|Home | About | Journals | Submit | Contact Us | Français|
In multicellular organisms, the cell cycle is a fundamental feature of cellular physiology that is critical for normal development, organogenesis and tissue homeostasis. Reflecting this central role, the molecular pathways that regulate cell division in eukaryotes are evolutionarily conserved. Aberrations in the control of the cell cycle are common in degenerative diseases and cancer. Therefore, analysis of the cell cycle in non-mammalian organisms can illuminate the processes underlying human development and disease. Forward genetic screens in yeast and Drosophila have been invaluable for gene discovery and have made important contributions to understanding pathways regulating cell proliferation. Importantly, it has been found that the human orthologs of some genes identified in these organisms are misexpressed in human tumors (Hariharan and Haber 2003). Zebrafish have proven to be an excellent model of early vertebrate development (Driever, Solnica-Krezel et al. 1996; Haffter, Granato et al. 1996) and also of a wide variety of human diseases such as cancer, anemia, cardiovascular defects, neuromuscular conditions, kidney disease and host-pathogen interaction, to name a few examples (Ackermann and Paw 2003; Bassett and Currie 2003; Lambrechts and Carmeliet 2004; Miller and Neely 2004; Drummond 2005; Goessling, North et al. 2007; Hsu, Wen et al. 2007).
The particular advantages that make zebrafish ideal for developmental embryology—including external fertilization of oocytes, transparent embryos, and rapid embryonic development—also provide the opportunity to study early cell divisions, tissue-specific cellular proliferation and, more broadly, the role of cell-cycle genes in development and disease. A number of methods and markers have been successfully applied to investigate the cell cycle in zebrafish embryos, including video microscopy (Kane, Warga et al. 1992; Kane 1999), histone-GFP fusions (Pauls, Geldmacher-Voss et al. 2001), BrdU labeling (Link, Fadool et al. 2000; Baye and Link 2007), Proliferating Cell Nuclear Antigen (PCNA) RNA and protein expression (Wullimann and Knipp 2000; Koudijs, den Broeder et al. 2005), phosphohistone H3 (pH3) immunohistochemistry (Shepard, Amatruda et al. 2005), and minichromosome manintenance protein expression (Ryu and Driever 2006).
Studies of the developing zebrafish embryo have revealed similarities to the early cell divisions of other vertebrates, such as Xenopus. In the zebrafish, the first seven cell divisions are synchronous and cycle rapidly between DNA replication (S phase) and mitosis (M phase) without the intervening gap phases, G1 and G2 (Kimmel, Ballard et al. 1995). The midblastula transition (MBT) ensues during the 10th cell division, which is approximately 3 h post fertilization. MBT is accompanied by loss of division synchrony, increased cell cycle duration, activation of zygotic transcription, and the onset of cellular motility (Kane and Kimmel 1993). Embryonic cells first exhibit a G1 gap phase between the M and S phases during MBT. Recently, Dalle Nogare et al. demonstrated that during cycles 11–13, embryonic cells acquire a G2 phase in a transcription-independent fashion, through inhibition of Cdk1 and its activating phosphatase, Cdc25a (Dalle Nogare, Pauerstein et al. 2008).
Further understanding of cell cycle regulation in zebrafish embryos was obtained by studying their responses to various cell cycle inhibitors, including aphidocolin, hydroxyurea, etoposide, camptothecin, and nocodazole (Ikegami, Rivera-Bennetts et al. 1997; Ikegami, Zhang et al. 1997; Ikegami, Hunter et al. 1999). Exposure to these agents after MBT induces cell cycle arrest, sometimes accompanied by initiation of an apoptotic program. However, prior to MBT, the embryonic cells continue to divide, often with deleterious effects, after exposure to cell cycle inhibitors. These studies indicate that zebrafish embryos do possess cell cycle checkpoints, but they are not functional until after MBT.
Later developmental stages of zebrafish embryogenesis provide the opportunity to study the cell cycle in distinct tissue types. Studies of cell cycle regulation in older embryos (10–36 h post fertilization) have focused on the developing eyes and central nervous system. Lineage analysis of CNS progenitor cells revealed a correlation between morphogenesis and cell cycle number, implying that the nervous system development may be at least partially regulated by the cell cycle (Kimmel, Warga et al. 1994). Whereas most developing vertebrate embryos exhibit a constant lengthening of the cell cycle duration throughout development, meticulous analysis of cell number in the developing zebrafish retina revealed a surprising mechanism of modulated cell cycle control. Li and colleagues (Li, Hu et al. 2000) reported that the retinal cell cycle duration temporarily slows between 16–24 hours post fertilization (hpf), followed by an abrupt change to more rapid cell divisions.
