Introduction
After 1952, when Briggs and King obtained normal hatched tadpoles by blastomere nuclear transfer, nuclear transplant technology began to be developed for reprogramming studies (Kikyo et al., Reference Kikyo, Wade, Guschin, Ge and Wolffe2000; Wade & Kikyo, Reference Wade and Kikyo2002; Li, Reference Li2002; Giraldo et al., Reference Giraldo, Hylan, Ballard, Purpera, Vaught, Lynn, Godke and Bondioli2008). Mammalian cloning by nuclear transfer has been successfully achieved in several species (Wilmut et al., Reference Wilmut, Schnieke, McWhir, Kind and Campbell1997; Wakayama et al., Reference Wakayama, Perry, Zuccotti, Johnson and Yanagimachi1998; Byrne et al., Reference Byrne, Pedersen, Clepper, Nelson, Sanger, Gokhale, Wolf and Mitalipov2007; French et al., Reference French, Adams, Anderson, Kitchen, Hughes and Wood2008) with varied somatic cell types as donors (Campbell et al., Reference Campbell, McWhir, Ritchie and Wilmut1996; Wilmut et al., Reference Wilmut, Schnieke, McWhir, Kind and Campbell1997; Wakayama et al., Reference Wakayama, Perry, Zuccotti, Johnson and Yanagimachi1998; Shiga et al., Reference Shiga, Fujita, Hirose, Sasae and Nagai1999).
Although fish cloning is less developed, several recent works using medaka (Oryzias latipes) have been reported, in which both blastomeres (Bubenshchikova et al., Reference Bubenshchikova, Ju, Pristyazhnyuk, Niwa, Kaftanovskaya, Kinoshita, Ozato and Wakamatsu2005) and somatic larval and adult cells (Bubenshchikova et al., Reference Bubenshchikova, Ju, Pristyazhnyuk, Niwa, Kaftanovskaya, Kinoshita, Ozato and Wakamatsu2005, Reference Bubenshchikova, Kaftanovskaya, Motosugi, Fujimoto, Arai, Kinoshita, Hashimoto, Ozato and Wakamatsu2007; Kaftanovskaya et al., Reference Kaftanovskaya, Motosugi, Kinoshita, Ozato and Wakamatsu2007) were used as donors, and non-enucleated and activated eggs were used as recipients in all cases. In this species, embryonic nuclear transplants using functionally enucleated and non-activated eggs have been achieved only with blastomeres as nuclei donors (Wakamatsu et al., Reference Wakamatsu, Ju, Pristyaznhyuk, Niwa, Ladygina, Kinoshita, Araki and Ozato2001).
In contrast, in zebrafish (Danio rerio), the first successful embryonic (10–15 somites) somatic cloning by nuclear transplant with mechanically enucleated and previously activated eggs was described by Huang and colleagues in 2003. Since this work and to date, to our knowledge, no additional improvement in nuclear transplant techniques in zebrafish has been published by these or any other authors.
To date, due to technical (Nüsslein-Volhard & Dahm, 2002) or biological (Westerfield, Reference Westerfield2003) limitations, fish somatic nuclear transplant in these two laboratory species with somatic embryonic (10–15 somites; Huang et al., Reference Huang, Ju, Lee and Lin2003) or adult cells (Bubenshchikova et al., Reference Bubenshchikova, Kaftanovskaya, Motosugi, Fujimoto, Arai, Kinoshita, Hashimoto, Ozato and Wakamatsu2007) has only been attempted using preactivated eggs as recipients. These limitations have hindered the study of the reprogramming effects of cytoplasmic factors characteristic of the metaphase II status in the oocyte, the effects of the synchrony degree between activation and nuclear transplant, egg aging or the previous donor cell reprogramming treatments. This situation is not the case in mammals, in which these reprogramming factors have been studied because somatic adult nuclear transplant has been more easily carried out before, at the same time and after oocyte activation (Cambell et al., Reference Campbell, McWhir, Ritchie and Wilmut1996; Wilmut et al., Reference Wilmut, Schnieke, McWhir, Kind and Campbell1997).
