The Early Development And Fate Determination Of T Cells In Zebrafish


T lymphocytes are key components of the adaptive immune system and play a central role in cell-mediated immunity. Although many subsets of T lymphocytes have been identified and their functions have been characterized, the early development and fate determination of T lymphocytes remain unclear. Based on the fact that zebrafish and vertebrate immunity has many aspects in common(Renshaw and Trede, 2011), we choose zebrafish as an experimental model to study the development of T lymphocytes. As indicated previously, the first wave of T cells are all CD4+ cells(Tian et al., 2017), we are eager to know why CD8+ T cells cannot be generated in the first wave and whether there exists differences between fetal and adult T cell fate determination in zebrafish.


Hematopoiesis – a process by which hematopoietic stem cells(HSCs) are formed and therefore blood cells are generated, is an essential process in the development of embryos. Studies in both mice and teleost have shown that hematopoiesis consists of multiple waves from distinct sources(Waas and Maillard, 2017). Generally, the first wave produces embryonic erythrocytes and macrophages(Palis et al., 1999 ), the second wave gives rise to erythromyeloid progenitors and the third wave generates HSCs. Most lymphocytes are produced after the third wave of hematopoiesis.

T lymphocytes are key components of the adaptive immune system and play a central role in cell-mediated immunity(Pancer and Cooper, 2006). The most common T cell subsets are helper T cells and cell toxic T cells, which express the co-receptor gene CD4 and CD8, respectively. CD4 and CD8 are cell surface molecules that are believed to work by forming co-receptor with T cell receptor(TCR) for recognition of MHC. It is generally believed that in adult mammals, CD4+ and CD8+ T cells are derived from CD4+8+ precursors by specific interaction of TCR with thymic major histocompatibility complex. This conclusion is based on the experiments did on transgenic mice(Teh et al., 1988). However, in the early stages of development, CD4+ and CD8+ T cells seems to be independent. Although it’s known that the first wave of CD4+ T cell arises from the aorta-gonad-mesonephros(AGM) and yolk sac(YS) in mice and from AGM and posterior blood island(PBI) in zebrafish(Nishikawa et al. ,1998; Tian et al., 2017). It remains elusive where and when the first wave of CD8+ T cells arises in both mammals and bony fish. If this problem can be solved, we’ll have a better understanding of T cell development and may have more methods to deal with diseases caused by immunity deficiency.

Recently, zebrafish (Danio rerio) is widely used as a model system to study human hematopoietic development. This is based on the fact that hematopoietic programs are largely conserved between mammals and zebrafish (Zon, 1995). Besides, zebrafish biology allows access to all developmental stages and provides convenience for performing high-resolution imaging of immune cells in vivo(Dee et al., 2016). Moreover, like mammals, zebrafish produce multiple classes of T cell subsets (including αβ and γδ T cells) and produce analogous mature blood cells. Therefore, zebrafish serves as a perfect vertebrate model to study the ontogeny of different waves of T lymphocytes.

In this study, we use zebrafish transgenic line combined with time-lapse imaging and Infrared-mediated temporal-spatial cell labeling to trace the CD8+ T cells and try to gain a better understanding of the development of T cells. To generate transgenic zebrafish, we use bacterial artificial chromosome(BAC) containing the reporter gene and use transposon-mediated method provided by others(Suster et al., 2011) to integrate the reporter gene into zebrafish genome. BACs are widely used in transgenesis as they can hold as large as 300kb genomic fragments, which often contains the complete structure of a gene(Monaco and Larin, 1994). Combined with Tol2 transposon element, single-copy integrations can usually be obtained, which means it can reveal the real expression level of the targeted gene(Chandler et al., 2007). Thus, our study will not only show the developmental details but also reveal the expression level of related genes in T cells.


