Transgenic frogs expressing the highly fluorescent protein venus under the control of a strong mammalian promoter suitable for monitoring living cells

Authors


Abstract

To easily monitor living cells and organisms, we have created a transgenic Xenopus line expressing Venus, a brighter variant of yellow fluorescent protein, under the control of the CMV enhancer/chicken β-actin (CAG) promoter. The established line exhibited high fluorescent intensity not only in most tissues of tadpoles to adult frogs but also in germ cells of both sexes, which enabled three-dimensional imaging of fluorescing organs from images of the serial slices of the transgenic animals. Furthermore, by using this transgenic line, we generated chimeric animals by brain implantation and importantly, we found that the brain grafts survived and expressed Venus in recipients after development, highlighting the boundary between fluorescent and nonfluorescent areas in live animals. Thus, Venus-expressing transgenic frogs, tadpoles, and embryos would facilitate their use in many applications, including the tracing of the fluorescent cells after tissue/organ transplantation. Developmental Dynamics 233:562–569, 2005. © 2005 Wiley-Liss, Inc.

INTRODUCTION

The use of fluorescent proteins such as a green fluorescent protein (GFP) allows the monitoring of dynamic processes in transgenic organisms (Chalfie et al., 1994). Thus far, transgenic organisms, including worm (Caenorhabditis elegans; Chalfie et al., 1994), the fish species zebrafish (Danio rerio; Amsterdam et al., 1995; Peters et al., 1995) and medaka (Oryzias latipes; Hamada et al., 1998), rodents such as mouse (Mus muscles; Ikawa et al., 1995) and rat (Rattus norvegicus; Takeuchi et al., 2003), fly (Drosophila melanogaster; Wang and Hazelrigg, 1994), slime molds (Dictyostelium; Hodgkinson, 1995), and plants such as maize and Arabidopsis thaliana (Hu and Cheng, 1995; Sheen et al., 1995), have been generated expressing GFP and examined for marker cells in several applications. In amphibians, both Xenopus laevis and tropicalis expressing GFP have been produced (Bronchain et al., 1999; Marsh-Armstrong et al., 1999; Hirsch et al., 2002). However, further investigations with respect to the tissues and cells expressing fluorescent proteins are necessary for these transgenic frogs. In Xenopus laevis, although larvae exhibit a transparent body during development, it is difficult to detect cells slightly expressing GFP because they express endogenous fluorescence that is not filtered out. Therefore, it is necessary to create a new transgenic line wherein exogenous and endogenous fluorescence are distinguishable.

Recently, many types of genetically engineered GFP mutants and variants, which emit blue, cyan, and yellow fluorescence, have been created. Among these, Venus, a variant of enhanced yellow fluorescent protein (EYFP), displays higher fluorescence due to its rate-limited and highly efficient maturation (Nagai et al., 2002). Venus can be used to monitor fluorescence in cells more easily than GFP. In this study, we have generated a novel transgenic Xenopus line expressing Venus. By placing Venus expression under the control of the strong CAG promoter, we generated transgenic frogs with detectable fluorescence in most tissues throughout development and in germ cells of both sexes. In addition, using an established novel system reconstructing a three-dimensional (3-D) fluorescent image, we displayed the fluorescing brain of transgenic animals. Finally, we performed transplant engraftment of brains derived from transgenic animals, resulting in the easy detection of the implanted brain after development. Thus, we present a new transgenic frog line that is more suitable for monitoring marker cells in living animals.

ABBREVIATIONS

CAG CMV enhancer/chicken β-actin CCD charge coupled device EYFP enhanced yellow fluorescent protein GFP green fluorescent protein ICCD intensified CCD MS-222 tricaine methanesulfonate polyA polyadenylation 3-D three dimensional 3D-ISM 3-D internal structure microscopy

RESULTS

Generation of Transgenic Frogs Carrying the Venus Fluorescence Gene and Germ Line Transmission