As the zebrafish model is a powerful genetic and developmental system in which the genome has already been sequenced, the aim of this work was to develop three methods to enable nuclear transplant to be carried out using adult cells prior, simultaneously with or following egg activation/fertilization in zebrafish to be used in reprogramming studies.
Material and methods
Care and maintenance of zebrafish colony
Two zebrafish (Danio rerio) colonies (wild and gold strains) were established in our laboratory from specimens purchased in a specialized establishment and kept in closed reproduction for five years. Adult zebrafish were kept in 20 litre tanks in a 2:1 ratio (females/males) and fed on granular food supplemented with recently defrosted hen egg yolks and shrimp meat (Simao et al., Reference Simao, Perez-Camps and Garcia-Ximénez2007). The light cycle was regulated at 14 h light/10 h dark.
Non-activated eggs and sperm collection
Eggs were collected after evaluation of the sexual behavior of both gold strain males and females at dawn. Only females that manifested this sexual behaviour were anesthetized in an oil clove solution (100 μl in 1 l of dechlorinated and decalcified water: system water) for a few minutes and the eggs were obtained by gentle extrusion of the ovary. It is important to prevent eggs coming into contact with fresh water, because they activate immediately. Only good eggs (yellow and translucent colours) were kept in Hanks’ buffered salt solution (H10) supplemented with 1.5% (v/v) of bovine serum albumin (BSA) and 0.1 g of NaCl/100 ml of Hanks’ medium (egg medium; pH: 7.4; osmolarity: 310–320 mOsm) at 8 °C until their use (1 h and 30 min as maximum time).
The gold zebrafish males that showed reproductive behaviour were also anesthetized as described before. The abdominal region was gently pressed while the sperm was being recovered from the genital pore into individual glass microcapillaries (1 × 90 mm Narishige Scientific Instrument Laboratory). A pool from 2–3 different males (0.5–2 μl/male) was diluted in 200 μl egg medium, which can also keep the sperm in a non-activated status, and then the dilution was stored at 8 °C until use.
In vitro fertilization
In zebrafish, the eggs quickly lose their post-ovulatory ability to be fertilized (90 min). Moreover, the time between complete egg activation and in vivo fertilization is extremely short (seconds) in zebrafish (Nüsslein-Volhard & Dahm, 2002). Non-activated eggs and sperm were mixed in egg medium and stored at 8 °C until fertilization, for all nuclear transplant methods (see Experimental design). To activate both gametes, 1 ml of system water at room temperature was added to the egg–sperm mixture. After 2–3 min, the time required for fertilization in zebrafish, the 35 mm Petri dish (Corning) was fully filled with the water system for achieving well developed embryos. Further culture was done at 28 °C.
Primary culture and somatic cell collection
Somatic cells used as nuclei donor came from wild zebrafish caudal fin primary cultures. The tissue was obtained by caudal fin amputation of adult specimens after they had been anesthetized in clove oil solution. The tissue was cleaned with a 0.2% bleach solution for 2 min, then washed twice in 10% Hanks’ buffered salt solution (H10) and then each tissue fragments were plated individually into a 35 mm Petri dish (Corning). Next, the tissue was incubated in Leibovitz medium supplemented with 20% of fetal bovine serum (FBS) and 0.036 g/l of glutamine (L15–FBS) at 28.5 °C (Westerfield, Reference Westerfield2003).
Before use, donor cells were incubated in Hanks’ buffered salt solution without Ca2+ and Mg2+ at room temperature for 30 min before performing the nuclear transplant. No additional detachment treatment was realized. Once the cells had come off the substrate, L15–FBS was added and the cell dish was preserved at 5 °C throughout the daily experimental session.
