SEARCH

SEARCH BY CITATION

Keywords:

  • β-actin;
  • Cre/loxP;
  • medaka fish;
  • transgenic;
  • zebrafish

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Conditional cell labeling, cell tracing, and genetic manipulation approaches are becoming increasingly important in developmental and regenerative biology. Such approaches in zebrafish research are hampered by the lack of an ubiquitous transgene driver element that is active at all developmental stages. Here, we report the isolation and characterization of the medaka fish (Oryzias latipes) β-actin (Olactb) promoter, which drives constitutive transgene expression during all developmental stages, and the analysis of adult organs except blood cell types. Taking advantage of the compact medaka promoter, we succeeded in generating a zebrafish transgenic (Tg) line with unprecedentedly strong and widespread transgene expression from embryonic to adult stages. Moreover, the Tg carries a pair of loxP sites, which enables the reporter fluorophore to switch from DsRed2 to enhanced green fluorescent protein (EGFP). We induced Cre/loxP recombination with Tg(hsp70l: mCherry-t2a-CreERt2) in the double Tg embryo and generated a Tg line that constitutively expresses EGFP. We further demonstrate the powerful application of Olactb-driven Tgs for cell lineage tracing using transplantation experiments with embryonic cells at the shield stage and adult cells of regenerating fin. Thus, the use of promoter elements from medaka is an alternative approach to generate Tgs with stronger and even novel expression patterns in zebrafish. The Olactb promoter and the Tg lines presented here represent an important advancement for the broader use of Cre/loxP-based Tg applications in zebrafish.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Conditional gene manipulation using Cre/loxP recombination is one of the important tools in molecular genetics. In particular, a transgene driver element that ubiquitously and strongly expresses the transgene in many cell types and stages is necessary for cell lineage tracing analysis or conditional gain-of-function and/or loss-of-function experiments. In mouse, conditional approaches using Cre/loxP recombination have been greatly accelerated by the discovery of the Rosa26 locus (Friedrich & Soriano 1991; Zambrowicz et al. 1997; Soriano 1999). Zebrafish (Danio rerio) has become an alternative and useful vertebrate model system that has the advantages of quick and inexpensive genetic manipulations and beautiful imaging analyses; however, the lack of an ubiquitous transgene driver applicable at both developmental and post-developmental stages has been a drawback.

So far, a number of ubiquitous promoter elements have been tested in zebrafish, and several promoters such as the h2afx, the TATA box-binding protein (tbp), versions of the actb promoters (Higashijima et al. 1997; Gillette-Ferguson et al. 2003; Kwan et al. 2007; Burket et al. 2008), and the Xenopus laevis elongation factor 1a promoter (Xlef1a1) (Johnson & Krieg 1994; Kawakami et al. 2004) have been used to generate transgenic (Tg) lines. However, these promoters are progressively inactivated in some neuronal, blood, and/or other cell types during development. For example, although Xlef1a1 is strongly expressed during early development, the expression is restricted to specific cell types (Hans et al. 2009; Collins et al. 2010). actb promoter fragments retain expression in adult tissues, yet show no or weak activity in blood cell types, fins, or several other cell types (Traver et al. 2003; Burket et al. 2008). This likely reflects specialization of differentiating cells, differences in cell-type-specific requirements for molecules controlling translation or chromatin maintenance, and/or the inactivation of foreign DNA sequences in the genome. Recently, two relatively successful examples of ubiquitous transgene expression in zebrafish have been reported. In one case, the long 9.8-kb promoter of actb2 was used to generate a Tg line that expressed the transgene in the adult heart and fin (Bertrand et al. 2010; Kikuchi et al. 2010; Liu et al. 2010; Singh et al. 2012). In another case, a 3.5-kb element of the ubiquitin promoter was successfully used to drive ubiquitous transgene expression not only in embryonic tissues, but also in adult tissues, including blood cell types (Mosimann et al. 2011).