To generate a novel reporter animal, we constructed plasmid DNA carrying the transgene that consists of the CAG promoter, the Venus fluorescent gene and the rabbit β-globin polyA signal sequence (Fig. 1A). We generated transgenic Xenopus by following the method as described previously (Kroll and Amaya, 1996; Amaya and Kroll, 1999) and established two founder males expressing Venus. As sperm derived from transgenic males carry the transgene, we were able to generate F1 progeny by in vitro fertilization of wild-type eggs. Expression of Venus in these progeny was observed in the late neurula stage and expanded to the entire animal at the tail bud stage (data not shown). This expression pattern was later than that of transgenic animals regulated by the CMV promoter (data not shown). However, a significant amount of Venus protein throughout the body was observed in tadpoles (Fig. 1B and Supplementary Figure S1, which can be viewed at http://www.interscience.wiley.com/jpages/1058-8388/suppmat). The intensity of Venus fluorescence in transgenic animals (upper animal in Fig. 1Ba) is clearly brighter than that of the GFP3 fluorescent protein in transgenic animals (lower animal in Fig. 1Ba and in b) under the control of the Xenopus cardiac actin promoter. Thus, Venus was highly expressed in transgenic animals. We designated these established transgenic lines as CAG-Venus and XCar-GFP3, respectively.

Figure 1.

Generation of transgenic animals expressing Venus. A: Structure of the constructed transgene. The transgene CAG-Venus is composed of the fluorescence reporter gene Venus and the rabbit β-globin polyA signal sequence under the control of the CAG promoter. Another transgene, XCar-GFP3, is composed of the fluorescence gene GFP3, the SV40 polyA signal sequence, and the Xenopus cardiac actin promoter (Mohun et al., 1986). B: Fluorescent analyses of the transgenic animals. Images of transgenic tadpoles carrying the CAG-Venus (upper animal in a) or the Xcar-GFP3 transgene (lower animal in a, and b) at stage 50 were captured as both brightfield (left panels) and fluorescence (right panels) under the microscope.

Histological Analysis of Venus Protein in Transgenic Tadpoles

We examined the expression profile of Venus in larvae by histological analysis. We prepared sections of the whole body of CAG-Venus tadpoles (stage 47) and analyzed Venus fluorescence under the microscope (Fig. 2). Venus was detected in a variety of tissues such as the brain, digestive organs, eye, fin, gill, heart, muscle, notochord, and spinal cord. We also observed Venus in the dorsal root ganglia and vessels (data not shown). Thus, Venus is expressed in most tissues in tadpoles during embryonic development. The CAG promoter is frequently used for ubiquitous expression of transgenes in the mice (Okabe et al., 1997; Kawamoto et al., 2000). Beyond our expectations, we found that this promoter is able to drive the ubiquitous expression of transgenes even in amphibians.

Figure 2.

Histological analysis of CAG-Venus transgenic tadpoles. AF: Longitudinal (A–E) and transverse (F) sections prepared from transgenic tadpoles expressing Venus at stage 47 were taken as fluorescence images under the microscope. b, brain; d, digestive organs; e, eye; f, fin; g, gill; h, heart; m, muscle; n, notochord; s, spinal cord. Scale bars = 200 μm in A–F.

3-D Analysis of Transgenic Animals

To recognize the 3-D fluorescent image of tissues in transgenic animals, CAG-Venus tadpoles (stage 50) were examined by the 3-D internal structure microscopy (3D-ISM) system. This system is able to reconstruct 3-D fluorescent images of fluorescing samples as described previously (Yokota et al., 1988). As shown in Figure 3A, we captured sliced surface images of the head of transgenic animals. We chose the area containing the brain (Fig. 3B), and stored serially sliced 100 images of samples. We then reconstructed a 3-D fluorescent image as shown in Figure 3C. We could show an image of the compacted and brightening brain. We also observed several skeleton-like structures highly expressing Venus (arrows in Fig. 3B,C). Thus our system enables the reconstruction of the structures of the fluorescing tissues.

Figure 3.

Three-dimensional fluorescence analysis of CAG-Venus transgenic animals. A,B: A transgenic tadpole at stage 50 embedded in OCT compounds was sliced, and images were taken. Images containing the head were arranged (A). Among these images, the area including the brain is magnified (B), indicated by the red box in A. C: This three-dimensional fluorescent image of the sample from the area indicated by the dotted line in B and its serial images was reconstructed as described in the Experimental Procedures section. Arrows indicate the bright portion observed in the sliced surface image. Scale bars = 100 μm in B,C.

