Figure 5. Paraquat treatment results in a graded loss of aboral ectoderm specification. Embryos were treated from fertilization through the pluteus stage with increasing concentrations of paraquat. A: Concentrations of paraquat between 5 and 50 mM caused a reduction in the size of embryos in a dose-dependent manner compared to controls. Spicules are reduced in size relative to the reduction in aboral ectoderm at 5 mM (B) and 25 mM (C) paraquat. However, at 50 mM paraquat (D), spicules were severely reduced in size and the PMCs misaligned. The top panels are DIC images and the bottom panels are polarization optic images of the same embryo to show the extent of spicule development.

Download figure to PowerPoint

The Effects of Paraquat on Different Embryonic Territories

To determine the effects of paraquat on different embryonic territories, embryos were cultured in paraquat until the pluteus stage, then fixed and immunostained with the oral ectoderm- and aboral ectoderm-specific antibodies, Ecto-V and Spec1, and the PMC and endodermal antibodies, 1d5 and Endo1. Endo1 expression was observed at all paraquat concentrations examined (Fig. 6B–D), suggesting that endodermal specification was not affected. Although 1d5 expression indicated that PMCs were present at all concentrations of paraquat examined (Fig. 6E–H), at higher concentrations of paraquat, PMCs were disorganized within the blastocoel (Fig. 6H).

thumbnail image

Figure 6. Immunohistochemical analysis reveals that endoderm and skeletogenic mesenchyme are correctly specified in paraquat-treated embryos. Paraquat-treated embryos were fixed and immunostained with Endo-1 and 1d5 to determine the effect of paraquat on endoderm and PMC specification. Endoderm is specified and archenteron formation occurs at all concentrations tested (BD). However, although PMCs are present and specified after 50 mM paraquat treatment (FH), their patterning is altered (H). Control embryos stained with Endo-1 and 1d5 are shown in panel A and E, respectively. Red immunostaining for each antibody is shown superimposed on the corresponding DIC image.

Download figure to PowerPoint

Ecto-V- and Spec1-positive immunostaining was observed in embryos cultured in 5 mM paraquat (Fig. 7B), but the extent of Spec1 staining was reduced compared to that in controls. In contrast, the extent of Ecto-V staining was comparable to that of controls (Fig. 7A). As the concentration of paraquat increased to 25 mM, the size of the aboral ectodermal territory was further reduced (Fig. 7C). At 50 mM paraquat, Spec1 expression was dramatically reduced and Ecto-V was expressed globally throughout the ectoderm (Fig. 7D). This suggested that paraquat effects ectodermal specification in a similar manner to 0.2% O2 treatment; as the concentration of paraquat increased, the region stained with Ecto-V antibody was expanded, while the region stained with Spec1 was reduced.

thumbnail image

Figure 7. Immunohistochemical analysis of ectoderm differentiation in paraquat-treated embryos shows dose-dependent reduction in aboral ectoderm–specific staining. Embryos were stained with the oral and aboral ectoderm–specific antibodies, Ecto-V (red) and Spec1 (green), after treatment with 5 mM (B), 25 mM (C), and 50 mM (D) paraquat and compared to untreated controls (A). Paraquat-treated embryos show reduced aboral ectoderm staining (B–D). At concentrations of 50 mM, Spec1 staining is severely diminished and Ecto-V is expressed globally throughout the ectoderm. The left panels are DIC images of the corresponding immunostained embryos.

Download figure to PowerPoint

Analysis of Gene Expression Levels in Paraquat-Treated Embryos

Both oral and aboral ectoderm genes were downregulated in embryos cultured in 0.2% O2, suggesting that the ectoderm was not properly specified (Figs. 3 and 4). Immunostaining with territory-specific markers showed that paraquat-treated embryos had expression patterns that resembled those obtained when embryos were cultured in a 0.2% O2 environment. We next determined whether gene expression in paraquat-treated embryos was similar to that of embryos cultured in low O2. When the oral ectoderm-specific genes, nodal, gsc, and dri were analyzed, all three showed a graded downregulation as the concentration of paraquat increased (Fig. 8). The aboral ectoderm–specific genes, cyIIIa and spec1 also showed a reduction in transcript levels in paraquat-treated embryos. The expression levels for foxA and brachyury, which are initially expressed in the endomesoderm and later in the stomodeal region of the oral ectoderm (Gross and McClay,2001; Oliveri et al.,2006), were also downregulated at 50 mM paraquat.

