|Home | About | Journals | Submit | Contact Us | Français|
Apoptosis is a genetically controlled process of cell suicide that plays an important role in animal development and in maintaining homeostasis. The nematode Caenorhabditis elegans has proven to be an excellent model organism for studying the mechanisms controlling apoptosis and the subsequent clearance of apoptotic cells, aided with cell-biological and genetic tools. In particular, the transparent nature of worm bodies and eggshells makes C. elegans particularly amiable for live cell microscopy. Here we describe a few methods for identifying apoptotic cells in living C. elegans embryos and adults and for monitoring their clearance during embryonic development. These methods are based on Differential Interference Contrast microscopy and on fluorescence microscopy using GFP-based reporters.
During an animal's development and adult life, a large number of unwanted cells are eliminated by programmed cell death, or apoptosis. Apoptotic cells are rapidly engulfed (via phagocytosis) by phagocytes, or engulfing cells, from the body and are degraded inside a membrane-bound structure referred to as a “phagosome” (Fig. 1) (1). Apoptosis plays important roles in sculpting structures; maintaining homeostasis, and eliminating abnormal, nonfunctional, or harmful cells (2). Efficient removal of dying cells is the necessary last step of apoptosis; in addition, it actively prevents harmful inflammatory and autoimmune responses (1).
The nematode C. elegans, a small free-living round worm, has been established as an excellent model organism for studying the mechanisms of apoptosis and the engulfment of apoptotic cells due to its simple anatomy, known cell lineage, well-established genetics, and easily distinguishable apoptotic cell morphology (3, 4). During the development of the C. elegans hermaphrodite, 131 somatic cells and approximately 300–500 germ cells undergo apoptosis (5–7). In the soma, due to the fixed cell lineage, both the identity of the cells that undergo apoptosis and the timing of death are invariable in C. elegans (5, 6). Apoptotic cells are rapidly engulfed and degraded by neighboring cells (5–7). Multiple types of cells can function as engulfing cells, including hypodermal cells, gonadal sheath cells, intestinal cells, and pharyngeal muscle cells (6–8). One particularly useful feature of C. elegans is that animals at all developmental stages are transparent. Apoptotic cells are thus easily recognized within living animals under the Nomarski differential interference contrast (DIC) optics as highly reflective, button-like objects that are referred to as “cell corpses”(Fig. 2a) (5–7). DIC microscopy is thus commonly used to detect cell corpses in C. elegans (4). DIC microscopy, however, is unable to distinguish engulfed cell corpses from unengulfed ones because the plasma membrane of an engulfing cell is typically not visible under DIC microscope.
Besides DIC microscopy, a number of methods have been used to recognize apoptotic cells at all developmental stages in C. elegans based on their distinct cellular features. These include the transmission electron microscopy (TEM) for detecting cell corpses in larvae and adults, the TUNEL (terminal transferase dUTP nick end labeling) assay that detects DNA ends generated during apoptosis in embryos, and the staining of larvae and adults with SYTO dyes. For an excellent review of these methods, please see ref. 9. Recently, several methods have been developed to detect the exposure of phosphatidylserine (PS), a membrane phospholipids kept in the inner leaflet of the plasma membrane of living cells, on the outer surface of C. elegans apoptotic cells using PS-binding proteins, such as MFG-E8 and annexin V, as reporters (10–12). In addition, apoptotic cells undergo chromatin condensation (13). A chromatin-associated histone H3 reporter (HIS-72::GFP) (14), which allows us to detect the distinct condensed chromatin morphology in apoptotic cells in C. elegans embryos, is another cell corpse-specific marker (15).
Recently, we developed CED-1::GFP and 2xFYVE::mRFP, two fluorescent markers that label the surface of phagocytic cups and that of maturing phagosomes (Fig. 1). These two markers not only offer new methods for distinguishing cell corpses but also enable us to determine whether a cell corpse is engulfed (8, 15–17). Using these markers, we further established time-lapse recording methods to monitor the processes of engulfment as well as degradation of individual apoptotic cells in developing embryos, a procedure that enables us to dissect the steps of apoptotic cell clearance on a subcellular level (15–17).
CED-1 is a single-pass transmembrane protein expressed in engulfing cells and acts on cell surfaces as a phagocytic receptor for neighboring apoptotic cells (8). CED-1 recognizes the cell-surface features of cell corpses, clusters on the phagocytic cup and then transiently on nascent phagosomes (Figs. 3B and and4B)4B) (8, 17). This feature enables a CED-1::GFP reporter (Fig. 3A) to specifically label cell corpses that are in the process of being engulfed (Fig. 4B). In addition, CED-1::GFP is particularly useful for detecting unengulfed or partially engulfed cell corpses in engulfment-defective mutants (except the ced-7 mutants), because the blockage or delay of pseudopod extension around cell corpses do not affect the ability of CED-1 to recognize cell corpses and cluster on phagocytic cups (8, 10). As a consequence, in these mutants, CED-1::GFP is observed as bright, distinct partial circles around cell corpses, which represent not-enclosed phagocytic cups (8, 10).