Here, we report on another ubiquitous promoter element that possibly drives the strongest reporter transgene expression in zebrafish. We demonstrate that a short 2.5-kb promoter element of the medaka actb gene drives strong transgene expression in medaka fish, and surprisingly this short element also drives ubiquitous and strong transgene expression in embryonic and adult zebrafish tissues. Given that the construct was designed to switch the transgene from dsRed2 (dsR2) to enhanced green fluorescent protein (EGFP) using Cre/loxP recombination, the Tg is useful for genetically labeling specific cell groups and tracking their long-term fates. We actually show that the strong and persistent expression of the transgene is useful for tracing cell lineage at single-cell resolution. Thus, promoter elements derived from foreign species are alternative and efficient tools to generate Tg animals in zebrafish with higher gene expression or novel expression patterns that unify duplicated gene expressions.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Plasmid constructs

p067Olactb:loxP-dsR2-loxP-EGFP

The 2.5-kb promoter region, the first untranslated (UTR) exon, the 1.5-kb first intron, and the sequence of the second intron before the initiation codon of the medaka fish actb (Olactb) gene was polymerase chain reaction (PCR)-amplified from the Oryzias latipes genome using KOD polymerase and cloned between the EcoRI and SalI sites of pDsRed2-1 (Clontech). Then, the DsRed2 gene between the SalI and NotI sites was replaced with the EGFP gene that was amplified by PCR from pEGFP-1 (Clontech). In the resultant plasmid, the loxP-DsRed2-polyA signal (pA)-loxP cassette, which was PCR-amplified from the modified pDsRed2-1 using the primers loxP-DsR2-FW1 (5′-CAGTGTCGACATAACTTCGTATAGCATACATTATACGAAGTTATACTAGTGATATCCATGGCCTCCTCCGAGAACGT-3′) and loxP-GFP-RV2 (5′-GACTCCGCGGATAACTTCGTATAATGTATGCTATACGAAGTTATCGATATCATATGTTAATTAACGCTTACAATTTACGCCT-3′), was inserted between the SalI and SacII sites. The NotI site of the template pDsRed2-1 was disrupted by blunt-end ligation.

pT2Olactb:loxP-dsR2-loxP-EGFP

The Xlef1a1 promoter region and the EGFP coding region were removed from pT2KXIGΔin (Kawakami 2005) by partial digestion with EcoRI and NotI and replaced with Olactb:loxP-dsR2-loxP-EGFP from p067Olactb:loxP-dsR2-loxP-EGFP. A 370-bp sequence of Xlef1a1 remained upstream of the Olactb promoter, but it did not affect actb promoter activity, because we did not observe any difference in expression when the construct lacking the Xlelf1a promoter region was injected (unpubl. obs., 2011).

pT2hsp70l:mCherry-t2a-CreERt2

A similar construct with hsp70l promoter and its Tg have been previously reported (Hans et al. 2009, 2011); however, we were unable to detect mCherry expression in adult fish after heat shock. To improve transgene expression, particularly in adult tissues, we used a modified construct and generated a Tg line. The hsp70l promoter was PCR-amplified from a vector in the tol2 kit (Kwan et al. 2007) and replaced with the promoter region of pT2pax2:mCherry-t2a-CreERt2 (Hans et al. 2009 at the SfiI and FseI sites. The resultant plasmid contains the 5′UTR sequence derived from the pax2 construct. The PCR reaction was carried out using KOD plus polymerase, according to the manufacturer's instructions (Toyobo).

Transgenesis

Fish were maintained in accordance with Animal Research Guidelines at Kyoto University and Tokyo Institute of Technology. Fish were maintained in an aquarium with recirculating water a 14 h/day and 10 h/night cycle at 28.5°C.

The medaka Tg was generated according to the method described by Kinoshita et al. (2000). Briefly, a circular form of plasmid (25 μg/mL) was injected into the cytoplasm of fertilized eggs of the d-rR medaka strain before the first cleavage.

Zebrafish Tgs were generated using tol2-mediated transgenesis (Kikuta & Kawakami 2009). Briefly, each transgenesis vector (25 μg/mL) was combined with 25 μg/mL tol2 mRNA and injected into one-cell-stage eggs derived from TL wild-type crosses, and the injected animals were grown to adulthood. Individual F0 founders were either outcrossed to TL or incrossed to other F0 founders, and their F1 progeny were screened for DsR2, EGFP, or mCherry fluorescence. Tg(hsp70l:mCherry-t2a-CreERt2) was identified by heat shock at 37°C for 1 h at 24 h post-fertilization (hpf) and screened for mCherry fluorescence at 48 hpf. Positive individual F1 adults were subsequently outcrossed to wild-type zebrafish until 50% transgene transmission, indicating single transgene insertions, was observed. Tgs were examined in each generation for uniform and strong transgene expression.