thumbnail image

Figure 8. The expression of genes known to influence ectodermal and endodermal specification is altered in paraquat-treated embryos. Embryos were treated with increasing concentrations of paraquat and the level of gene expression determined at the pluteus stage. Expression of the oral ectoderm–specific genes, nodal, goosecoid (gsc), and deadringer (dri) show a dose-dependent downregulation and the aboral specific genes, cyIIIa and spec1 are similarly downregulated. FoxA and brachyury, which are expressed in the stomodeal component of the oral ectoderm, are also downregulated at 50 mM paraquat. However, the endoderm-specific genes, endo16, krox, and wnt8 were not affected at the concentrations tested. Otx-alpha, which is normally localized to the oral ectoderm and endoderm, is upregulated at all concentrations tested. For each gene examined, the paraquat concentration tested is presented in the following order: 5, 25, and 50 mM (graph at top left). The gray transparent region represents levels of expression in control embryos; values above or below that region indicate upregulation or downregulation, respectively.

Download figure to PowerPoint

Although immunostaining showed normal PMC and endoderm specification at high paraquat concentrations, transcript levels for many ectodermal genes were suppressed by paraquat. We were, therefore, concerned that paraquat treatment caused general toxicity. In contrast to other ectoderm genes, otx-alpha transcript expression was upregulated by paraquat treatment. In normal development, otx-alpha is uniformly expressed in the ectodermal territories at the blastula stage but is later restricted to the endoderm and oral ectoderm, and targets of otx-alpha are competitively inhibited by gsc (Mao et al.,1994; Wang et al.,1996; Li et al.,1997; Angerer et al.,2001). Downregulation of genes involved in specification and differentiation of ectoderm after paraquat treatment argues that increasing ROS levels by paraquat (and possibly 0.2% O2) allowed the initial specification of the ectoderm but inhibited the expression of later genes involved in ectoderm polarization.

To determine the effect of paraquat on endoderm-specific gene expression, endodermal genes were analyzed by Q-RT-PCR. Wnt8, which is required for endomesoderm specification and is expressed at later stages (Wikramanayake et al.,2004), was maintained at levels similar to controls at all concentrations of paraquat (Fig. 8). Endo16 and krox are similarly involved in endomesoderm specification with krox expression restricted to cells of the developing vegetal plate (Wang et al.,1996) and endo16 expression throughout the invaginating archenteron during gastrulation (Nocente-McGrath et al.,1989; Romano and Wray,2006). Transcript levels of endo16 and krox were not affected by paraquat treatment, consistent with the immunostaining observed with Endo1, and arguing that paraquat does not appreciably affect endoderm specification (Fig. 8).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

ROS and the oxidation of proteins have important roles in signal transduction (An and Blackwell,2003; Shibata et al.,2003; Chen et al.,2008). Several lines of evidence suggest that redox signaling influences ectoderm specification and OA axis patterning in sea urchin embryos. Redox signaling may act through the initiation of nodal transcription, as binding sites for redox-sensitive bZIP transcription factors have been identified in the cis-regulatory region of nodal (Range et al., 2006; Nam et al.,2007). In addition, Ni treatment, which mimics Nodal overexpression, has been shown to generate ROS and influence transcriptional activation of certain genes in a redox-dependent manner (reviewed in Lu et al.,2005; Das et al.,2008). We reasoned that altering ROS levels by various treatments might similarly influence ectoderm specification and alter nodal expression. However, we found that altering ROS levels affects all ectoderm-specific gene expression, but did not severely affect gene expression in other territories. Embryos cultured in 0.2% O2 or treated with paraquat, both of which have the potential to generate ROS in cells (Tsukamoto et al.,2002; Schumacker,2003), downregulated expression of nodal and other oral and aboral ectoderm–specific genes.

