Brief Communication
849
Inducible gene expression in transgenic Xenopus embryos Grant N. Wheeler, Fiona S. Hamilton and Stefan Hoppler The amphibian Xenopus laevis has been successfully used for many years as a model system for studying vertebrate development. Because of technical limitations, however, molecular investigations have mainly concentrated on early stages. We have developed a straightforward method for stage-specific induction of gene expression in transgenic Xenopus embryos [1,2]. This method is based on the Xenopus heat shock protein 70 (Xhsp70 [3]) promoter driving the expression of desired gene products. We found that ubiquitous expression of the transgene is induced upon relatively mild heat treatment. Green fluorescent protein (GFP) was used as a marker to monitor successful induction of gene expression in transgenic embryos. We used this method to study the stage specificity of Wnt signalling function. Transient ectopic Wnt-8 expression during early neurulation was sufficient to repress anterior head development and this capacity was restricted to early stages of neurulation. By transient over-expression at different stages of development, we show that frizzled-7 disrupted morphogenesis sequentially from anterior to posterior along the dorsal axis as development proceeds. These results demonstrate that this method for inducible gene expression in transgenic Xenopus embryos will be a very powerful tool for temporal analysis of gene function and for studying molecular mechanisms of vertebrate organogenesis. Address: Division of Cell and Developmental Biology, The Wellcome Trust Biocentre, University of Dundee, Dundee DD1 5EH, Scotland, UK. Correspondence: Stefan Hoppler E-mail:
[email protected] Received: 12 April 2000 Revised: 31 May 2000 Accepted: 31 May 2000 Published: 30 June 2000 Current Biology 2000, 10:849–852 0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
Results and discussion Wnt signalling function has been implicated in several processes during early Xenopus development [4]. However, the genes encoding Wnt ligands and their frizzled receptors are expressed in both early and later stage embryos [4]. It is not feasible to study the function of Wnt signalling components in later developmental stages via conventional methods such as mRNA injections [5], DNA construct injections [6] or even integration of constitutively expressing DNA constructs in transgenic embryos [1]. These manipulations promote expression from early
Figure 1 (a) HSP
Xwnt-8
(b)
HSP
GFP
(c) Con
Heat shock stage 13
(d)
(e) Exp
(f)
(g) Con
Heat shock stage 20
(h)
(i) Exp
Morphology
GFP
Current Biology
Induction of ubiquitous Xwnt-8 expression in transgenic Xenopus embryos. (a) An illustration of the DNA construct containing Xwnt-8 cDNA and GFP cDNA under the transcriptional control of the Xhsp70 promoter, which was used to generate transgenic embryos. Embryos that expressed GFP or were developing abnormally were removed from the experiment before induction (see Table 1). Normally developing embryos with no GFP expression were heat treated during (b–e) early (stage 13) and (f–i) late (stage 20) neurulation to induce transient gene expression. These embryos were fixed at stages (b–e) 29 and (f–i) 36. The same embryos were photographed under (b,d,f,h) visible light and (c,e,g,i) UV light. (d,e,h,i) Transgenic embryos with induced Xwnt-8 expression were (e,i) recognised by detecting GFP under UV light (note bluish green GFP colour, specifically in eye, fin and axial tissue) and separated from (b,c,f,g) non-transgenic embryos (note (c,g) only yellowish autofluorescence in yolky tissue). Note that development of the cement gland (the darkly pigmented organ at most anterior position) and anterior development were inhibited by transient ubiquitous Xwnt-8 expression in early neurula ((d); compare with (b)). Transient ubiquitous expression of Xwnt-8 at the end of neurulation did not visibly effect anterior development ((f); compare with (h)). Embryos are oriented with anterior to the left.
stages onwards and will severely disrupt axis or germlayer patterning in early embryos and therefore make functional analysis of later stage events impossible.
850
Current Biology Vol 10 No 14
Figure 2 (a)
gene expression of reporter genes such as chloramphenicol transferase (CAT) or GFP (for example, compare Figure 1e with c; Figure 2; see Materials and methods). We have designed our DNA constructs in such a way that the induction of gene expression of any desired cDNA is marked by the concomitant expression of GFP (for example Figure 1a). The transgenesis procedure we employed is not completely efficient and produces some embryos that have not integrated the transgene [2]. We exploited these embryos to serve as controls in the present study because they were detected by the absence of GFP expression and because they had been treated identically to those embryos that did integrate and express the transgene.