Histology

Transgene expression in adult tissues was assessed in transverse sections. A young adult Tg fish, approximately 2 months old, was euthanized with ice and immediately fixed with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) overnight at 4°C. The fixed fish was equilibrated with 20% sucrose in PBS, embedded in Tissue-Tek compound (Miles), and sectioned at 20 μm using a cryostat. Cut sections were picked up on glass slides and washed two to three times with PBS followed by PBS supplemented with 0.1% Tween 20 (PBST). Sections were stained with polyclonal anti-DsRed antibody (Invitrogen) and Alexa568-labelled anti-rabbit antibody and further counterstained with the SYTO Green 11 Nucleic Acid Stain (5 mmol/L in PBST; Invitrogen) for 10 min at room temperature and mounted with 80% glycerol containing 2.5% 1,4-diazabicyclo [2,2,2] octan (DABCO, Nacalai Tesque) as an anti-fading reagent. Pictures were taken using a confocal microscope.

Blood cell analysis

Adult fish blood cells were collected from the tail fin vessels by amputating the fin on a piece of parafilm. Blood cells were mixed with a drop of PBS and mounted on a slide glass. A cover slip was placed on the sample using pieces of vinyl tape as a spacer.

Cre/loxP recombination

Respective female Tg(hsp70l:mCherry-t2a-CreERt2) and male Tg(Olactb:loxP-dsR2-loxP-EGFP) were mated to obtain embryos that carried both transgenes. Fertilized eggs with DsRed fluorescence at 24 hpf were selected and heat-shocked at 37°C for 1 h. The embryos were returned to 28.5°C and incubated overnight in medium containing 5 μmol/L tamoxifen (TAM; Sigma-Aldrich). Approximately 50% of embryos became EGFP-positive. These embryos were raised to adulthood and mated with wild-type fish to obtain embryos that constitutively expressed EGFP.

Transplantation experiments

Cell transplantation at early embryonic stage was performed according to standard procedures. Briefly, female Tg(Olactb:loxP-dsR2-loxP-EGFP) were crossed with male wild-type fish. All fertilized eggs had maternal contribution of DsR2 and were used as donors. The donors and the wild-type host embryos were dechorionated at 30% epiboly stage and further incubated in embryo medium supplemented with 100 units of penicillin and 0.01% streptomycin until they attained the appropriate stage for transplantation. Donor and host embryos were placed in small wells on a 2% agarose platform that accommodated one embryo per well. Transplantation was performed between 40% epiboly and shield stages using glass needles with tip diameters of 30–50 μm.

For adult blastema transplantation, the fins of adult Tg(Olactb:loxP-dsR2-loxP-EGFP) and wild-type fish were amputated in the middle and fin regeneration was induced. At 2 days post-amputation (dpa), the regenerating tissue of the donor was collected by re-cutting the fin at a slightly proximal site. The blastema and wound epidermis were manually separated with 30-guage needles in Hank's buffer and further cut into small pieces. The donor tissue was transplanted into the blastema region of the host using a transplantation needle with a wide tip. For transplanting the wound epidermis, the donor epidermis was cut into small pieces, approximately 1/10 of the blastema size, and inserted into the blastema or the space between the epidermis and blastema. The donor blastema was also cut into small pieces and inserted into the host blastema region.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Medaka β-actin promoter transgenic

actb is one of the genes that exhibits strong and ubiquitous expression in many cell types and stages. Taking advantage of its ubiquitous and strong expression, we used the 2.5-kb promoter region along with the first UTR exon, 1.5-kb first intron, and a short sequence before the initiation codon in the second exon (Fig. 1A) to generate the medaka Tg. The Tg displayed strong dsR2 expression such that the fish had a visible red body color under daylight (Fig. 1B). Furthermore, transgene expression was observed in a variety of adult tissues, such as the brain, spleen, kidney, gill, gut, ovary, and testis (data not shown). Compared to the long (9.8 kb) promoter element of the zebrafish actb2 required for stable adult tissue expression, a 2.5-kb region of the medaka β-actin was sufficient to drive an even stronger expression of the transgene in embryonic and adult tissues.