Somatic cell nuclear transplant equipment
The nuclear transplant was performed using a Nikon inverted microscope equipped with two Leitz micromanipulators. During the manipulation process, the non-dechorionated eggs were held with a 260 μm outer diameter holding pipette and the cells were picked, lysed and injected into the eggs by means of a 10–12 μm inner diameter microinjection pipette. The microinjection pipette was fire polished, beveled and sharpened.
To perform the nuclear transplant, two separated drops were deposited in a Petri dish (Corning) (90 mm) and covered by mineral oil. One of them contained the donor somatic cells and was in L-15–FBS medium (300 mOsm). The other drop was the handling drop, which means the place where the nuclear transplant was performed, so the medium differed depending upon the nuclear transplant method tested in each case (see below).
The donor cell was picked up and lysed by aspiration with the injection micropipette before injection. The exact place where the cellular content was to be deposited was dependent on the nuclear transplant method tested in each case (see below).
Nuclear transplant was performed at different temperatures depending on the nuclear transplant method (see below).
Nuclear transplant methods
In order to carry out nuclear transplant whatever the status of the egg activation, three nuclear transplant methods were developed in which the somatic nuclear transplant was performed prior, simultaneous or posterior to the egg activation by the spermatozoa. As the aim of this work was to establish these methods technically and they were independently performed, no comparison of their technical efficiencies was made.
Method A: nuclear transplant prior to egg activation/fertilization
The somatic cell nucleus was inserted into the central region of the egg. To prevent egg activation, the transplant was performed in a handling drop composed of egg medium and the micromanipulation area was cooled down to 8 °C. This temperature around the handling zone was reached by cooling air cooled with liquid N2. Then, transplanted eggs were individually in vitro fertilized and cultured at 28.5 °C in the system water (Westerfield, Reference Westerfield2003).
Method B: nuclear transplant simultaneously with egg activation/fertilization
In this case, previously mixed non-activated eggs and sperm were kept at 8 °C and deposited individually in the handling drop that contained the system water, so that gametes were activated and fertilized. The micromanipulation area was not cooled (room temperature). The donor nucleus was injected in the incipient animal pole, just where the zygote nucleus was found (Wolenski & Hart, Reference Wolenski and Hart1987). The reconstructed embryos were incubated under the same conditions described previously.
Method C: nuclear transplant following egg activation/fertilization
In order to carry out nuclear transplant after fertilization, eggs and sperm were previously mixed and activated, then fertilized at room temperature as described in the second technique but, in this case, nuclear transplant was realized a few minutes after fertilization, just the time required for visualizing the completely defined animal pole. After injecting the donor nucleus at the animal pole, reconstructed embryos were incubated under the same conditions as described above.
Experimental design
The three techniques tested (A, B and C) were not carried out simultaneously, but were developed and assessed independently. In the three cases, in order to evaluate how post-ovulatory aging affects reconstructed embryo survival, two batches consisting of 3–5 eggs each were transplanted sequentially and compared in each session (A1, A2; B1, B2; C1, C2). Overall, the length of each experimental session did not exceed 90 min in all cases, the maximum time for efficient egg fertilization (Nüsslein-Volhard & Dahm, 2002). In this way, A1, B1 and C1 were manipulated during the first 45 min and A2, B2 and C2 during the last 45 min. A non-manipulated control group was fertilized at the end of each experimental session, at 90 min (CA, CB and CC) to test the ability of the egg to be fertilized at this time.
Given that the aim of the present work was mainly technical, only the embryo and larval survival rates of reconstructed embryos from the three techniques (A, B and C) were evaluated at different developmental stages: at mid blastula transition (MBT) stage (2 h after nuclear transplant); at 50% epiboly stage (7 h after nuclear transplant); at 24 h post-nuclear transplant; at 48 h post-nuclear transplant; and at larval stage (5 days after nuclear transplant) (Westerfield, Reference Westerfield2003). Moreover, at 24 h, 48 h and at 5 days post-nuclear transplant, normal and abnormal development was registered. In the non-manipulated control group, only the fertilization rate was evaluated by the survival rate at the MBT stage.