(b)
GFP-positive
GFP-negative
To test our inducible system in studying gene function in Xenopus development, we turned to Wnt signalling in the Xenopus neurula. Wnt over-expression in the dorsal ectoderm leads to a reduction of anterior structures [8]. mRNAmediated over-expression of Xenopus Wnt-8 (Xwnt-8) has been shown to mimic this effect by producing loss of dorso-anterior structures in embryos [9]. This activity has been suggested to reflect Wnt function during neurulation in antero-posterior patterning of the neural plate [10], but it has so far not been possible to test this prediction. To test whether the observed ability of ectopic Xwnt-8 to reduce dorso-anterior structures [9] required its expression before but not after neural tube formation, we induced stage specific Xwnt-8 expression in transgenic embryos (Figure 1; Table 1). We found that development of anterior structures was reduced by transient Wnt over-expression at early neurula stages. If, instead, Wnt was transiently over-expressed during later stages when the neural tube has formed, head development was not visibly affected. This finding confirms that the dorsal ectoderm is responsive to Wnt patterning at the neural plate stage. It further shows that this responsive phase is limited to the neural plate stage and does not extend to later stages of neurulation. Furthermore, our inducible expression system also
Current Biology
Induction of GFP expression in transgenic Xenopus embryos. Transgenic embryos were generated with a DNA construct containing Xwnt-8 cDNA and GFP cDNA under the transcriptional control of the Xhsp70 promoter (see Figure 1a). Normally developing embryos with no GFP expression were heat treated during early neurulation (stage 13) to induce transient gene expression. The embryos were fixed after neurulation (stage 21) and divided into GFP-expressing and non-expressing groups as detected under UV light. After sectioning, GFP expression in internal tissues was confirmed by detection with an anti-GFP antibody (see Materials and methods). Note (a) GFP protein expression in a section from a GFP-expressing embryo and (b) compare with the absence of GFP protein expression in a nonexpressing embryo. Sections are oriented with dorsal to the top.
A heat-shock promoter has been very successfully used for many years in Drosophila to induce ubiquitous gene expression [7]. We have developed a protocol for inducible gene expression in Xenopus transgenic embryos using the Xenopus heat-shock protein 70 promoter (Xhsp70 [3]). Because Xenopus is cold-blooded, we could temporarily raise the ambient temperature to induce ubiquitous gene expression under the control of Xhsp70. With only a few short heat treatments we were able to induce transiently Table 1 Phenotype of embryos overexpressing Xwnt-8 after gastrulation. Heat shock at stage
Experiment
n
% head reduced
% normal
None
GFP-positive before heat shock
13
76
24
13/14
GFP-positive after heat shock GFP-negative after heat shock
81 88
66 10
34 90
20/21
GFP-positive after heat shock GFP-negative after heat shock
34 75
13 0
87 100
A DNA construct containing Xwnt-8 cDNA and GFP cDNA under the transcriptional control of the Xhsp70 promoter was used to generate transgenic embryos (Figure 1a). Deformed embryos and embryos with uninduced GFP expression were separated and not subsequently heat treated. Where uninduced GFP expression was ubiquitous, embryos were severely deformed with little anterior or dorsal structures, probably due to concomitant constitutive and ubiquitous Xwnt-8
expression. All the other embryos were heat treated (see Materials and methods) at different stages of development to transiently induce Xwnt-8 expression. Xwnt-8 over-expressing embryos were identified by concomitant GFP expression. Embryos in which GFP expression could not be detected served as an internal control. The ‘head reduced’ phenotype consisted of absence of cement gland tissue and reduction of anterior head structures including developing eyes (Figure 1d).