image

Figure 1. The medaka (Oryzias latipes) and zebrafish (Danio rerio) transgenic (Tg) lines with widespread expression of dsRed2 (dsR2) driven by the medaka fish actb promoter (Olactb). (A) Schematic of the plasmid DNA construct used to generate the Tgs. The construct contains a 2.5-kb upstream region, 1.1-kb first intron, and a 7-bp sequence of the second exon before the actb initiation codon. The dsR2 cassette is flanked by 2 loxP sites and followed by the enhanced green fluorescent protein (EGFP) cassette. (B) Bright field (upper) and fluorescent (lower) images of medaka Tg(Olactb:loxP-dsR2-loxP-EGFP) with strong and widespread expression of DsR2. The strong DsR2 expression is clearly seen under daylight. (C) Bright field (upper) and fluorescent (lower) images of zebrafish Tg(Olactb:loxP-dsR2-loxP-EGFP) carrying a single copy of the transgene. The stable Tg line displays red-orange color due to strong DsR2 expression. Lines and numbers indicate the approximate locations of the cross-sections shown in Figure 2A. (D) Zebrafish Tg carrying homozygous transgene copies.

Download figure to PowerPoint

Generation of a zebrafish transgenic using the medaka β-actin promoter

Zebrafish has 2 actb genes, actb1 and actb2. The promoters of both genes have been used to generate Tg lines (Gillette-Ferguson et al. 2003; Burket et al. 2008; Bertrand et al. 2010; Liu et al. 2010). In our hands, zebrafish actb1 or actb2 promoters <5 kb in length did not drive stable and ubiquitous transgene expression and displayed a mosaic and/or weak transgene expression in adult tissues (data not shown).

To generate a zebrafish Tg line with strong and stable transgene expression for use as a reliable genetic marker for cell lineage tracing, we used the medaka actb promoter element. To facilitate transgene integration efficiency in zebrafish, we inserted the Olactb:loxP-dsR2-loxP-EGFP cassette into the tol2 transposon vector (Kawakami 2005, 2007; Kikuta & Kawakami 2009) and injected the construct into fertilized zebrafish eggs. The injected embryos exhibited brilliant DsR2 fluorescence from the somite stage until the adult stage in a variety of tissues. More than 50% of founder fish produced F1 offspring that expressed the transgene in the whole body at 1 dpf. Some adult F1 fish only retained weak transgene expression possibly due to gene silencing (data not shown). But, others maintained the strong and widespread expression. We selected and raised two independent F1 carriers that exhibited strong DsRed expression in adult tissues and outcrossed them with wild-type fish. By repeated selection of Tgs with stable and uniform transgene expression and outcross with wild-type fish, we established a Tg line, which stably expressed the transgene more than five generations (Fig. 1C). Furthermore, we also generated other transgenics using the Olactb promoter and observed a similar level of stability of transgene expression. These observations indicate that the observed promoter activity is reproducible. The line transmits the transgene to about 50% of offspring when outcrossed with the wild-type strain, indicating that the transgene is inserted into a single locus of the genome. Furthermore, as we used the transposon-mediated transgenesis, the number of transgene insertions per locus is thought to be single. We actually confirmed it using Cre/loxP recombination (Fig. 3C). Despite the single copy of the transgene, the zebrafish Tg had an apparent red body color under daylight (Fig. 1C). Furthermore, adult fish carrying homozygous copies of the transgene displayed a brighter red color (Fig. 1D).

Transgene expression in various stages and tissues

To further characterize Tg(Olactb:loxP-dsR2-loxP-EGFP), we examined transgene expression in a variety of tissues and stages. We first examined transgene expression in adult cross-sections. As shown in Figure 2A, DsR2 expression was observed in most tissues and cell types. Although tissues such as the nervous system and somites appeared to exhibit stronger transgene expression, the overall expression of the transgene was, by and large, uniform and ubiquitous.