Several studies have elucidated a role for the zebrafish cell-cycle machinery in tissue differentiation during development and in the regenerative response to injury. Bessa et al. found that Meis1, a marker of the eye primordium, promotes G1-S progression and a block of differentiation in the zebrafish eye through regulation of Cyclin D1 and c-myc expression (Bessa, Tavares et al. 2008). Fischer and co-workers showed that loss of caf1b in zebrafish (by mutation or MO injection) leads to an S-phase arrest and eventual apoptosis that can be rescued by p53 deficiency. However, loss of caf1b also leads to a block in differentiation in tissues that express caf1b, implicating caf1b in the switch from proliferation to differentiation (Fischer, Prykhozhij et al. 2007). The effect of loss of early mitotic inhibitor 1 (emi1) on somite formation was evaluated by Zhang et al. These authors found that cell-cycle progression was required for proper somite morphogenesis, but not for formation of the segmentation clock (Zhang, Kendrick et al. 2008). The role of the cell cycle in regeneration has also been assessed. Certain traumas result in loss of hair cell precursors, which results in deafness in vertebrates. Hernández and co-workers used BrdU labeling and transgenic GFP reporter lines to study hair cell regeneration, identifying proliferation- dependent and –independent mechanisms of hair cell renewal (Hernandez, Olivari et al. 2007).
Several groups have carried out forward-genetic screens to identify mutations that alter cell proliferation in embryos. Shepard et al. used phosphohistone H3 (pH3) as a marker of cell proliferation in a 2-generation haploid genetic screen. They identified seven mutant lines with different alterations in pH3 immunoreactivity. At least two of these lines demonstrate aneuploidy and increased cancer susceptibility as heterozygotes (Shepard, Amatruda et al. 2005; Shepard, Amatruda et al. 2007). Using a similar screening strategy, Pfaff et al. identified a further set of genes required for cell proliferation mutants, among which was SIL ( for Scl-Interrupting Locus), which was identified as a novel, vertebrate-specific regulator of mitotic spindle assembly (Pfaff, Straub et al. 2007). Koudijs and co-workers used Proliferating Cell Nuclear Antigen (PCNA) expression in the CNS as a readout to identify new mutations in repressors of the Hedgehog (Hh) signaling pathway (Koudijs, den Broeder et al. 2005). Finally, another screen for genes that control eye growth uncovered two zebrafish lines mutant for the anaphase-promoting complex/cyclosome (APC/C) (Wehman, Staub et al. 2006). Loss of APC/C results in a loss of mitotic progression and apoptosis; in this study, co-labeling with BrdU and pH3 revealed cells undergoing mitotic catastrophe.
In this chapter, we provide protocols to characterize the various phases of cell division in zebrafish embryos, and protocols to detect DNA damage, senescence and cell death. Assays discussed in this chapter include: DNA content analysis by flow cytometry, whole-mount embryonic antibody staining, mitotic spindle analysis, BrdU incorporation, cell death analysis, and in situ hybridization with cell cycle regulatory genes. Each assay targets different phases of the cell cycle and in total create a detailed picture of zebrafish embryo cell proliferation. Although our studies have focused on embryonic assays for cell cycle characterization, it is likely that these protocols can be modified to study adult tissues. These protocols can be applied to a variety of experiments, such as characterization of the cell cycle phenotypes of mutants or the analysis of RNA overexpression and morpholino knockdown of cell cycle regulatory genes. Furthermore, the genetic tractability of the zebrafish system (Patton and Zon 2001) makes it an excellent organism in which to pursue forward genetic screens for mutations or chemical screens for novel compounds that alter cell division using one or more of these cell cycle assays.
A profile of the cell cycle in disaggregated zebrafish embryos or adult tissue can be obtained through DNA content analysis. In this technique, cells are stained with a dye that fluoresces upon DNA binding, such as Hoechst 33342 or propidium iodide. The intensity of fluorescence is proportional to the amount of DNA in each cell (Krishan 1975). Analysis by fluorescence activated cell sorting (FACS) generates a histogram showing the proportion of cells that have an unreplicated complement of DNA (G1 phase), those that have a fully replicated complement of DNA (G2 or M phase) and those that have an intermediate amount of DNA (S phase).
All steps are performed on ice except for the dechorionation (step 1) and RNAse incubation (step 9).