In wild-type C. elegans embryos, the clustering of CED-1::GFP around a cell corpse is detectable throughout the entire engulfment process (~5 min) and the first 9 min of phagosome maturation, which lasts 50–70 min in total (17). As a result, at any given time point, only a small portion of cell corpses is labeled by CED-1::GFP in animals that display normal engulfment activity. Recently, we established 2xFYVE::GFP as a marker for cell corpses that remains on the phagosomal surface until the complete degradation of the cell corpse inside and therefore labels almost all cell corpses.
Phosphatidylinositol-3-phosphate (PI3P) is a phosphoinositide species that is specifically enriched on the surface of endosomes and phagosomes and that acts as a signaling molecule for vesicle trafficking events (18). The FYVE domain of C. elegans EEA-1, in a tandem repeat, specifically associates with PI3P (19). The 2xFYVE::GFP and 2xFYVE::mRFP fusion proteins, which are expressed in engulfing cells under the control of the ced-1 promoter Pced-1 (Fig. 3A) (8), are localized to the cytoplasmic puncta. Immediately after the closure of a phagocytic cup, these markers are enriched on the surface of nascent phagosomes and persist on phagosomal surfaces until the complete degradation of the cell corpse (17). In embryos co-expressing CED-1::GFP and 2xFYVE::mRFP, the entire cell-corpse removal process can be monitored in real time (Fig. 4B).
In addition, the 2xFYVE::GFP or 2xFYVE::mRFP reporters are excellent tools for scoring the number of cell corpses at all stages of C. elegans development. We found that in wild-type embryos, the disc-like DIC morphology of a cell corpse appears when engulfment starts, and disappears ~30 min after the initiation of cell-corpse degradation (N. L. and Z. Z., unpublished observation). Compared to the DIC morphology, the 2xFYVE::GFP signal persists on the surface of a phagosome until its complete disappearance, it is thus able to detect cell corpses that partially or totally lose their distinct DIC morphology.
The materials and methods described here are specific for the detection of apoptotic cells in C. elegans. For general materials and methods for raising and handling C. elegans, please see ref. 20. For general introduction of using DIC microscopy in C. elegans, please see ref. 21.
Cell corpses can be recognized as reflective, disc-like objects in living animals using DIC microscopy (Fig. 2a). As the execution of cell death and the clearance of apoptotic cells are dynamic processes, it is critical to score the number of cell corpses at defined developmental stages and within the defined regions of an animal for meaningful comparison of results obtained from different genetic backgrounds. To assay for the pattern of the apoptosis events of somatic cells during embryonic development, we score the number of cell corpses in the entire embryo at the following stages: bean, comma, 1.5-fold, 2-fold, 3-fold, and late 4-fold stages. Embryos at these stages, which correspond to ~320, ~380, ~420, ~460, ~520 – ~605, and ~700 – ~790 min after the first cleavage (the first cytokinesis), respectively (6, 16), are easily recognized using DIC microscopy by their distinct body morphology (Fig. 2A). Two independent sets of data, obtained from the same wild-type strain (N2) (20) by two different individuals in the laboratory, are similar to each other (Fig. 2B) (15, 16). These results are consistent with the invariable embryonic cell lineage described in (6).
To study the effect of mutants defective in cell-corpse removal, the number of persistent cell corpses is often scored in the head (the area between the anterior end of the worm and the anterior boundary of the intestine) of L1 Larvae. In the head of newly hatched wild-type L1 larvae (hatched within 1 h), no cell corpses are observed (Fig. 2B); on the other hand, mutations that induce extra cell deaths or block cell-corpse removal result in the persistent presence of cell corpses in the head (22, 23).
Germ cells that undergo apoptosis during germline development or are induced to die by DNA damaging agents can be scored in the adult hermaphrodite gonad using DIC microscopy (7, 24). Again, to obtain reproducible results, it is critical to score in animals of defined age. The most commonly used samples are adult hermaphrodites that are aged 48 h post the mid-L4 larval stage.
The DeltaVision Deconvolution Microscope is a white-light microscope that relies on specially designed computer deconvolution algorithm to achieve high resolution (25). Comparing with conventional confocal microscope, the DeltaVision results in less photobleaching of images and less photodamage to living specimens and offers comparable, under some conditions even superior, resolution and sensitivity. Here we described a specific protocol for image capture and time-lapse recording that we developed using the DeltaVision. For step-by-step operation of the DeltaVision microscope and the SoftWoRx software, see the manufacturer's instruction (http://www.appliedprecision.com).
ZH814 is an unc-76(e911) mutant strain carrying reporter constructs Pced-1ced-1::gfp and Pced-12xfyve::mrfp as well as pUNC-76(+), a plasmid containing the wild-type unc-76 gene, in the same transgenic array. Transgenic animals are normal for locomotion, whereas nontransgenic animals are Unc (Uncoordinated). To cross the transgenic array to the strains of your interest, follow standard genetic operation (20).
Among the 113 cells that undergo apoptosis during embryogenesis (6), we choose to monitor the clearance of three apoptotic cells referred to as C1, C2, and C3 (Fig. 3B). These three cells are located at the ventral surface of an embryo, in approximately the same or adjacent focal planes, and are engulfed at approximately the same time, between 320–330 min post-first cleavage (16). C1, C2, and C3 are each engulfed by a different ventral hypodermal cell, ABplaapppa, ABpraapppa, and ABplaapppp, respectively, while these hypodermal cells extend their cell bodies to the ventral midline (Fig. 3b) (16). These temporal and spatial features make it easy to identify C1, C2, and C3, and their engulfing cells; furthermore, they allow the recording of the clearance of all three cell corpses in the same time-lapse series, using a z-stack containing 8–12 serial z-sections (at 0.5 μm/section) at every time point.
The dynamic changes of the signal intensity of CED-1 and PI3P indicate the progress of engulfment and phagosome maturation; in addition, alteration of the dynamic pattern of these signals in mutant backgrounds suggests specific defects in phagosome formation and maturation (15, 17). The signal intensity of CED-1 and PI3P on phagosomal surfaces is quantified by measuring the fluorescence intensity of CED-1::GFP and 2xFYVE::mRFP, respectively. The absolute fluorescence signal intensity, however, varies from embryos to embryos due to the different expression levels of the transgene. Thus, we use the relative signal intensity represented by the ratio of the intensity on phagosomal surface to that in an adjacent area inside the cytosol to indicate the enrichment of CED-1 or PI3P on phagosomal surfaces. We use the software ImageJ to quantify fluorescence signal intensity.
During phagosome maturation, the volume of a phagosome decreases as the content is gradually digested, and is a reliable index that reflects the progression of the degradation of apoptotic cells (15, 17).
Using the protocols described above, mount embryos on slides and identify embryos at the stage of your choice. Capture serial z-section images of an entire embryo at 40 × 0.5 μm/s optic interval (see Note 7). Other parameters for image recording are the same as described above except a time course for recording is not necessary. Scoring the number of 2xFYVE::GFP(+) rings using deconvolved serial z-section images. The results represent the number of phagosomes, or engulfed cell corpses. As mentioned before (Subheading 1.2), this method is highly sensitive in identifying apoptotic cells, including those that lose their distinct DIC morphology.
Z. Z. was supported by NIH (GM067848), the Cancer Research Institute, the Rita Allen Foundation, and a Basil O’ Connor Starter Scholar award from March of Dimes Foundation. X. H. was supported by NIH (GM068676).
1NaN3 anesthetizes and immobilizes animals. Larvae and adults are immobilized within a few minutes after incubation with the 30 mM NaN3 solution. It takes NaN3 a much longer time to penetrate eggshells. For scoring embryos younger than the 2-fold stage, it is not necessary to use NaN3, since vigorous body movement of embryos does not start until that stage. Note that after 1-h incubation in the NaN3 solution, the DIC morphology of larvae and adults starts to become abnormal, whereas that of embryos are not affected.
2Anesthetization is not necessary since at the particular embryonic stages for recording, there is minimum embryonic body movement. NaN3 stops embryonic development and should be avoided.
3How to ensure that embryos develop normally during time-lapse recording? (a) To ensure normal embryonic development in a chamber with limited oxygen supply, load no more than 50 eggs onto the glass slide, disperse eggs thoroughly in the M9 solution, and carry over as little bacteria as possible. (b) To avoid photodamage of embryonic development and photobleach of fluorescence signals, use a highly sensitive CCD camera so that the light exposure time could be minimized, and restrain or avoid direct observation of fluorescent light under the eyepiece. Instead, “snapshots” with the camera should be used for finding and setting the focal plane to begin the recording. For weak fluorescence signals, use the “2 × 2 binning” function to keep the exposure time minimal. In addition, include only the necessary number of z-sections at each time point. As a rule of thumb, for recording of two channels, the exposure time of each channel should be kept below 0.2 s per z-section. (c) Signs of the photodamage of embryonic development. Data obtained from those embryos whose development is arrested due to photodamage are not useful. We rely on a few embryonic morphology changes to determine whether the development is proceeding in the normal time course. For example, the period from the bean- to the comma-stage, lasts for 60–70 min (Fig. 2A)(a, b). During this period, an embryo rotates 90° (Fig. 2A)(a, b). In addition, a period from the comma stage to the 1.5-fold stage lasts for ~40 min. A significant elongation of any of these time intervals is a sign of developmental arrest.
4Two distinct features that can help identify embryos at this stage are (1) the ventral surface slightly invaginates on both sides, and (2) the three soon-to-be engulfing cells are located at the lateral sides, in a edge shape, with the tip of each cell less than halfway extended toward the ventral midline (Fig. 4A).
5The Median and mean values are usually very similar. An abnormally bright pixel on phagosome surface, however, is largely ignored in median value, whereas it is counted and significantly increases mean value. Therefore, the median value is more resistant to signal noise and reflects the signal intensity more accurately.
6The μm/pixel scale designates the size of each pixel in the image, which can be obtained from the program with which the image is captured.
7The average thickness of an embryo is 20 μm (Z. Z., unpublished observation).