image

Figure 2. Widespread transgene expression driven by the Olactb promoter in various tissues and stages of zebrafish. (A) Cross-sections of a heterozygous Tg(Olactb:loxP-dsR2-loxP-EGFP) zebrafish stained with anti-DsRed antibody, SYTO Green 11 (SG11) and their merged images at levels shown in Figure 1C. Ubiquitous expression of DsR2 is seen in all identifiable tissues and internal organs, and particularly stronger fluorescence is observed in the central nervous tissues, germ cells and muscles. M, mandible; br, brain; he, heart; hi, hind brain; li, liver; in, intestine; ov, ovary; ki, kidney; sm, skeletal muscle; spc, spinal cord; tf, tail fin. (B) Absence of transgene expression in blood cell lineage. Blood cells were collected from amputated tail fin of Tg(Olactb:loxP-dsR2-loxP-EGFP). Although a few DsR2-positive dots were seen (arrowhead), they did not appear to be live cells given their small size and the absence of a nucleus. (C) Transgene expression in endothelial cells of regenerating blood vessels. In some cases of adult regenerating cell transplantation from Tg(Olactb:loxP-dsR2-loxP-EGFP) to wild-type fish, the transplanted cells are incorporated into blood vessels. When regeneration is re-induced by fin amputation, the transplanted cells proliferate to become a part of the regenerated blood vessels, confirming that the transgene is expressed in the endothelial cells. Note that the endothelial cells only produce endothelial cells, and not other cell types, during fin regeneration. dpa, days post-amputation. (D) Maternal DsR2 contribution in 1–2-celled embryos derived from outcrossing a stable female Tg(Olactb:loxP-dsR2-loxP-EGFP) with wild type. (E and F) Comparison of maternal and zygotic contributions of transgene expression at 24 h post-fertilization (hpf) (E) and 48 hpf (F). Zygotic transgene expression begins around the somite stage and gradually increases in intensity, although the maternal DsR2 fluorescence is stronger than the zygotic one even at 48 hpf. The zygotic transgene expression is uniform and ubiquitous throughout (also see Fig. 3C). Scale bars, 400 μm in A (1–9); 200 μm in A (10); 50 μm in B and C 1 mm in D–F.

Download figure to PowerPoint

Blood cell lineages of zebrafish actb Tgs have been reported to lack transgene expression (Traver et al. 2003; Burket et al. 2008); therefore, we collected blood cells from adult fin blood vessels and carefully examined transgene expression using fluorescent and confocal microscopes. However, we did not detect any blood cells expressing DsR2. Although a few DsR2-positive particles were observed (Fig. 2B, arrowhead), they were smaller than red blood cells and may not be live cells or even cells. Thus, consistent with the findings in other actb Tgs generated in zebrafish, Tg(Olactb:loxP-dsR2-loxP-EGFP) also lacked transgene expression in hematopoietic cell types.

In contrast to the lack of transgene expression in blood cells, strong transgene expression was observed in blood vessels (Fig. 2C). As shown in Figure 4D, we established a procedure for cell transplantation between regenerating fins. In some cases, the transplanted Tg cells were incorporated into blood vessels and exhibited an elongated shape (n = 5). To further confirm the identity of these cells, regeneration was re-induced by amputating a slightly distal site of the implanted cells. Upon regeneration, the implanted cells proliferated to form the regenerated blood vessels (Fig. 2C), indicating that the medaka actb promoter indeed drives transgene expression in endothelial cells. More intriguingly, the transplanted endothelial cells only contributed to the regenerated blood vessels, and never differentiated into other cell types, indicating the fate restriction of endothelial cells during fin regeneration.

As for transgene expression during development, the embryos produced by the female Tg displayed strong DsRed expression from the 1-cell stage (Fig. 2D), indicating maternal contribution of the transgene. This maternal DsR2 fluorescence was sustained for more than 48 h (Fig. 2E,F). On the other hand, zygotic transgene expression was observed from the somite stage (Fig. 2E). Although the fluorescence intensity of the zygotic expression was weaker than that of the maternal expression (Fig. 2E,F), it was stronger than that of many other Tg lines, such as Tg(Xlef1a:EGFP) at 24 hpf, and further increased with time (Fig. 2F). Most importantly, the zygotic as well as maternal expression of the transgene was uniform and ubiquitous in most tissues, including neuronal tissues and fin fold, irrespective of the developmental stage. Collectively, these data indicate that the medaka actb promoter drives transgene expression in every life stage, from one cell to adulthood, and in most cell types, excluding blood cell types.