At least three replicates were done in all experimental groups. Results were analysed using the chi-squared test. When a single degree of freedom was involved, Yates’ correction for continuity was performed.
Results and discussion
The technical aspects for three methodologies for zebrafish somatic nuclear transplant prior to, simultaneously with and following egg activation/fertilization have been established in our laboratory. All three were developed and evaluated using non-irradiated eggs activated/fertilized by non-irradiated spermatozoa. In this way, the effects on survival and further embryo and larval development can be attributed exclusively to the transplant methodology employed, because the background noise due to the exigency of the spermatozoa for egg activation in zebrafish (Lee et al., Reference Lee, Webb and Miller1999) and the developmental limitations caused by a haploid condition (Nüsslein-Volhard & Dahm, 2002) would not exist in this case. On this subject, it has to be taken into account that, in these fish species, the triploid or tetraploid condition that could occasionally be derived from the addition of a somatic nucleus to the resident zygote pronuclei does not affect the embryonic, and even larval, development in a relevant way (Diter et al., Reference Diter, Guyomard and Chourrout1988; Peruzzi & Chatain, Reference Peruzzi and Chatain2003).
As a first general comment regarding the efficiency of the in vivo artificial collection of ovarian oocytes, the sexual behaviour synchrony showed by the separate fish colonies must be pointed out, in such a way that a large number of eggs could be collected in some sessions whereas no eggs might be obtained in others. Another relevant consideration concerns the fact that MBT nuclear transplant embryos were obtained in the great majority of the daily experimental sessions, whatever the transplant method used.
A technical advantage common to the three methodologies developed in our laboratory was the avoidance of previous dechorionation. In fish nuclear transplant, when the oocyte is activated but not enucleated, the donor nucleus is usually inserted into the perinuclear region of the oocyte, the closest as possible to the female nucleus, which is located in the cytoplasm subjacent to the plasma membrane under the micropyle (Amance & Iyengar, Reference Amanze and Iyengar1990). In various teleost species (such as medaka, catfish or tilapia), the animal pole position can easily be detected in preactivated oocytes because the micropyle can be visualized at this stage, although, unfortunately, this is not the case in zebrafish (Poleo et al., Reference Poleo, Denniston, Reggio, Godke and Tiersch2001).
One possible way to obviate such biological difficulty in zebrafish was attempted in method A, in which the somatic cell nucleus was inserted into the central region of the egg. This point of transplant was chosen because, in terms of probability, the central region will be closer to the female nucleus, whatever its real localization. So, the donor nucleus will be more likely to migrate to the microvilli cluster in animal pole. This is the place where fertilization occurs (Wolenski & Hart, Reference Wolenski and Hart1987) through the cytoplasmic flows together with the pronuclei at the time of the activation.
The earliest manipulated group (A1) reached significant higher survival rates compared with the most aged group (A2) both at epiboly and 24 h stage (Table 1). However, in more advanced stages (48 h and larval stage), the observed differences did not reach significant levels, probably due to the low number of embryos that developed to these stages. The egg control group (CA) fertilized at the end of each experimental session showed MBT rates equal or higher than the two experimental timing groups, indicating the maintenance of the egg fertilization ability until the end of the transplant session. Anyway, the larval survival rate was 7% (A1:6 larvae from 82 manipulated) in the first group and 5% (6 larvae from 132 manipulated) taking into account the overall two groups.
Method A: survival rates of nuclear transplant prior to the egg activation/fertilization.
a,bRows with different superscripts are statistically different.
It must be underlined that to cool down to 8 °C the temperature of the micromanipulation area was critical in maintaining the egg in a non-activated state during transplant. This initial strategy permits the impregnation of the donor nucleus in the reprogramming factors present in the egg at metaphase stage. In further experiments the effect of different times of donor nucleus impregnation before the activation will be tested.