As the study is currently in progress, we haven’t finished constructing the transgenic lines. After finishing the Tg(CD8: mCherry; CD4: eGFP) line, we’ll first check when CD8+ T cells first emerge from thymus. Then, more importantly, we’ll check if there are immature T cells co-expressing CD4 and CD8. If so, it is very likely that T cell fate determination in teleost is similar with that in mammals(Teh et al., 1988). So the difference between the first wave and adult T cells(the first wave only gave rise to CD4+ cells) may owe to the fact that:1. The fetal thymus is not well developed, i.e. lacking some kinds of antigen; 2. The fate of the first wave of T cells have been determined much earlier; 3 .The primitive macrophages in thymus are functionally different with mature macrophages. For condition 1, we might label some MHC related genes and the T cell homing genes(CCR9a/b) or inject some antibodies to analyze if the fate determination of early T cells can be changed by thymus related materials. For condition 3, we should label both macrophages and T cells and focus on their contact and communication in thymus, compare it with the conditions in adult fish to make sure whether there is any difference between macrophage-T cell connection. For, condition 2, maybe we can extract fetal T cells and culture them in vitro or in adult zebrafish, it’ll be a lot of work.

What’s more, if in adult fish, there are no cells co-expressing CD4 and CD8, we can conclude that fate determination of T cell in zebrafish is very likely to be different from that in mammals. Then we can not only focus on the earlier waves of T cells but also the entire development details of T lymphocytes in zebrafish. We can also conclude that T cell fate determination is not conserved in mammals and teleost. Then, in order to make clear the underlying mechanism in teleost T cell fate determination, maybe we’ll pay attention to genes regulating the expression of MHC molecules or regulating T cell homing(such as the CCL class genes). Which method to use depends on experiment results.



All zebrafish were maintained according to standard protocol(Wester eld,2000 ). In this study, zebrafish lines Tg(CD8: loxp-IFP-loxp-eGFP), Tg(lck: loxp-IFP-loxp-DsRed), Tg(CD4: mCherry), Tg(CD8: eGFP), Tg(lck: NTRO-eGFP), Tg(lck: eGFP) and Tg(hsp70: mCherry-T2a-CreERT2) were used. All animal experiments were conducted following the guidelines of the Animal and Plant Care Facility of the Hong Kong University of Science and Technology (HKUST) and were approved by the Animal Ethics Committee of HKUST.

Generation of BACs

All BACs are generated according to Maximiliano L Suster’s method(Suster et al., 2011). Taking the BAC(lck: eGFP) as an example, the lck BAC clone(CH211-230I4) was purchased via ZFIN. To construct the BAC used to generate Tg(lck: eGFP), we first prepared the eGFP cassette through PCR. In order to generate the exon 1 recombination vector, two homologous arms were also PCR amplified. The primers used are as following: left HA FP: 5’-ATTGGGTACCGGGCCCCCCCAATTGTTAGCAGAGTATCTGTCG-3’, left HA RP: 5’- CCCTTGCTCACCATGGTGGCTTTTTCCAGTTTCCAGCACAAGCAT-3’, right HA FP: 5’- GCAGCTCCAGCCTACACGCGGGTGATTTTATCCCTCTTCTATG-3’, right HA RP: 5’- CGGTGGCGGCCGCTCTAGAAACAATTAACTTCTACGGTGCT-3’. The length of left and right homologous arm are 658bp and 548bp respectively. As nonspecific binding exists(Figure 1), we cut the gel and extracted DNA using OMEGA Gel Extraction Kit. After collecting all PCR products, we linked left and right HA, reporter cassette and pBluescript SK(PBSK) plasmid together using Gibson assembly according to standard protocol(Perkel, 2014). The assembled plasmid was then introduced into competent E.coli cells for selection.

Correctly assembled PBSK plasmid was selected through colony PCR, primers used are: M13 FP: 5’- GTAAACGACGGCCAGT-3’ and M13 RP: 5’- GTCATAGCTGTTTCCTG-3’(Figure 2). To insert the target cassette through homologous recombination, the pSIM vector, which contains heat-inducible recombinase functions, was electroporated into cells that contain the BAC. These bacteria were maintained at 32 °C for 24 hours. After that, the cells were heat shocked at 42 °C for 15min and were transformed into competent cells according to the chemical transformation protocol provided by others(Suster et al., 2011). Then, the cells were mixed with iTol2 cassette to construct BACs that containing iTol2 transposon gene. The reporter gene cassette was inserted into BAC backbone using the same method. Final BACs were extracted and cells containing the final BAC were raised for further usage.