Color switch and cell labeling by Cre/loxP recombination

One of the important applications of an ubiquitous promoter element is the conditional and permanent labeling of specific cells and/or their genetic manipulation. In zebrafish, it has been shown that Cre/loxP system-based lineage tracing tools are effective with single-insertion loxP cassettes generated via tol2-mediated transgenesis (Kawakami 2007; Yoshikawa et al. 2008). Here, we used the Tg(hsp70l:mCherry-t2a-CreERt2) to evaluate the recombination efficiency. It expresses a bicistronic mRNA encoding mCherry and CreERt2 separated by a viral self-cleavable T2A peptide sequence (Provost et al. 2007). The CreERt2 recombinase fusion protein is induced to translocate into the nucleus by exposure to TAM or its active metabolite 4-OHT (Feil et al. 1996, 1997) and is thus ideal for spatiotemporal tracing of cell types during zebrafish development (Hans et al. 2009). We designed the construct such that the reporter fluorescent protein switched from DsR2 to EGFP by Cre/loxP recombination (Fig. 1A). Once Cre recombinase excises the DsR2 cassette, the cells and their descendants are permanently labeled with EGFP expression.

We crossed the heterozygous male Tg(Olactb:loxP-dsR2-loxP-EGFP) with the heterozygous female Tg(hsp70l:mCherry-t2a-CreERt2) and obtained double Tg embryos. Among the produced embryos, approximately 50% of embryos were DsR2-positive and half of them were expected to carry the CreERt2; therefore, we selected the DsR2-positive embryos at 24 hpf and heat-shocked them at 37°C to induce CreERt2 expression, and further incubated them with 5 μmol/L TAM until 4 dpf. Though we could not confirm the mCherry fluorescence due to the strong DsR2 expression, approximately 50% of these embryos, which were thought to carry both transgenes, indeed displayed a highly mosaic EGFP expression, indicating the occurrence of Cre/loxP recombination. We also confirmed the presence of the Cre transgene by PCR genotyping. Intriguingly, these embryos retained a significant level of DsR2 fluorescence (Fig. 3A), suggesting that the Cre/loxP recombination occurred in a mosaic pattern. This incomplete penetrance of Cre/loxP recombination is possibly due to the unequal hsp70l promoter activity and CreERt2 transgene expression. We indeed observed an uneven mCherry expression in the Tg(hsp70l:mCherry-t2a-CreERt2) even after a longer heat induction. In addition, it could also be possible that proteins connected by the T2a sometimes expressed at a lower level (unpubl. obs., 2011).

image

Figure 3. Switch in transgene expression from DsR2 to enhanced green fluorescent protein (EGFP) by Cre/loxP recombination. (A) Three days post-fertilization (dpf) larvae of Tg(Olactb:loxP-dsR2-loxP-EGFP;hsp70l:mCherry-t2a-CreERt2) (upper) and Tg(Olactb:loxP-dsR2-loxP-EGFP) (lower) after induction of Cre/loxP recombination at 24 hpf. Larvae carrying both the Cre transgenes become EGFP-positive, whereas those lacking the Cre transgene do not express EGFP. (B) Adult fish with recombination induced at 24 hpf. DsR2 and EGFP are expressed in a mosaic pattern. (C) Embryos (24 hpf) obtained from the outcross of the DsR2/EGFP mosaic fish with wild-type TL zebrafish. Both panels show the same field. Arrowheads indicate the embryos with EGFP expression. Half of the produced embryos exclusively express DsR2 (left panel) or EGFP (right panel), suggesting single copy insertion into the genomic locus. Scale bars in A and C, 0.5 mm.

Download figure to PowerPoint

image

Figure 4. Cell tracing after transplantation of transgenic cells into wild-type host. (A–C) Transplanted cells during development. Transplantation was performed at 40% epiboly stage. Donor embryos were obtained by mating female Tg(Olactb:loxP-dsR2-loxP-EGFP) with male wild-type TL. Due to strong fluorescence of maternal DsR2, individual transplanted cells and their progenies are clearly recognizable at single-cell resolution at 40% epiboly (A), 24 hpf (B), and 4 dpf (C). Even after 4 days post-fertilization (dpf), individual progenies could be traced by the increasing zygotic transgene expression (data not shown). (D) Tracing of transplanted cells in the adult tissue. Regenerating fin blastema cells transplanted into wild-type blastema integrate into the host and proliferate to contribute to the regenerated tissue. The donor cells and their progeny are traceable during the course of regeneration. The progeny of transplanted blastema cells contribute to both intra-ray and inter-ray tissues. Arrow indicates the site of amputation. dpi, days post-implantation. Scale bars, 0.5 mm in A–C; 50 μm in D.