Brief Communication
851
Figure 3
(a)
(b)
(c) Con
Exp
Heat shock:
Stage 13
Stage 20
Stage 25 Current Biology
Induction of ubiquitous Xfz7 expression in transgenic Xenopus embryos. A DNA construct containing Xfz7 cDNA and GFP cDNA under the transcriptional control of the Xhsp70 promoter was used to generate transgenic Xenopus embryos. Embryos that expressed GFP or were developing abnormally were removed from the experiment before induction. Normally developing embryos with no GFP expression were heat treated during (a) early neurulation (stage 13), (b) late neurulation (stage 20) and (c) shortly before the tailbud stage (stage 25) to induce gene expression. These embryos were fixed at approximately stage (a,b) 34 or (c) 36. Gained GFP expression (data not shown) distinguishes transgenic embryos with induced Xfz7 expression (Exp) from non-transgenic embryos (Con). Ubiquitous Xfz7 expression affected the length of the dorsal axis, but different parts of the dorsal axis were affected by Xfz7 over-expression at different stages of development. (a) During early neurulation, transient Xfz7
over-expression affected the development of dorsal axial structures at the level of the hindbrain (arrowhead). In the illustrated experiment, 53% of a total of 17 embryos with GFP expression had a strong anterior trunk phenotype, whereas 92% of 109 embryos without any detectable GFP expression were normal. (b) During late neurulation, however, transient Xfz7 over-expression affected dorsal axial structures at the level of the posterior trunk (arrowhead). In the illustrated experiment, 33% of a total of 18 embryos with GFP expression had a strong posterior trunk phenotype, whereas 95% of 126 embryos without any detectable GFP expression were normal. (c) Transient Xfz7 over-expression at a stage just before the tailbud formation affects the developing tail (arrowhead). In the illustrated experiment, 71% of a total of 14 embryos with GFP expression had a strong tail phenotype, whereas 84% of 72 embryos without any detectable GFP expression were normal. Embryos are oriented with anterior to the left.
allows us to study Wnt signalling during neurulation without the effects of Wnt over-expression during gastrulation. We can therefore show that Wnt signalling functions during neurulation to pattern the anteroposterior axis in the dorsal ectoderm separately from Wnt signalling during gastrulation, which functions in dorsoventral patterning of the mesoderm [11,12].
positions. Ectopic Xfz7 signalling thus interferes with the organisation of dorsal axial tissues in situ when they form, in addition to being able to disrupt gastrulation movements at an earlier stage [14].
We have recently cloned the novel Xenopus Wnt receptor frizzled-7 (Xfz7) and described its dynamic expression during Xenopus development [13]. mRNA-mediated overexpression of Xfz7 perturbs the morphogenesis of the trunk and tail [14]. As these effects were different from the ones caused by Xwnt-8 over-expression, Xfz7 is thought to function in a different Wnt pathway from Xwnt-8. It is not clear whether this mRNA-mediated effect is a consequence of Xfz7 over-expression during gastrulation or whether the exogenous Xfz7 mRNA and protein product were stable enough to disrupt morphogenesis at later stages. To address this question we over-expressed Xfz7 at different stages of development after gastrulation with our inducible gene expression system in transgenic embryos (Figure 3). We found that transient Xfz7 over-expression in early neurula caused defective dorsal axis development at the hindbrain level. Over-expression of Xfz7 at progressively later stages resulted in locally defective development of dorsal axis tissue at gradually more posterior
In conclusion, we present heat-inducible gene expression in transgenic Xenopus embryos that can be applied to the study of molecular mechanisms of vertebrate development. A method for hormone-inducible protein function has been used in Xenopus for some years [15]. It is applicable only to a relatively small group of proteins, which does not include secreted signals (like Xwnt-8) or membrane receptors (like Xfz7). The heat-inducible gene expression system can potentially be used for the over-expression of any gene product during Xenopus development. Our experiments with Xwnt-8 and Xfz7 provide an example for temporal analysis of gene function during Xenopus development that has not previously been possible. Thus, heat-inducible gene expression in transgenic Xenopus provides a powerful method for studying gene function during later stages of Xenopus development.