In Vitro Assays for Fluorescence in Transgenic Animals

We examined tail extracts prepared from wild-type, CAG-Venus, and XCar-GFP3 tadpoles (stage 58) for the amount and fluorescence intensity of fluorescent proteins. We detected high amounts of Venus proteins in extracts prepared from the tail of transgenic animals due to the strong CAG promoter (Supplementary Fig. S2A). The amount of fluorescent proteins was 10 times higher in CAG-Venus animals than in Xcar-GFP3 animals. Furthermore, we compared the fluorescence intensity between Venus and GFP3 fluorescent proteins at the same concentration (Supplementary Fig. S2B). Clearly, Venus showed higher intensity with the sharp peak compared with GFP3 at around 530 nm. Thus, these data verified that the easy detection of fluorescence in our transgenic animals is due to the combination of the strong promoter and bright fluorescent protein.

Venus Expression in Tissues of Juvenile Frogs

We examined CAG-Venus transgenic animals that had finished metamorphosis. Under the microscope, a juvenile frog showed high fluorescent intensity in both dorsal and ventral sides of its whole body (Fig. 4A,B). Other transgenic frogs also showed a similar expression pattern and fluorescent intensity. These data suggested that high amounts of Venus are produced in the muscle and skin. We dissected several tissues from these transgenic frogs and examined fluorescence (Fig. 4C–L). Venus was detected in all the examined tissues, including the brain, eye, hand, heart, kidney, liver, lung, ovary, testis, and thyroid gland. We also detected the expression of Venus in the digestive organs such as intestine and stomach (data not shown).

Figure 4.

Morphological and histological analyses of Venus fluorescence in the tissues of CAG-Venus transgenic frogs. A,B: Observation of the fluorescence in the whole body; the dorsal (A) and ventral (B) sides of the wild-type (left side) and transgenic (right side) juvenile frogs were taken for the brightfield (left panels) and fluorescence (right panels) images under the microscope. CL: Fluorescent analysis of Venus proteins in a variety of tissues dissected from transgenic animals; images of the brain (C), eye (D), hand (E), heart (F), kidneys (G), lobe of the liver (H), lung (I), thyroid gland (J), ovaries (K), and testis (L) were captured in brightfield (left panels) and fluorescence (right panels) under the microscope. M,N: Histological analysis of the reproductive organs; sections prepared from specimens of the transgenic ovary (M) and testis (N) were taken as fluorescence images and then stained with hematoxylin and eosin and photographed under the microscope. O, immature oocyte; Sc, spermatocytes; Sg, spermatogonia; Sp, sperm. OT: Immunohistochemical analysis of the brain; as indicated in O, transverse sections derived from the olfactory bulb (P), cerebrum (Q), diencephalons (R), mesencephalon (S) and hypophysis (T) of the brain were immunostained with anti–green fluorescent protein antibody (red) and counterstained with Hoechst (blue). cp, choroid plexus; e, ependymal cell; gl, glomerulus; gr, granule cell; h, hypophysis; lv, lateral ventricle; m, mitral cell; n, nerve cell; tev, tectal ventricle; thv, third ventricle. Scale bars = 25 μm in M,N, 50 μm in O–T.

We further examined Venus expression by histological analyses of the dissected tissues (Fig. 4M,N and Supplementary Fig. S3). Venus was brightly detected in most cells of these tissues. Importantly, we found that oocytes in the ovary and spermatogenic cells in the testis clearly express Venus (Fig. 4M,N). To characterize Venus-positive cells in the brain, we investigated the expression pattern of Venus by immunohistochemical analysis with anti-GFP antibody and found that most of cells in the brain express Venus (Fig. 4O–T). Neurons and other cells in the olfactory bulb, cerebrum, diencephalons, mesencephalon, and hypophysis are Venus-positives. Thus, Venus was detected in most tissues and various cell types. Our results also indicate that the established transgenic lines express Venus from eggs to adults.