Note: Samples can also be fixed in Ethanol, allowing multiple samples or time points to be collected for subsequent analysis.
Histone H3 phosphorylation is considered to be a crucial event for the onset of mitosis and this antibody has been widely used in Drosophila and mammalian cell lines as a mitotic marker (Hendzel, Wei et al. 1997). Two members of the Aurora/AIK kinase family, Aurora A and Aurora B, phosphorylate histone H3 at the serine 10 residue (Chadee, Hendzel et al. 1999; Crosio, Fimia et al. 2002). Increased serine 10 phosphorylation of histone H3 has been seen in transformed fibroblasts (Chadee, Hendzel et al. 1999), suggesting that this antibody could make an excellent marker for cell proliferation in the zebrafish as well as detecting cell cycle mutations that may result in transformed phenotypes. In zebrafish, the phospho-histone H3 antibody (pH3) stains mitotic cells throughout the embryo (Figure 1A). pH3 staining in developing organs like the nervous system increases as they undergo proliferation during distinct developmental stages.
Study of the mitotic spindle and centrosomes is an important step in understanding mutants with cell cycle defects, particularly those whose phenotypes appear to be related to problems in mitosis. Genomic instability is one of the main alterations seen in human cancers and such unequal segregation of chromosomes can be caused by problems in mitotic spindle formation or centrosome number (Kramer, Neben et al. 2002). In this protocol, anti-a-tubulin labels the mitotic spindle, anti-g-tubulin the centrosome, and DAPI the DNA.
5-bromo-2-deoxyuridine (BrdU) is a nucleoside analog that is specifically incorporated into DNA during S-phase (Meyn, Hewitt et al. 1973) and can subsequently be detected with an anti-BrdU specific antibody. This technique has been used to label replicating cells in zebrafish embryos (Larison and Bremiller 1990) and adults (Rowlerson, Radaelli et al. 1997). The following protocol is designed to label a fraction of proliferating cells in zebrafish embryos, to allow comparison of the replication fraction of different embryos (Figure 1 B). If the embryos are chased for varying amounts of time after the BrdU pulse, then fixed and stained for both BrdU and pH3 (section B), the transit of cells from S phase into G2/M can be assessed.. This is useful in analyzing mutants with mitotic phenotypes.
The Comet Assay, also known as the Single Cell Microgel Electrophoresis (SCGE) assay, is a highly sensitive technique that is used to detect DNA damage at the single cell level(Singh, McCoy et al. 1988). Cells are embedded into a thin agarose gel, through which a current is run allowing for migration of DNA. Smaller fragments of DNA, resulting from DNA damage, will travel more quickly and appear as a tail to the nucleus “comet head”. The comets can be visualized using a nuclear stain, such as SYBR green, and visualized under a fluorescent microscope. The following protocol is designed to isolate cells from zebrafish embryos and detect any kind of DNA damage. Variations of this technique allows for specific detection of double-stranded break
The study of cellular senescence was initiated by Hayflick and Moorhead and was first described in 1961(Hayflick and Moorhead 1961). Cellular senescence pertains to the cessation of cell replication and certain morphological and transcriptional changes that occur when cells permanently cease dividing. Although it is unclear whether the events that occur during in vitro cellular senescence also occur during organismal aging (Masoro 2006; Hayflick 2007) studies have revealed strong connections between cellular senescence, cancer, and age-related diseases (Campisi 2005). Cellular senescence most likely arose evolutionarily as a mechanism to defend against tumorigenesis (Shay and Roninson 2004). When a cell is afflicted by stress that may result in transformation (such as oxidative stress, DNA damage or overepxpression of oncogenes) tumor-suppressor genes such as p53 may force the cell to undergo senescence-induced arrest. Arrested cells are functional but are not a risk for tumor initiation. Senescence also occurs as the ends of chromosomes, the telomeres, shorten. During reach replication cycle, if no active telomerase is present (Bodnar, Ouellette et al. 1998), the telomeres shorten, leading eventually to critically short telomeres which may interfere with gene expression and genomic stability (Shay and Wright 2006). Normal cells senesce before telomeres shorten to the point of causing genomic instability, therefore instilling a counting mechanism which confirms Hayflick’s observation in 1961 (Shay and Wright 2006).