Materials and methods DNA constructs Experiments with CAT as reporter protein under the control of Xenopus hsp70 promoter (data not shown) were performed in embryos transgenic with the plasmid pHSP-CAT [16]. The base for all subsequent plasmids was pHS1, in which the hsp70 promoter (a HindIII–XbaI fragment from
852
Current Biology Vol 10 No 14
pHSP-CAT) replaced the simian cytomegalovirus (CMV) promoter in pCS2+ (a gift from Dave Turner and Ralph Rupp). Xwnt-8 cDNA [17] (a BamHI–NotI fragment from pCS2+Xwnt-8, a gift from Jeff D. Brown and Randall T. Moon) and GFP3 cDNA (a BamHI–NotI fragment from pCSGFP3 [18]) were separately subcloned into pHS1 to obtain HS1Xwnt-8 and HS1GFP3, respectively (see also [19]). pHS1Xwnt8HSG was created by combining plasmid sequences containing Xhsp70 and SP6 promoters upstream of Xwnt-8 cDNA and SV40 polyadenylation sequences (ScaI–NotI(blunt) from pHS1Xwnt-8) with plasmid sequences containing Xhsp70 and SP6 promoters upstream of GFP3 cDNA and SV40 polyadenylation sequences (HindIII(blunt)–ScaI from pHS1GFP3). To facilitate further experiments, we generated the pHSHSG2 by subcloning a fragment containing Xhsp70 and SP6 promoters upstream of GFP3 cDNA and SV40 polyadenylation sequences (HindIII(blunt)–Asp718 from pHS1GFP3) into pHS1 downstream of SV40 polyadenylation sequences. Xfz7 cDNA ([13], cut with SacI and BglII) was first subcloned into pGem-1 (Promega, cut with SacI and BamHI) and then integrated (as a EcoRI–HincII fragment) into pHSHSG2 (StuI–EcoRI digested) to generate pHSXfz7(S/B)HSG2.
Transgenic procedure In order to obtain embryos transgenic for these constructs we followed the procedures that were previously described [1,2]. Plasmids were linearised with either HindIII for pHSP-CAT or NotI for pHS1Xwnt-8HSG and pHSXfz7(S/B)HSG2.
Induction of gene expression Embryos were allowed to develop at 16°C to the desired stage for induction of gene expression. Prior to reaching the desired stage, embryos were removed from the experiment that were either deformed or express GFP uninduced (as observed with a stereomicroscope equipped with UV lamp and GFP filters; see Table 1). The uninduced GFP expression was probably due to enhancer trap integration of the transgene or due to cellular stress in deformed embryos. When embryos were kept at 16°C, we did not detect any background activity of the heat-shock promoter. For induction of gene expression, embryos were transferred between a precooled petri dish of 0.1´ MMR [20] at 16°C and a prewarmed petri dish of 0.1´ MMR [20] at 34°C or 37°C. Embryos between embryonic stages 13 [21] and approximately 20 [21] were treated 3 to 4 times for 15 min at 34°C with an interval of 15 min at 16°C. Embryos older than approximately stage 20 [21] were treated 3 to 4 times for 30 min at 37°C with an interval of 30 min at 16°C. Artefacts due to heat treatment were only observed in embryos that were treated during gastrulation stages (stages 10–12 [21]). These included incorrect or incomplete blastopore closure, which resulted in arrest of development at this late gastrulation stage or further development with spina bifida. We also observed a transient ruffling of the ectoderm, which did not appear to affect later development, however.
Detection of GFP expression Induced expression of the marker gene GFP was observed with a Leica stereomicroscope equipped with UV lamp and GFP filters. Maximal expression was usually observed 4–12 h after induction, but GFP was usually still visible in live embryos several days later. When embryos had reached the desired stage, they were fixed in MEMFA [20]. If fixed embryos were subsequently stored in PBS [20] for morphological analysis, GFP could still be observed. Experiments were documented with an RS Photometrics CoolSNAP digital camera and with Improvision Openlab and Adobe Photoshop software on a Macintosh G3 computer. Embryos for immunostaining were fixed in MEMFA and stored in PBS [20]. They were subsequently embedded in gelatin and frozen [22]. 10 mm sections were cut on a Leica Cryostat. The sections were immunostained with affinity purified sheep anti-GFP sera (a gift of Dr F. Barr, Glasgow University) and alkaline phosphatase-conjugated anti-sheep antibody (Sigma). Localisation was visualised using NBT and BCIP [22]. The result was documented on a Leica DMR microscope with an MTI3 CCD digital camera and with Improvision Openlab and Adobe Photoshop software on a Macintosh G3 computer.
Acknowledgements We would like to thank Enrique Amaya, Mariann Bienz and Andrea Münsterberg for encouragement and discussions in the initial phase of this project, Cheryll Tickle for comments on the manuscript and Andrew Bain for technical assistance. This research has been supported by The Wellcome Trust and by an equipment grant from The Royal Society.