Transplant Engraftment With the Transgenic Brains

In this study, we noticed that most of cells in the brain from larvae to adults express Venus. Therefore, we performed graft transplantation using brains derived from CAG-Venus transgenic tadpoles following the manipulation and time schedule outlined in Figure 5A. One week after surgery, all tadpoles (n = 15) developed to stage 59. We examined the implanted brain in some recipients under the microscope (Fig. 5B,C). We could detect the fluorescence in the anterior region of the head in four of seven animals. Similarly, we detected some fluorescence under the heads of three frogs (of eight animals) that had finished metamorphosis at 3 and 4 weeks posttransplantation (Fig. 5D,E). Furthermore, we confirmed Venus expression in the implanted brain by histological analysis. We detected Venus in the restricted area in slice sections of the surgical brain (Supplementary Figure S4A). In addition, nuclear counterstaining with Hoechst dye clearly showed that cells derived from a donor exhibit normal morphology of nuclei, suggesting they are surviving after transplantation. Moreover, we examined the 3-D reconstruction of the brain based on Venus expression (Supplementary Figure S4B). By building up the 3-D fluorescent image from serially sliced surface images, we could show a donor brain fragment as a lump in the recipient. These data clearly indicated that the transplantation was successful in tadpoles. This finding suggests that the regeneration of the brain may occur by transplantation in amphibians. Thus, we could monitor the implanted tissues, suggesting that the Venus-expressing transgenic animals are useful as a reporter organism.

Figure 5.

Transplant engraftment with the brain from transgenic animals. A: Schematic diagram displaying the protocol for brain transplantation. Wild-type tadpoles (n = 15) at stages 55–58 were ectomized by removing the anterior one third of the forebrain, and a fragment of the whole brain from CAG-Venus transgenic tadpoles (stages 56–58) was implanted. After the transplantation, the recipient larvae were kept in water for 1, 3, and 4 weeks. BE: Morphological analysis of the implanted animals; heads of the implanted tadpoles (B,C) and frogs (D,E) were photographed in the light (left), fluorescent (middle), and magnified (right) fields under the microscope at 1 week (B,C), 3 weeks (D), and 4 weeks (E) after surgery. Arrows indicate an engrafted area.

DISCUSSION

In this study, we have generated a novel transgenic frog expressing the yellow fluorescent protein Venus under the control of the CAG promoter. We investigated in detail the expression profile of Venus in transgenic animals throughout development, and we found that Venus is detected in most tissues and from eggs to adults in this transgenic line. In addition, the fluorescent intensity of Venus in cells was very high, resulting in the acquisition of vivid fluorescent images in living animals and histological samples. Therefore, our established transgenic frogs expressing ubiquitously and brightly fluorescent proteins are useful as a reporter animal. Indeed, we confirmed the advantage of this transgenic line by applications such as the generation of chimeras and tissue transplantation. In particular, transplantation of brains from transgenic tadpoles demonstrated that the brain of amphibian larvae has a regenerative activity. Thus, we demonstrated the utility of our transgenic frogs for biomonitoring in living cells and organisms.

We propose the feasibility of using Venus as a reporter molecule in transgenic animals as follows. As Xenopus produces endogenously fluorescent proteins displaying the same spectra as that of GFP, it is difficult to distinguish between transgenic and nontransgenic cells when transgenic animals express low levels of GFP. In contrast, Venus does not completely overlap with endogenous fluorescent proteins; thereby, it is not difficult to detect the fluorescence of Venus in transgenic animals. Fluorescent images through a YFP filter were very clear. As described in a previous report (Nagai et al., 2002), Venus shows characteristics associated with the improved speed and efficiency of maturation. Furthermore, although YFP and its derivatives are acid-sensitive and quenched by chloride ions, Venus is tolerant to acidosis and halides, suggesting stable fluorescence in a wide range of environmental conditions (Nagai et al., 2002; Rekas et al., 2002). Thus, Venus exhibits many advantages compared with other fluorescent proteins.

It has been known that the CAG promoter powerfully drives transgene expression (Niwa et al., 1991). Using this promoter, transgenic mice display ubiquitous expression of GFP in the whole body of adults and in embryos at various developmental stages (Okabe et al., 1997; Kawamoto et al., 2000). In addition, it has been shown that this promoter induces transgene expression in mouse embryonic stem cells, whereas the CMV promoter could not (Niwa et al., 2000; and data not shown). In this study, we verified that the CAG promoter works in amphibians as well as mammals. By our histological analyses of reproductive organs, we detected the expression of Venus in eggs and sperm, suggesting the possibility of monitoring both somatic and germ cells during an animals' lifetime using our transgenic frogs.