Senescent cells lose sensitivity to mitogens or growth factors, repress cell cycle genes, such as cdk2, and become insensitive to apoptotic signals. Morphological changes occur resulting in an enlarged shape and flattened body (Ben-Porath and Weinberg 2005), as well as expression of unique markers, many of unknown function, such as β-galactosidase activity at pH 6.0 (Dimri, Lee et al. 1995). Kishi and co-workers have used senescence-associated as β-galactosidase staining in several studies to characterize senescence in normal and mutant zebrafish embryos and during aging of zebrafish adults (Kishi, Uchiyama et al. 2003; Kishi 2004; Tsai, Tucci et al. 2007; Kishi, Bayliss et al. 2008).
We have used the Senescence-Associated Beta-Galacotsidase Detection Kit from Sigma (CS 0030). The following protocol adapts the manufacturer’s instructions specifically for use with zebrafish embryos, and is kindly provided by Jenny Richardson and Dr. Elizabeth Patton, Edinburgh Cancer Research Centre:
An alternative protocol was described by Dr. Shuji Kishi and co-workers in a recent paper describing a senescence-based genetic screen (Kishi, Bayliss et al. 2008). The following protocol is adapted from Kishi, S et al., “The identification of zebrafish mutants showing alterations in senescence-associated biomarkers”, PLoS Genet. 2008 Aug 15;4(8):e1000152:
Apoptosis is a form of programmed cell death that eliminates damaged or unneeded cells. It is controlled by multiple signaling pathways that mediate responses to growth, survival or death signals. Cell cycle checkpoint controls are linked to apoptotic cascades and these connections can be compromised in diseases, including cancer. The defining characteristics of apoptosis are membrane blebbing, cell shrinkage, nuclear condensation, segmentation and division into apoptotic bodies that are phagocytosed (Wyllie 1987). The DNA strand breaks that occur during apoptosis can be detected by enzymatically labeling the free ends with modified nucleotides which can then be detected with antibodies (Gavrieli, Sherman et al. 1992).
Another method of apoptotic cell detection that can be performed on living embryos is acridine orange staining. The basis of this method is that the ATP-dependent lysosomal proton pump is preserved in apoptotic but not necrotic cells therefore apoptotic cells will take up the acridine orange dye whereas living or necrotic cells will not (Darzynkiewicz, Bruno et al. 1992). This method is useful for identifying mutants based on an apoptotic phenotype in order to further characterize them in living assays.
RNA expression analysis by in situ hybridization of antisense probes in whole-mount zebrafish embryos is a commonly used technique to localize expression of developmental regulatory genes. While the technique is not exceptionally quantitative, it can reveal stark differences in gene expression. More quantitative analysis of gene expression, such as Northern blotting, RT-PCR, or real-time PCR do not permit the examination of alterations in tissue-specific expression or an expression pattern.
Cell division is a highly controlled process that involves regulation at both the transcriptional and post-translational stages. Cyclins are a class of proteins that play critical roles in guiding cells through the G1, S, G2, and M phases of the cell cycle by regulating the activity of the cyclin-dependent kinases. The name cyclin alludes to the fact that their expression levels oscillate between peaks and nadirs that are coordinated with particular phases of the cell cycle (reviewed in (Murray 2004). The tightly regulated expression of these important cell cycle genes incorporates transcriptional, translational, and posttranslational controls. Many genes involved in cell cycle regulation are specifically expressed during the cell cycle phase in which they act.
Zebrafish orthologs of cell cycle regulatory genes such as PCNA and cyclins have been found to possess similar expression patterns throughout the proliferative tissues of developing zebrafish embryos (C. Thisse, B. Thisse, unpublished and www.zfin.org). In situ hybridization for cell cycle regulatory genes can be performed using previously published in situ hybridization protocols (Thisse, Thisse et al. 1993; Thisse, Thisse et al. 1994; Jowett 1999).
Another way to probe the cell cycle is via chemical agents. Chemical screens could identify novel compounds that are useful tools for studying the cell cycle. Furthermore, mutations in cell cycle genes are commonly found in human cancer. Given the need to improve upon current cancer therapy, one approach is to identify small molecule suppressors that bypass the consequences of specific cell cycle gene mutations. Akin to the use of genetic modifier screens to identify secondary mutations that enhance or suppress a primary defect (St Johnston 2002), chemical suppressor screens would directly identify small molecules that rescue a genetic phenotype. If the phenotype is disease-related, such compounds might represent lead therapeutic agents.