References 1. Kroll KL, Amaya E: Transgenic Xenopus embryos from sperm nuclear transplantations reveal FGF signaling requirements during gastrulation. Development 1996, 122:3173-3183. 2. Amaya E, Kroll KL: A method for generating transgenic frog embryos. Methods Mol Biol 1999; 97:393-414. 3. Bienz M: Xenopus hsp 70 genes are constitutively expressed in injected oocytes. EMBO J 1984, 3:2477-2483. 4. Gradl D, Kuhl M, Wedlich D: Keeping a close eye on Wnt-1/wg signaling in Xenopus. Mech Dev 1999, 86:3-15. 5. Harvey RP, Melton DA: Microinjection of synthetic Xhox-1A homeobox mRNA disrupts somite formation in developing Xenopus embryos. Cell 1988, 53:687-697. 6. Christian JL, Moon RT: Interactions between Xwnt-8 and Spemann organizer signaling pathways generate dorsoventral pattern in the embryonic mesoderm of Xenopus. Genes Dev 1993, 7:13-28. 7. Lis JT, Simon JA, Sutton CA: New heat shock puffs and betagalactosidase activity resulting from transformation of Drosophila with an hsp70-lacZ hybrid gene. Cell 1983, 35:403-410. 8. McGrew LL, Lai CJ, Moon RT: Specification of the anteroposterior neural axis through synergistic interaction of the Wnt signaling cascade with noggin and follistatin. Dev Biol 1995, 172:337-342. 9. Tian Q, Nakayama T, Dixon MP, Christian JL: Post-transcriptional regulation of Xwnt-8 expression is required for normal myogenesis during vertebrate embryonic development. Development 1999, 126:3371-3380. 10. McGrew LL, Hoppler S, Moon RT: Wnt and FGF pathways cooperatively pattern anteroposterior neural ectoderm in Xenopus. Mech Dev 1997, 69:105-114. 11. Hoppler S, Brown JD, Moon RT: Expression of a dominant-negative Wnt blocks induction of MyoD in Xenopus embryos. Genes Dev 1996, 10:2805-2817. 12. Hoppler S, Moon RT: BMP-2/-4 and Wnt-8 cooperatively pattern the Xenopus mesoderm. Mech Dev 1998, 71:119-129. 13. Wheeler GN, Hoppler S: Two novel Xenopus frizzled genes expressed in developing heart and brain. Mech Dev 1999, 86:203-277. 14. Medina A, Reintsch W, Steinbeisser H: Xenopus frizzled 7 can act in canonical and non-canonical Wnt signaling pathways: implications on early patterning and morphogenesis. Mech Dev 2000, 92:227-237. 15. Kolm PJ, Sive HL: Efficient hormone-inducible protein function in Xenopus laevis. Dev Biol 1995, 171:267-272. 16. Ranjan M, Tafuri SR, Wolffe AP: Masking mRNA from translation in somatic cells. Genes Dev 1993, 7:1725-1736. 17. Christian JL, McMahon JA, McMahon AP, Moon RT: Xwnt-8, a Xenopus Wnt-1/int-1-related gene responsive to mesoderminducing growth factors, may play a role in ventral mesodermal patterning during embryogenesis. Development 1991, 111:1045-1055. 18. Bronchain OJ, Hartley KO, Amaya E: A gene trap approach in Xenopus. Curr Biol 1999, 9:1195-1198. 19. Marsh-Armstrong N, Huang H, Berry DL, Brown DD: Germ-line transmission of transgenes in Xenopus laevis. Proc Natl Acad Sci USA 1999, 96:14389-14393. 20. Sive HH, Grainger RM, Harland RM: Early Development of Xenopus laevis: a Laboratory Manual. New York: Cold Spring Harbor Laboratory Press; 2000. 21. Nieuwkoop PD, Faber J: Normal Table of Xenopus laevis (Daudin). Amsterdam: North-Holland Pub. Co.; 1967. 22. Stern CD: Immunocytochemistry of embryonic material. In Essential Developmental Biology. Edited by Stern CD, Holland PWH. Oxford: Oxford University Press; 1993:194-212.