It is now possible to visualize entire animal samples at high resolution in three dimensions (Ewald et al., 2002; Weninger and Mohun, 2002). Recently, for constructing the fluorescence image in three dimensions, we further improved a system established previously (Yokota et al., 1988). Our 3D-ISM system is feasible to acquire data rapidly and construct 3-D fluorescence images with more accuracy and high resolution. By using this system, we could reconstitute a 3-D fluorescent image of transgenic animals. In this study, we successfully showed the structural image of highly fluorescing tissues such as the brain. In addition, we were able to detect a fluorescing brain fragment derived from the transgenic donor in the recipient. Consequently, evidence from these results suggests that our system is able to create the 3-D fluorescence image and detect the fluorescing cells in the tissues or a whole-body.

In this study, we succeeded in the brain transplantation even in congenic frogs. This procedure depended on the growth of tadpoles used as recipients and donors. During stages 55–58, the immune system seems to be suppressed (Tozaki et al., 2000) thereby the rejection of donor cells did not occur drastically in the recipient. By histological analyses, however, we observed a damaged brain in several implanted animals. Therefore, it is necessary to use animals with the same genetic background in transplantation experiments for the further investigation. Despite difficulties and stress associated with transplantation, we detected that the engrafted brain fragments survive even after 3–4 weeks after surgery. Although the transplantation of adult brains has been performed, these studies have not been entirely successful (Margotta et al., 1993). By using our transgenic animals, we demonstrated the novel finding that the amphibian brain has regenerative activity from a morphological aspect. As a further study, we will examine whether the implanted brain recovers normal function.

The combination of Venus as a reporter and the CAG promoter for expression of the transgene enabled us to observe fluorescence in live cells, intact tissues and living animals without difficulty. Although we have generated transgenic frogs expressing EGFP under the control of the CMV promoter, they do not show strong fluorescent intensity compared with animals established in this study (data not shown). This significant advantage is not restricted to Xenopus but also applies to other animals. The enhancement of fluorescence enabled us to easily distinguish between fluorescent and nonfluorescent cells in these animals (data not shown). Thus, we are currently attempting to visualize the chimerism in living animals and to monitor the chimeric pattern from embryo to adult stage. Consequently, we believe CAG-Venus transgenic frogs would be useful for a variety of applications as a source of marker cells associated with the bright fluorescence to visualize events and phenomena throughout development.

EXPERIMENTAL PROCEDURES

Maintenance of Frogs

The African clawed frog Xenopus laevis and transgenic lines and tadpoles were maintained in water at 22°C. Developmental stages of animals were followed according to Nieuwkoop and Faber (1967). Transgenic progeny was obtained by in vitro fertilization of wild-type eggs with sperm derived from founder males as described previously (Suzuki et al., 1994).

Construction of the Transgene and Generation of Transgenic Frogs

For construction of a plasmid carrying the transgene named CAG-Venus, a DNA fragment encoding an YFP variant Venus (Nagai et al., 2002) and the rabbit β-globin polyadenylation (polyA) signal sequence were inserted into the pCAGGS expression vector (Niwa et al., 1991). The plasmid named XCar-GFP3 contains the gene encoding GFP3 (a variant of GFP) and SV40 ployA signal sequence driven by control of the Xenopus cardiac actin promoter (Mohun et al., 1986) and is a gift from Dr. E. Amaya. For generation of transgenic frogs, we followed the method established previously (Kroll and Amaya, 1996; Amaya and Kroll, 1999), but no restriction enzymes were added.

Fluorescent Stereomicroscopic Observation of Transgenic Animals and Their Tissues

Tissues, tadpoles, and frog fluorescence were analyzed under a dissecting microscope (MZFLIII, Leica Mycrosystems Wetzlar; SZ61-GFP-D, OLYMPUS, Tokyo) with charge coupled device (CCD) cameras (DC 300F, Leica Microsystems; Axiocam, Carl Zeiss GmbH, Jena, Germany; Orca C7780-22, Hamamatsu photonics, Hamamatsu, Japan).