In method B (Table 2), nuclear transplant and fertilization were performed at the same time, which meant transplanting the donor nucleus while the egg was activating. Egg activation and fertilization are both marked by elevation of the chorion and a dramatic reorganization of the yolk cytoplasm. In this way, the animal pole is segregated through the place where the female nucleus will be located (Wolenski & Hart, Reference Wolenski and Hart1987). This approach enabled detection of the incipient animal pole in order to deposit the somatic nucleus in the female perinuclear region.
Method B: survival rates of nuclear transplant simultaneously to the egg activation/fertilization.
a,bRows with different superscripts are statistically different.
The survival rate differences between the first (B1) and second group (B2) did not reach significant levels in any assessment. However, it must be pointed out that these differences decreased over time and that the survival rates were finally similar at the 48 h stage (B1: 19% vs B2: 20%). In this case, the fertilization rate of the final control group (CB) was lower than the Total B, even than that found in the B2, which involved an obvious effect of the egg aging and a very slightly negative effect of the assayed nuclear transplant technique.
When nuclear transplant was performed following egg fertilization, method C (Table 3), the variability in the time required for showing the perivitelline space as an activation signal should be pointed out. This factor represented a critical point due to the technical difficulty involved in fixing the egg with the holding pipette, because the egg rotated inside the chorion while this space was increasing. With activation, the micropyle can be more easily detected but, a few seconds after the fertilization, the chorion hardened and the micropyle sealed (Poleo et al., Reference Poleo, Denniston, Reggio, Godke and Tiersch2001). This factor made it very difficult to insert the donor nucleus through this point even if the microinjection pipette was fire polished, beveled and sharpened, as in our case. A possibility to obviate this difficulty could be to dechorionate the egg after fertilization, but this technique is time consuming and the first cleavage is very early (minutes) in zebrafish. Moreover, the reduction of the temperature to arrest development before the MBT stage involves embryo lethality (Francisco-Simao et al., Reference Simao, Perez-Camps and Garcia-Ximénez2007).
Method C: survival rates of nuclear transplant posterior to the egg activation/fertilization.
a,bRows with different superscripts are statistically different.
Regarding survival differences between both handling groups (C1 and C2), it should be emphasized that, as in technique B, the differences observed did not reach significant levels in any case. However, these differences were relevant, ranged from a differential of 10% at MBT stage up to 15 points at the larval stage. The explanation for not reaching significance level could be the low number of surviving embryos that developed to this stage, owing to the aforementioned technical difficulty. The survival rate of the control group (CC), compared with the C1 and C2 groups, showed again a considerable egg aging effect, which means that the time for performing the nuclear transplant in zebrafish must be shorter in order to avoid (or minimize) such a pronounced negative effect.
Adult specimens showed the gold phenotype. This fact does not discard a possible mosaicism or that the reconstructed embryos with the donor nucleus incorporated more efficiently did not reach adult stages. However, it has to be taken into account that the nuclear fate of the transplanted nuclei was not analysed because, as mentioned previously, the main aim of this work was to establish three nuclear transplant protocols in zebrafish by a technical approach. In this way, after the establishment of these three techniques presented, the nuclear fate will be studied, as well as its integration degree and form in the specimens.
In conclusion, the reasonable technical efficiencies achieved in the present work make the use of these three methods interesting for future reprogramming studies by nuclear transplant in this species.
Acknowledgements
We would like to thank Mr Javier Rubio for technical support regarding colony maintenance, microinstrument making and egg micromanipulation. The authors thank Mr Neil Macowan for revising the English version of the manuscript.
This work is part of the project no. AGL2008–03275. Financial support for this project was denied by the CICYT of the Ministerio de Ciencia e Innovación del Gobierno de España.