Download figure to PowerPoint

We raised these mosaic embryos to adulthood, and the adult fish continued to display the mosaic DsR2 and EGFP expression all over the body (Fig. 3B). The male mosaic fish were further crossed with wild-type females to obtain progeny. These embryos showed either DsR2 or EGFP at 24 hpf, indicating that the germ line cells of the mosaic fish were also a mosaic of recombined and non-recombined cells. More importantly, none of the embryos exhibited both DsR2 and EGFP fluorescence (Fig. 3C). This strongly supports a single copy of transgene integration per genome locus occurring via tol2-mediated transgenesis. Furthermore, we did not detect any EGFP-positive cells in the DsRed-expressing embryos, although half of them carried the Cre transgene. This indicates that no detectable leaky recombination occurred in our Tg(hsp70l:mCherry-t2a-CreERt2) without TAM and heat shock, and is consistent with the previous observation by Hans et al. (2011).

Cell lineage tracing by cell transplantation

For use as a lineage tracing marker, the intensity of transgene expression needs to be strong enough to easily track single cells. We evaluated the utility of Tg(Olactb:loxP-dsR2-loxP-EGFP) as a lineage tracing tool using cell transplantation experiments. We first performed the cell transplantation in early embryonic stage. The embryos produced by the female Tg(Olactb:loxP-dsR2-loxP-EGFP), which expressed maternal DsR2, were used as donors, and the cells were transplanted into wild-type embryos at 40% epiboly stage. Due to the strong and long-lasting fluorescence of maternal DsR2, the transplanted DsR2-positive cells were clearly recognized at single-cell resolution immediately after the operation (Fig. 4A), at 24 hpf (Fig. 4B), and 4 dpf (Fig. 4C), demonstrating that the established Tg is a useful tool for tracking cell lineage during embryonic development.

We further examined the versatility of Tg(Olactb:loxP-dsR2-loxP-EGFP) in tracking cell lineages in adult tissues. To this end, we transplanted cells of the regenerating fin, which is one of the easily accessible tissues for cell transplantation in adult fish. We established a procedure for cell transplantation of regenerating cells such as the blastema and wound epidermis. Two days after amputation of the adult fin, the regenerating tissue was isolated and separated into epidermal and mesenchymal parts and transplanted into the host regenerating tissue. Although the transplanted wound epidermis was stably integrated into the host tissue at 1 h after transplantation, the donor epidermal cells almost disappeared on the next day (n = 8) with a few remaining donor cells in the epidermal region in some cases (data not shown), suggesting that the transplanted wound epidermis cells may be rejected or dead. On the other hand, the transplanted blastema cells easily integrated into the host tissue in most cases, began to proliferate, and contributed to a wide region of the regenerated tissue including the mesenchymal cells inside and outside of the fin ray bones and seemingly the osteocytes outlining the fin ray bone (n > 20; Fig. 4D). During the course of regeneration, even a single cell was clearly traceable, demonstrating that the cells derived from Tg(Olactb:loxP-dsR2-loxP-EGFP) were also traceable at single-cell resolution in adult tissues. Thus, the zebrafish Olactb-driven Tg can be seamlessly used as a useful cell lineage tracer throughout the life cycle.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Conditional gene manipulation using Cre/loxP recombination is one of the important tools for molecular genetic analysis in model organisms. The number of potent zebrafish tools to manipulate genetic events and trace cell lineages through development has steadily increased in recent years; however, the lack of an ubiquitous transgene driver has been a common obstacle (Hans et al. 2009; Blackburn & Langenau 2010; Collins et al. 2010). Although the actb gene promoter has been one of the strong candidates, it has been difficult to generate a Tg line that uniformly and stably expresses the transgene (Gillette-Ferguson et al. 2003; Burket et al. 2008; Bertrand et al. 2010; Liu et al. 2010). Here, we found that the short 2.5-kb medaka actb promoter strongly drives transgene expression in every life stage from one cell to adulthood and in most cell types in the medaka Tg, and further used it to generate a zebrafish Tg that ubiquitously expresses the transgene in all stages and tissues, except the blood cell lineage. Intriguingly, the transgene was expressed in endothelial cells (Fig. 2C), despite their closely related cell lineage with hematopoietic cell types. Although we cannot exclude the potential existence of other cell types that lack transgene expression, our Tg line using the medaka actb promoter is a breakthrough in the limitation of Cre/loxP technology in zebrafish genetics.