Previous studies reported that culturing sea urchin embryos under hypoxic conditions resulted in reduced nodal transcript expression, but no effect was observed on aboral ectoderm–specific genes (Coffman et al.,2004). We found that culturing sea urchin embryos in 0.2% O2 also influences ectoderm specification, but not endoderm or skeletogenic mesenchyme specification, since both endoderm and PMCs are formed and express territory-specific markers. We also found that spiculogenesis is disrupted by culture in a low O2 environment, presumably due to the disruption of normal ectoderm polarity. Our results are consistent with the previous report that culturing embryos in low O2 reduced oral ectoderm-specific gene expression (Coffman et al.,2004), since we observed that the oral genes nodal, BMP2/4, lefty (Duboc et al.,2004; Flowers et al.,2004), and gsc (Angerer et al.,2001) were reduced in embryos cultured in 0.2% O2. However, our results differ from the results of Coffman et al. (2004), as we found all the aboral ectoderm–specific genes that were analyzed; cyIIIa, tbx2/3, and spec1 (Hardin et al.,1985; Kirchhamer et al.,1996; Gross et al.,2003) were also downregulated after culturing embryos in 0.2% O2 (Fig. 4). These differences may be due to the differences in the experimental conditions used in Coffman et al. (2004). In the present study, the observed effects clearly depended on the concentration of O2 used and the length of exposure. Embryos cultured in 0.2% O2 show expanded expression of the oral ectoderm marker, Ecto-V (Coffman and McClay,1990), while expression of the aboral marker, Spec1 (Wikramanayake et al.,1995), is diminished. Transcript levels of oral ectoderm-specific genes were also reduced, suggesting that 0.2% O2 treatment, unlike Ni, did not oralize embryos. Aboral ectoderm genes were also downregulated by 0.2% O2. However this reduction is not due to expansion of the oral ectoderm as is seen in Ni-treated embryos. Expansion of Ecto-V staining and the reduced oral ectoderm gene expression in embryos cultured with low O2 concentrations are counterintuitive. A possible explanation for the expansion of Ecto-V in the absence oral gene expression may be that the ectoderm is in a non-polarized state. When animal half embryos are isolated from 8-cell-stage embryos by microsurgery and cultured, they fail to polarize into oral and aboral ectoderm. In these manipulated embryos, Ecto-V expression is expanded throughout the ectoderm (Wikramanayake et al.,1995). Thus, the global expression of Ecto-V that we observe in treated embryos may be indicative of unspecified ectoderm.

Another way of altering redox signaling is to generate ROS by treatment with paraquat (Bus et al.,1976; Tsukamoto et al.,2002). We chose concentrations of paraquat that have previously been reported to affect development of C. elegans and D. melanogaster (An and Blackwell,2003; Clark et al.,2006). We observed that increasing ROS with paraquat mimics the effects observed after culture in low O2 conditions. Paraquat-treated embryos also show expanded Ecto-V expression and diminished Spec1 expression and transcript levels for most ectoderm genes, both oral and aboral, were reduced. These results argue that ectoderm specification is affected by paraquat in a similar manner to that observed when embryos are cultured in a reduced O2 environment, and also suggest that the effects of reduced O2 and paraquat on ectoderm specification are probably related to elevated ROS levels in embryos.

Nonetheless, elevation of ROS does not inhibit all ectodermal genes. Our data indicate that otx-alpha gene expression is upregulated in paraquat-treated embryos. Otx-alpha is a transcriptional activator that is uniformly and ubiquitously expressed in ectodermal territories at the blastula stage but is restricted to the gut and oral ectoderm at later stages (Li et al.,1997). In both paraquat and 0.2% O2-treated embryos, ectoderm forms but does not become polarized into oral and aboral domain based on immunostaining with territory-specific antibodies and the reduced expression of both oral and aboral ectoderm–specific genes. The elevation of otx-alpha transcript levels after paraquat treatment is probably the result of the disruption of normal ectoderm polarity since endoderm appears morphologically intact and patterned based on the expression of region-specific antigens and the expression of other endoderm-specific genes. However, we do not exclude the possibility that otx-alpha expression is elevated in the endoderm in paraquat-treated embryos as a result of the disruption in ectoderm polarity.