Histological and Immunohistochemical Analyses of Transgenic Animals

Transgenic tadpoles at stage 47 and tissues dissected from young transgenic frogs were fixed in formaldehyde neutral buffer (Nacalai tesque, Kyoto, Japan) overnight, washed three times with phosphate buffered saline (PBS), and embedded in glycol methacrylate (Technovit 7100 or 8100, Heraeus Kulzer GmbH, Wehrheim, Germany). Sections of the specimens were prepared at a thickness of 8–10 μm, and images were taken with a microscope (Axioplan or Axioskop 2 plus, Carl Zeiss GmbH). In sections prepared from the testis and ovary, samples were stained with hematoxylin and eosin according to standard procedures.

To detect Venus protein in the brain of transgenic frogs immunohistochemically, brains dissected from animals were fixed in 4% paraformaldehyde/PBS at 4°C overnight, washed, dehydrated, and embedded in paraffin (Tissue Prep, Fisher Scientific, Pittsburgh, PA). Transverse sections were prepared at 10 μm, blocked in PBS containing 10% fetal calf serum and 0.3% Triton-X for 15 min. After removing the blocking solution, sections were incubated with anti-GFP rabbit polyclonal antibody (BD-Clontech, Palo Alto, CA; 1:20 in fresh blocking solution) for 2 hr at room temperature. Thereafter, the sections were washed in PBS containing 0.3% Triton-X and incubated with Cy3-conjugated anti-rabbit IgG goat antibody (1:800 in PBS; Chemicon International, Inc., Temecula, CA) for 2 hr at room temperature. After washing in PBS, samples were counterstained with Hoechst33342.

3-D Image Analysis of Transgenic Animals

CAG-Venus transgenic tadpoles at stage 50 were fixed in formaldehyde neutral buffer overnight, washed with PBS, and exchanged with PBS containing 20% sucrose. Samples were embedded in OCT compound and then analyzed with the 3D-ISM system. This system linking a computed drive cryotome, a fast-gated intensified CCD (ICCD) camera (ICCD300/DF, Hamamatsu Photonics), a confocal laser scanning unit (CSU-10, YOKOGAWA, Musashino, Japan), a laser disc recorder (LVR-3000AN, SONY, Tokyo, Japan), and a personal computer for image processing is able to reconstruct 3-D fluorescent images of fluorescing samples as described previously (Yokota et al., 1988). In brief, surfaces of each frozen and embedded sample were captured under the ICCD camera and repeated by slicing and removing with a rotary slicer driven by a computer. Serially sliced surface images were stored and processed with a PC. The 3-D reconstitution was performed by the ray casting method using 3-D visualization software (Voxcel Viewer, Toshiba Machine, Numazu, Japan). A similar operation was also performed in brain specimens that were engrafted with a brain fragment from CAG-Venus tadpoles.

Transplantation

Stages 55–58 wild-type tadpoles (n = 15) were anesthetized with 0.2% tricaine methanesulfonate (MS-222, Sigma Chemicals Co., St. Louis, MO) in water. For bulbectomy, their skin, dura mater, and meninges covering the forebrain were torn and the anterior one third of the forebrain gently aspirated through a needle. Two transgenic tadpoles at stages 56–58 were also anesthetized in 0.2% MS-222 solution, and the whole brains were dissected from these larvae by using a fine dissecting knife and fine forceps. Dissected brains were cut into small pieces and each fragment was implanted into the ectomized place of recipient wild-type animals. After the transplant, the larvae were left undisturbed for 30 min to allow the transplant to heal to the surrounding tissue. Larvae were kept to 2 days without bait in water for complete recovery. Animals were then fed and examined after 1, 3, and 4 weeks.

Acknowledgements

We thank Drs. E. Amaya (University of Cambridge) and J. Miyazaki (Osaka University) for gifts pXCar-GFP3 and pCAGGS plasmids; and Drs. A. Miyawaki (Riken), S. Tochinai (Hokkaido University), M. Maeno (Niigata University), M. Nozaki (Osaka University), H. Kubota, and H. Mori (Kyoto University) for their experimental advice.

Ancillary