Zebrafish have recently been utilized in chemical screens to identify compounds that perturb specific aspects of development (Peterson, Link et al. 2000; Khersonsky, Jung et al. 2003; Peterson, Shaw et al. 2004; den Hertog 2005; Bayliss, Bellavance et al. 2006; Anderson, Bartlett et al. 2007). The zebrafish system offers several advantages for chemical screens, providing information on tissue specificity and toxicity, and accounting for compound activation via drug metabolism. Furthermore, cells are not transformed and are in their normal physiological milieu of cell-cell and cell-extracellular matrix interactions. Murphey and co-workers carried out a high-throughput chemical screen to detect small molecules capable of perturbing the cell cycle during zebrafish development, identifying several compounds that were not previously detected in cell-based screens of the same library (Murphey, Stern et al. 2006). As another application of this technique, Stern et al. screened a 16,000-compound library to identify small molecules capable of suppressing the cell proliferation defect in the crash&burn cell cycle mutant (Stern, Murphey et al. 2005). This technology could easily be applied to other cell cycle mutants and could be modified to use cell cycle assays other than pH3 staining. In addition, such chemical suppressor screens could be applied to any zebrafish model of human disease (Dooley and Zon 2000). For these reasons, we provide a detailed protocol below.
The following protocol can be repeated weekly giving a throughput of over 1000 compounds per week for a recessive lethal mutation. In the case of homozygous viable mutants, the throughput could be improved by using fewer embryos (3–5) per well in 96-well plates.
Given the power of zebrafish in forward vertebrate genetics and organism-based small molecule screens, the system will nicely complement traditional model organisms for studying the cell division cycle. Many of the assays that are commonly used to probe the cell cycle in systems such as yeast, Drosophila, and mammalian cells can be used in the zebrafish. The protocols outlined in this chapter can be utilized to characterize known mutants for alterations in cell proliferation or, alternatively, can be used to screen for more cell cycle mutants. Given that zebrafish embryos are amenable to gene knockdown via antisense morpholino-modified oligonucleotides and overexpression by mRNA injection, these protocols can also be used to study cell cycle genes in the zebrafish without generating a mutant.
|Alkaline Solution||0.6g NaOH, 250uL 200mM EDTA pH10, to 50mL ddH20|
|Buffer||12g NaOH, 2mL 500mM EDTA pH 8, to 1L ddH20|
|Anti-BrdU||Roche cat # 1170 376|
|Block||2% blocking reagent (Roche 1096 176), 10% fetal calf serum, 1% dimethylsulfoxide in PBST.|
|BrdU Block||0.2% blocking reagent (Roche 1096 176), 10% fetal calf serum, 1% dimethylsulfoxide in PBST. The lower concentration of blocking reagent improves detection.|
|E3||5mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4.|
|Mesh||Small Parts, inc. 105 µm mesh is cat # U-CMN-105D. 40 µm mesh is cat # U-CMN-40D.|
|PBS||Phosphate-buffered saline, pH 7.5.|
|PBST||1X PBS with 0.1% (v/v) Tween-20|
|Pellet pestle & tubes||Fisher, cat # K749520-0090|
|PFA||4% paraformaldehyde buffered with 1× PBS|
|Propidium Iodide||0.1% Sodium Citrate, 0.05mg/mL propidium iodide, 0.0002% Triton X-100 (added fresh)|
|Sigma Senescence Fixation Buffer 10X (Catalog Number F1797)|
|Contains 20% formaldehyde, 2% glutaraldehyde, 70.4 mM Na2HPO4, 14.7 mM KH2PO4, 1.37 M NaCl, and 26.8 mM KCl|
|Sigma Senescence Staining Mixture|
|(Prepare just prior to use)|
|Mix the following for preparation of 10 ml of the Staining Mixture:|
|1 mL of Staining Solution 10X Buffer (Catalog Number S5818)|
|125 µL of Reagent B (Catalog Number R5272)|
|125 µL of Reagent C (Catalog Number R5147)|
|0.25 mL of X-gal Solution (Catalog Number X3753)|
|8.50 mL of ultrapure water|
|Screening medium||E3 supplemented with 1% DMSO, 20 µM metronidazole, 50 units/mL penicillin, 50 mg/mL streptomycin, and 1mM Tris pH 7.4.|
We thank Len Zon and members of the Zon laboratory for useful discussions. Original versions of many of these protocols were worked out by Jennifer L. Shepard, Ryan Murphey, Howard M. Stern, Kathryn L. Pfaff and J.F.A. D.V. was supported by NIH Training Grant 5 T32 GM008203 and J.F.A. was supported by grants from the Lance Armstrong Foundation, the Amon G. Carter Foundation, the Welch Foundation and NIH/NCI grant 1R01CA135731.
1Items in boldface indicate reagents and supplies listed in Section V.