Our results indicate that nodal transcription is reduced in 0.2% O2-treated embryos, as are the downstream targets of nodal, BMP2/4, lefty, and gsc. This raises the possibility that elevating ROS over-stimulates mechanisms that normally prevent expansion of Nodal signaling to other embryonic territories. In control embryos, it is thought that negative feedback between Nodal signaling and the Nodal antagonist Lefty blocks the expansion of Nodal signaling throughout the ectoderm (Duboc et al.,2008) as Lefty competes with Nodal for the Nodal receptor (Sakuma et al.,2002; Schier,2003; Duboc et al.,2004,2005). Since Lefty is a downstream target of Nodal and our results indicate that lefty transcript levels are downregulated when embryos are cultured in 0.2% O2, the reduction of oral ectoderm-specific genes is unlikely to be related to an antagonistic effect of Lefty and must be through some other mechanism. Another possible antagonistic mechanism may operate via the Dr-1/DrAP-1 complex, which has been known to repress nodal by attenuation of its autoregulatory feedback loop (Iratni et al.,2002). The activity of Dr-1/DrAP-1 is enhanced in human cell lines cultured in 0.2% O2 (Denko et al.,2003). This is similar to our experimental conditions and correlates with our results, since embryos cultured in 0.2% O2 resulted in reduced nodal transcript levels. Orthologs of DrAP-1 and Dr-1 are present in the S. purpuratus genome (XM_001193886, XM_793823), suggesting that this mechanism may operate in sea urchin embryos cultured in low oxygen to suppress the nodal autoregulatory feedback loop.

Specification of the oral ectoderm is thought to be dependent on an oxidative redox signal created by respiratory asymmetry in different blastomeres with a more oxidative state leading to the formation of oral ectoderm (Coffman and Davidson,2001). Although we show that hypoxia and paraquat treatments affect sea urchin development similarly, and both influence redox signaling through the generation of ROS (Tsukamoto et al.,2002; Schumacker,2003), the reduction of nodal transcript levels we observed after these treatments argues that an increase in ROS does not activate nodal. Our data argue that altered redox signaling through elevation of ROS levels suppresses ectodermal territory specification and does not oralize ectoderm. However, it was proposed that ectoderm specification and initiation of nodal transcription are influenced by an oxidative redox state through asymmetric mitochondrial inheritance (Coffman and Davidson,2001; Coffman et al.,2004; Nam et al.,2007; Range et al.,2007), since our treatments generate uniform levels of ROS, this may stimulate mechanisms that negatively regulate nodal or alter a necessary redox asymmetry.

While Ni treatment is also thought to generate ROS in cells (Kawanishi et al.,2002; Cavallo et al.,2003; Pourahmad et al.,2003), we show that Ni overcomes the effects of 0.2% O2 treatment on ectoderm gene expression. If Ni and hypoxia both act through the generation of ROS, then these combined treatments should further increase ROS levels. Since Ni appears to counteract the effects of 0.2% O2, this argues that the effects of Ni on Nodal signaling are not through the generation of ROS. The mechanism of Ni's action on Nodal and OA axis patterning, therefore, remains enigmatic. Transcription factors, matrix metalloproteinases, and heat shock proteins can all be induced by Ni (Salnikow and Costa,2000). Ni has also been shown to block calcium channels (Lee et al.,1999), to interfere with metal-dependent proteins (Hartwig et al.,2002), and to influence histone acetylation, methylation, and ubiquitylation (Broday et al.,2000; Golebiowski and Kasprzak,2005; Karaczyn et al.,2005,2006a,b) as well as DNA methylation (Lee et al.,1995). A recent microarray analysis of the effects of Ni on mammalian endothelial cells has shown that in addition to activating the HIF pathway, Ni stimulates the NF-kappa B signaling pathway and that these two independent signal transduction pathways are modulated by the activation status of p38 (Viemann et al.,2007). However, the precise role of NF-kappa B signaling in sea urchin embryos and the effects of Ni on this pathway remain to be determined.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Animals

Adult Strongylocentrotus purpuratus were obtained from Santa Barbara Biological (Venice, CA) or The Cultured Abalone (Goleta, CA). Adults were maintained at 10°C. Eggs and sperm were harvested after intracoelomic injection of 0.5 M KCl. Eggs were fertilized with a dilute suspension of sperm, washed, and the resulting embryos cultured in Millipore filtered artificial seawater (MFSW) at 14.5°C.