Because the size of the medaka genome is approximately half of that of zebrafish (Kasahara et al. 2007), it is reasonable to assume that the promoter/enhancer elements are located within a compact region. Indeed, compared to the long zebrafish actb2 promoter (9.8 kb) required to sustain adult tissue expression, 2.5 kb of the Olactb promoter was sufficient to drive a stronger expression of the transgene in embryonic and adult tissues. Recently, almost the same promoter region was successfully used to generate a uniform Tg in a related medaka species, Oryzias dancena (Cho et al. 2011). In addition, medaka has a single actb gene, whereas zebrafish has two actb genes, which may have redundant functions and expression. Therefore, the zebrafish actb genes may have a weaker transcriptional activity than that of medaka fish.

Still, the zebrafish Tg(Olactb:loxP-dsR2-loxP-EGFP) displayed a weaker red color in comparison to the bright red body color of the medaka Tg (Fig. 1). This may be due to the multiple copy number of the transgene integrated into a locus in the medaka Tg rather than the promoter activity itself, because the medaka Tg was generated by a simple plasmid injection, which leads to the integration of multiple copies of the transgene into a locus. By contrast, the zebrafish Tg was generated via tol2-based transgenesis, which introduces a single transgene insertion per locus. Thus, the overall intensity of Olactb-driven transgene expression appears to depend on the number of transgene copies. The observation that adult zebrafish Tgs carrying homozygous transgenes displayed a brighter red color supports this notion (Fig. 1D). Given the single copy of the transgene in the zebrafish Tg, its orange-red body color is surprising; therefore, we speculate that the transcription efficiency per transgene copy may be comparable to that of medaka. Thus, our established Tg displayed an unprecedentedly strong and widespread transgene expression in zebrafish. Moreover, the insertion of a single transgene copy in zebrafish is an advantage when using Cre/loxP recombination, because it enables complete color switch from red to green in every cell.

To further demonstrate the actual application of the zebrafish Olactb-driven Tg as a cell tracing reagent, we performed a series of transplantation experiments in early embryonic stage and adult regenerating fin. We successfully demonstrated that the established Tg is a useful tool for tracking cell lineage throughout the life cycle. Taking advantage of the established Tg, we were able to observe an intriguing cell behavior of transplanted blastema cells during adult fin regeneration. As shown in Figure 4D, the donor blastema cells proliferated to distribute into a wide region of the regenerated tissue (n > 20). However, in some cases, the donor cells stably incorporated into the host tissue, but did not proliferate (data not shown). Between the proliferated and un-proliferated cases, it seemed likely that the host blastema region that received the transplantation was decisive for activating/maintaining cell proliferation. Irrespective of the place of origin of donor blastema cells, transplantation in the distal region of the host blastema induced explosive cell proliferation, suggesting the presence of a localized signal for regenerative cell proliferation. Another intriguing outcome of our adult cell transplantation was that the transplanted endothelial cells only contributed to the regeneration of blood vessels, and did not differentiate into other cell types (Fig. 2C), indicating the fate restriction of endothelial cells during fin regeneration.

In summary, the widespread and, particularly, strong expression of the transgene driven by the medaka actb promoter can be seamlessly used as a useful cell lineage tracer throughout the life cycle. In addition, due to the small size of the Olactb promoter element, it is easily applicable in other constructs for conditional loss-of-function or gain-of-function analyses. Thus, foreign promoter elements derived from the compact medaka genome are alternative and efficient tools to generate Tg animals in zebrafish with higher gene expression or new expression patterns that unify multiple gene expressions. The medaka actb promoter now provides an alternative choice for constructing elaborate recombinase-dependent or drug-inducible transgene systems for all developmental stages in zebrafish.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This work was supported by research grants from the Ministry of Education, Sports, Science and Technology of Japan, KAKENHI (Grant-in-Aid for Scientific Research). We thank K. M. Kwan and C. B. Chien for providing the Tol2 kit, and S. Hans and M. Brand for generously providing the CreERt2constructs and Tg(hsp70l:mCherry-t2a-CreERt2).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References