Generation of transgenic lines

The BAC DNA was purified and injected into one-cell stage zebrafish embryos together with Tol2 mRNA which was synthesized previously. Ten hours after microinjection, in each group, 8 embryos were randomly picked to confirm the excision of Tol2-mediated BAC insertion in injected embryos(Figure3). If the injection is confirmed to be successful, the remained embryos will be raised to adulthood. To perform Infrared-mediated temporal-spatial cell labeling, zebrafish line Tg(lck: loxp-IFP-loxp-Dsred) was crossed with line Tg(hsp70: mCherry-T2a-CreERT2) and then was further crossed with Tg(lck: NTRO=eGFP) to generate Tg(hsp70: mCherry-T2a-CreERT2; lck: loxp-IFP-loxp-eGFP; lck: NTRO-eGFP) embryos. These embryos are used to study the recovery of thymus localized T cell. To identify the different development track of CD4+ T cells and CD8+ T cells, Tg(CD4: mcherry) and Tg(CD8: eGFP) was crossed to generate Tg(CD4: mCherry; CD8: eGFP) line.

Infrared-mediated temporal-spatial cell labeling

The infrared-mediated temporal-spatial cell labeling system was set according to previous study(Xu et al., 2015). The power of the 1345nm light on the sample surface is around 65-80mW and the time of each shock is roughly 2min. Zebrafish line Tg(hsp70: mCherry-T2a-CreERT2; lck: loxp-IFP-loxp-eGFP; lck: NTRO-eGFP) was used. One day before labeling, the fish were treated with metronidazole(MTZ) to kill all thymus localized T cells according to standard protocol(Mathias et al., 2014). Then the PBI area was labeled by irradiating two or three positions along the PBI region at around 5 dpf. The recovery of thymus localized T cells was analyzed by time-lapse imaging or in situ hybridization. In another experiment, the Tg(CD8: lozp-IFP-loxp-eGFP) line was also labeled at PBI to make clear where the first wave of CD8+ T cells come from and whether HSC-independent hematopoietic progenitors are able to give rise to CD8+ T cells.

Mutant identification

To choose F0 fish with germline transmission, embryos were first observed under fluorescent microscope, the ones with stronger fluorescence response were selected and raised to adulthood. Then, the fish were crossed with wild-type(ABS) line, those who can produce offspring with fluorescence signals are believed to have stable germline integration(Suster et al., 2011).

Whole-mount in situ hybridization

Antisense DIG-labeled RNA probe was synthesized in vitro using specially designed DNAs. Fish were dehydrated and fixed in 4% PFA at room temperature. The staining process was according to a previous study (Thisse and Thisse, 2008). Fluorescent microscopy was used to detect the signals.


Adaptive immunity has been considered as one of the key evolutionary innovations in the emergence of vertebrates(Boehm, 2012) which improved the efficiency of immune activity significantly. In this study, we use zebrafish as an experimental model to expand our understanding of the development and fate determination of T lymphocytes. We’ll try to identify why there are no CD8+ T cells in early development stages and add more details to T cell fate determination.

Our understanding of hematopoiesis in zebrafish lags far behind our state of knowledge for mammals(Dee et al., 2016). However, this gap is likely to narrow rapidly as zebrafish serves as a better experimental model than mammals in many research areas. In addition, advances in technology such as transgenic technology and the optimizing of fluorescent proteins have provided significant convenience in molecular analysis and in single cell tracking. We envisage more great discoveries will be performed using zebrafish, taking advantage of its rapid generation time, visualization of single cells in situ and availability for genetic manipulation.

Leave a Comment

Your email address will not be published. Required fields are marked *