Hypoxia Treatment

Embryos were cultured until the pluteus stage in different O2 concentrations ranging from ambient (∼21%) to 0% (100% nitrogen). The different O2 concentrations were obtained using a Biospherix ProOx Model 110 controller unit (Biospherix, Redfield NY). The system contains three basic elements: the controller unit, the O2 sensor, and a C-Chamber. The controller unit is fully automated and works by sensing and displacing air that has higher O2 levels inside the chamber by releasing Nitrogen (N2) into the chamber. The controller unit processed O2 sensor readings, and the release of N2 into the C-Chamber was adjusted automatically. Prior to hypoxia experiments, the MFSW used for culture was treated with N2 for 2 hr inside the C-Chamber to displace the existing O2. The O2 sensor was calibrated twice with atmospheric air and with 100% N2, respectively. The accuracy of the sensor was in the range of ±0.1% O2. As 0% O2 was lethal, embryos were not cultured in less than 0.2% O2. The C-Chamber was placed in an incubator that was set to 14.5°C. The atmospheric pressure inside the chamber is always the same as the pressure outside the chamber. The chamber is not perfectly sealed to displace the air in the chamber with the N2. To assure that different concentrations of O2 did not alter the pH of the MFSW, the pH was determined before and after treatment and no difference was observed.

Immunohistochemistry

Embryos were fixed at the pluteus stage in 4% paraformaldehyde in phosphate buffered saline (PBS) and stained as described previously (Agca et al.,2009). Monoclonal antibodies Ecto-V (Coffman and McClay,1990), 1d5 (McClay,1983), and Endo1 (Wessel and McClay,1985) were obtained from Dr. David McClay, Duke University, and Spec-1 antibody was used as described in Wikramanayake et al. (1995). Nuclei were stained with Draq5 to show the general morphology of the embryos (Smith et al.,1999). Draq 5 was diluted 1:1,000 in PBS with 0.1% Tween20 (PBST) and added before the final wash. Immunostained embryos were mounted after the final wash with ProLong Gold Antifade Reagent, (Molecular Probes, Eugene, OR) and imaged using a Fluoview 500 Confocal system (Olympus America Inc., Center Valley, PA).

Q-RT-PCR Analysis of Sea Urchin Embryos

The Gen-Elute Mammalian Total RNA Purification Kit (Sigma, St. Louis, MO) was used to isolate RNA from ∼500–600 embryos according to the manufacturer's instructions. cDNA was prepared by using the Applied Biosystems RT reaction kit (Applied Biosystems, Foster City, CA). Q-RT-PCR reactions were conducted as described (Agca et al.,2009) using two different internal controls for each treatment, ubiquitin and SpZ12-1 (Nemer et al.,1991; Wang DG et al.,1995). Amplification reactions were performed on an ABI 7900 HT with ABI SYBR Green PCR 2× Master Mix (Applied Biosystems, Foster City, CA). Analyses were performed in triplicate from samples taken from three separate experiments. Primers with amplicon lengths between 100–200 bp were designed with Primer3 software (primer sequences for nodal, BMP2/4, lefty cyIIIa, tbx2/3, and spec1 are listed in Agca et al.,2009), which is available online at http://fokker.wi.mit.edu/primer3/input.htm. The primer pairs for deadringer, gsc, foxA, brachyury, krox, wnt8, otx-alpha, and endo16 were designed by the Davidson Lab (California Institute of Technology, Pasadena, CA) and details can be found at: http://www.spbase.org/SpBase/resources/methods/q-pcr.php.

Paraquat and Ni Treatments

Embryos were treated with concentrations of paraquat dichloride (Sigma, St. Louis, MO) ranging from 5 to 50 mM. A 1 M stock solution was prepared by dissolving paraquat in de-ionized water. NiCl2 (Sigma) was dissolved in de-ionized water to obtain a 10-mM stock solution, which was diluted in sea water to give 50- and 500-μM working concentrations Stock solutions were diluted in MFSW to give twice the desired working concentration and then an equal volume of fertilized eggs in MFSW was added to bring the final concentration to 1×. Embryos were cultured at 14.5°C until the pluteus stage.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Dr. Peter Cserjesi, Tulane University, and Brandee Ann Price, Baylor College of Medicine, for helpful comments on the manuscript. We thank members of the Davidson Laboratory at California Institute of Technology for assistance with Q-RT-PCR. We also acknowledge the Robert A. Welch Foundation (G-0010) for support to W.H.K. and the LSUHSC Enhancement Fund for support to J.M.V.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES