We have isolated a group of monoclonal antibodies that specifically recognize either apoptotic or engulfment cells in the interdigit areas of chicken hind limb foot plates, and throughout the embryo. Ten of these antibodies (anti-apogens) detect epitopes on dying cells that colocalize to areas of programmed cell death, characterized by the presence of apoptotic cells and bodies with typical cellular and nuclear morphology. Our results indicate that cells destined to die, or that are in the process of dying, express specific antigens that are not detectable in or on the surface of living cells. The detection of these apoptotic cell antigens in other areas of programmed cell death throughout the chick embryo indicates that different cell types, which form specific tissues and organs, may utilize similar cell death mechanisms. Six of the monoclonal antibodies (anti -engulfens) define a class of engulfment cells which contain various numbers of apoptotic cells and/or apoptotic bodies in areas of programmed cell death. The immunostaining pattern of the anti-engulfen R15F is similar to that of an antibody against a common leukocyte antigen, suggesting the participation of cells from the immune system in the removal of apoptotic cell debris. These novel monoclonal antibody markers for apoptotic and engulfment cells will provide new tools to assist the further understanding of developmental programmed cell death in vertebrates.

The general phenomenon of cell death during vertebrate devel-opment was documented in a survey performed by A. Glucksmann over 40 years ago (Glucksmann, 1950). Using various vital dyes, he and others (Saunders, 1966; Bowen, 1981) identified numerous areas in developing vertebrate embryos that contain dying cells. Additional studies, particularly in the chick nervous system, demonstrated that excess numbers of neurons, initially generated early in development, subsequently die at predicted times (Oppenheim, 1991). Some very basic biological questions about developmental programmed cell death have been raised, including: why do so many cells die during development? What determines the life or death of a cell? How are dead cells removed? We have addressed these questions by using the developing chicken foot plate as a model in which the interdigit areas invariably undergo programmed cell death at predicted times (stage 30-32) (for reviews see Hinchliffe, 1981; Saunders and Fallon, 1967; Hammar and Mottet, 1971). When interdigit cells die during limb development, they exhibit the hallmark features of apoptotic cells (Kerr, 1971; Kerr et al., 1972), and these dead cells are rapidly engulfed by resident neighboring cells and/or invading immune cells (Fallon and Saunders, 1968; Hurle and Hinchliffe, 1978). Many apoptotic nuclei can be observed within an engulfing cell that has taken up one or many dying or dead cells (Dawd and Hinchcliffe, 1971). The nuclei of apoptotic cells undergo the most drastic changes, which include margination and condensation of chromatin and nuclear fragmentation, in all areas where cell death is occurring. These nuclear changes have been observed particularly in the foot plate but also in many other areas of cell death in the chicken embryo, such as the mesonephros (Salzbeger and Webver, 1965), developing brain and peripheral nervous system (for recent reviews see Oppenheim, 1991; Clarke, 1990), the aortic arches and bulbus cordis in the heart (Pexieder, 1975; Manasek 1969). In addition to the nuclear changes, vacuole formation in the cytoplasm, and cellular fragmentation into pieces, which may or may not contain nuclear DNA, are two other consistent features of chick embryonic cell death (Saunders and Fallon, 1966; Dawd and Hinchcliffe, 1971, Hinchcliffe and Ede, 1973). The morphological changes of apoptotic cells as well as the engulfment process appear to be conserved among species, including the nematode Caenorhabditris elegans (Robertson and Thomson, 1982).

A genetic pathway for programmed cell death in vertebrates has not been nearly as well established as the one described in C. elegans (Ellis et al., 1991). Eleven genes have been identified in C. elegans that intiate and execute cell death autonomously (Hengartner et al., 1991; Yuan et al., 1993), direct engulfment, remove dead cells (Hedgecock et al., 1983; Ellis et al., 1991) or degrade dead cell DNA (Sulston et al., 1980). Two vertebrate gene families have been identified that may function in developmental programmed cell death in mammals. The bcl-2 family is emerging with members that cooperate or antagonize the cell death-suppressing activity of the orginal bcl-2 protein (Boise et al., 1993; Oltvai et al., 1993). The mammalian interleukin-1β converting enzyme (Thornberry et al., 1992), which has been shown to participate in the initiation of cell death in fibroblasts in vitro (Miura et al., 1993) may be a funtional vertebrate homolog of ced-3 protein (Yuan et al., 1993).

Many components of the pathway leading to programmed cell death in vertebrates remain to be identified and questions about mechanisms of initiation and execution, cell and tissue homeostasis, recognition and removal of dead cells remain unanswered. We approached the problem by generating monoclonal antibodies that recognized antigens specifically expressed in the interdigit areas of the chick foot plates during the period of programmed cell death (Hurle and Colvee, 1982). These markers identify cells dying by apoptosis, and will potentially help to define individual steps in the process. In this report, we describe a total of 16 monoclonal antibodies that define 2 categories of cell death antigens specified by different immunostaining patterns at the light microscope level, that are either associated with dying cells or with cells that are in the process of phagocytosing cell debris. Our studies indicate that many of the cell deaths occurring during embryonic development possibly share a common mechanism.

Chicken egg incubation

Virus-free standard fertilized eggs from (Spafas Inc., Norwich, Conn.) were kept at 10-11°C, in a mini wine cellar (T.W.E. Wholesales Inc., Pleasantville, NY), for no more than 7 days, or incubated at 38-39°C and 85-87% humidity in a egg incubator (Kuhl Corp., Flemington, New Jersey) which was equipped with rotating trays. The chicken embryos were staged according to the Hamburger and Hamilton (1951) tables.

Day 7 and 10 antigen preparation

Chick foot plates, stage 31-32 (day 7-7.5), consisting of digits and interdigit tissue, and digits only from stage 36 embryos (day 10) were dissected from respective hindlimbs and washed in ice-cold PBS. Tissues were homogenized individually with a glass:glass tissue grinder (Fisher Scientific, USA) in freshly prepared 4% buffered paraformaldehyde (pH 7.4) for fusion A and B, and for the suppression injections in protocol W. The antigen (digit and interdigit tissue) for immunization injections in protocol W, was prepared in a protease inhibitor cocktail made of PBS without magnesium and calcium salts, consisting of 1 mM PMSF, 25 mM EDTA, 1 μg/ml pepstatin, 20 μg/ml each of aprotinin, leupeptin and bestatin (Boerhringer Mannheim, Indianapolis, IN). Each tissue homogenate was sonicated at 0°C with five 15 second pulses at 30 cycles per second. The homogenate was centrifuged at 4°C, 45,000 rpm for 30 minutes and washed and sonicated in PBS. The final pellet was resuspended in PBS and mixed 1:1 with complete Freunds adjuvant (Organon Teknika, Cappel, Durham, NC) in a glass:glass syringe (Popper and Sons Inc., New Hyde Park, NY). Protein concentration was determined according to Bradford (1976).

Neonatal tolerization and immunization

Methods for neonatal tolerization and immunization were performed as described by Hockfield (1987). For fusion A and B, a total of three 1-day-old Balb/c mice were tolerized by intraperitoneal injections with unwanted antigen, day 10 digits, approximately 4.5-9.0 mg tissue wet weight-PBS/mouse, 50-60 μl injection volume. A total of seven injections for immune tolerization were performed every other day. Starting at day 25 after birth, mice for fusion A were immunized intraperitoneally with 200 μg protein, and for fusion B, mice were injected intradermally with 25-30 μl of a solution containing 50-100 μg of day-7 foot plate antigen. In each of the aforementioned immunizations, which were repeated four more times every fourth day, antigen was mixed thoroughly 1:1 with Freunds complete adjuvant (Cappel).

Cyclophosphamide immunosuppression

For fusion W, two 5-week old female Balb/cJ mice (Jackson Labs, Bar Harbor, ME) received intraperitoneal injections of 500 μg of paraformaldehyde-fixed day-10 chick hindlimb digit, prepared as described. For each injection, the antigen was mixed 1:1 with Ribi adjuvant MPL and TDM emulsion, (Ribi Immunochemical Research Inc., Hamilton, MT) to a total volume of 400 μl. Immediately after, each mouse received an intraperitoneal injection of cyclophosphamide, 110 mg/kg (Sigma Chemical Co., St. Louis, MO). Cyclophosphamide was administered intraperitoneally at 24 and 48 hours later. After 2 weeks, the same sequence of antigen and cyclophosphamide administration was repeated. At 2 weeks, mice were immunized intraperitoneally with 300 μg of day-7 digit and interdigit tissue, mixed 1:1 with Ribi adjuvant and PBS. This immunization was repeated 2 weeks later. Mouse sera from tail bleeds were tested for selective suppression with indirect immunoflourescence. A final boost was given intraperitoneally 2 weeks later with day-7 antigen solubilized only in PBS.

Media and additives

The medium for NS-1 myeloma and hybridoma cells was D-MEM with 4.5 g/l glucose, 15 mM Hepes, supplemented with nonessential amino acids, 110 mg/ml sodium pyruvate, penicillin, streptomycin and neomycin (Gibco Life Technologies Inc., Grand Island, NY). NS-1 cells were cultivated for 2 weeks prior to the fusion in D-MEM, 5% characterized fetal calf serum (Hyclone, Logan, Utah). Hybrids were plated in D-MEM, 20% fetal calf serum, OPI supplement (Sigma) and HT and aminopterin (Gibco). Media for limiting dilutions consisted of the above, except fetal calf serum was reduced to 15% and Balb/c mouse (3- to 4-weeks old) thymocyte-conditioned medium was added 1:1 with the basic medium, and aminopterin was removed.

Fusions A, B and W

For fusions A and B, day 41 after birth, inguinal and popliteal lymph nodes (source of antibodies designated R) as well as spleen (source of antibodies designated A and B) were dissected and sieved separately through a nylon screen (Falcon) with a blunt-ended glass syringe plunger. With respect to fusion W, 3 days after the final immunization only the spleen was taken from these animals and processed. Red blood cells from the spleen only were lysed with a red blood cell lysis buffer (Sigma). The ratio of myeloma cells to spleen cells was 1:5. 1 ml of polyethylene glycol 4000 (Gibco) was added over 30 seconds while mixing, and diluted to 12 ml with serum-free DMEM over 4 minutes (1 ml first minute, 2 ml the second, 3 ml the third and 6 ml the 4th). Fused cells were centrifuged 800 g, for 4 minutes and 40 ml selection medium was added without disturbing the pellet and allowed to sit 12 minutes. At this time, the cell pellet was resuspended and diluted with enough medium to plate in a total of fifteen 96 well plates, 150-200 μl/well. At approximately 10- to 17-days post-fusion, wells with hybridoma growth were identified and screened on previously prepared frozen chick foot plate sections.

Indirect immunofluorescence screening

Hybridoma supernatants were tested on stage 31-32 foot plate tissue sections. Foot plates as well as embryos were fixed in fresh 4% paraformaldehyde for 2 hours at room temperature and subsequently cryoprotected in 30% sucrose-PBS for 24-48 hours at 4° C. Tissue was then embedded in OCT compound (Miles Inc., Diagnostic Division Elkart, IN) in bottle-neck embedding capsules (Polysciences Inc., Warrington, PA) and quick-frozen in dry ice and isopentane. Frozen sections (8-10 μm) of chick foot plates or embryos were collected, 4 per slide, on superfrost plus slides (Fisher). Wells were made around tissue sections with a PAP pen (Daido Sangyo Co., LTD. Japan) and nonspecific sites on tissue sections were blocked with 1% BSA-PBS and 10% goat serum for 30 minutes to 1 hour at room temperature in a humidified chamber for R antibodies and 1-2 hours for A, B and W antibodies. Apoptotic morphology in the interdigit and embryo was identified by including a DNA dye, Hoechst 33258 (Sigma) at 0.1 μg/ml in the blocking buffer, or propidium iodide at 4 μg/ml. 75-125 μl hybridoma supernatants were added and incubated overnight at 4°C in a humidified chamber for R antibodies, while A, B and W supernatants were incubated at room temperature for 4-6 hours. The samples were washed twice with 1% BSA-PBS for 3 minutes at room temperature. Fluorescein-conjugated secondary antibodies, goat antimouse IgGAM (Cappel) were added at 1:200 in 1% BSA-PBS for 1 hour at room temperature. Sections were washed twice with 1% BSA-PBS for 3 minutes and slides were coverslipped using 90% glycerol, 10% PBS and 1mg/ml o-phenylenediamine (Sigma). Antibodies recognizing dying cells were identified with a Zeiss Epifluorescence Axioplan microscope using the 40× and 100× oil immersion phasecontrast objectives. In Fig. 3, the cells indirectly labelled with FITC-conjugated goat anti-mouse IgM (μ chain specific) and Texas red-conjugated goat anti-mouse IgG specific antibody, were imaged with a Biorad MRC-600 laser-scanning confocal imaging system and Bio-Rad COMOS software. Images were collected with Kalman averaging and intensified with contrast enhancement and/or edge sharpening. Tissue sections were stained with propidium iodide when desired, as described by Rodriguez-Tarduchy et al. (1990). Briefly, sections were incubated in 20 mM Hepes, pH 7.0, 100 μg/ml DNase free RNase and 4 μg/ml propidium iodide, for 30 minutes at 37°C and then processed for immunostaining. The isotypes of the isolated antibodies were detected using a rat anti-mouse immunoglobulin kit (Gibco).

Immunostaining of interdigit apoptotic and engulfment cells by two classes of monoclonal antibodies

Our objective was to isolate monoclonal antibodies that recognize specific antigens associated with dying cells in the interdigit areas of stages 31-32 (day 7-7.5) chick foot plates. To this end, we took advantage of two protocols that enhance the mouse immune response to desired antigens (Hockfield, 1987; Matthew and Patterson, 1983). The protocol developed by Hockfield involved neonatal tolerization with stage 35-36 (day 10-11) digits where interdigit cell death has concluded, and immunization with stage 31-32 (day 7-7.5) interdigit dying tissue. The second method developed by Matthew uses immunization and chemical immunosuppression by cyclophosphamide to suppress immune responses to undesired antigens. Antibodies were isolated by immunostaining with hybridoma supernatants on frozen sections of day 7-7.5 chick footplates. The Hoechst DNA dye 33258, as well as propidium iodide was employed as a double stain, which labelled the characteristic apoptotic nuclear and cellular morphologies respectively. From a total of 6,048 hybridomas screened, 1,238 positive clones staining different anatomical areas of the foot plate were initially identified; from this group, 16 cell death-specific hybridomas were cloned.

These antibodies were classified into two categories according to the staining patterns observed by fluorescence microscopy at 40× and 100× magnifications (See Tables 1 and 2 for summaries). Fig. 1A-F shows stainings of selected cell death antibodies from the two classes. The same sections were costained with the Hoechst (Fig. 1A,D) and propidium iodide (Fig. 1B,E) fluorescent dyes to illustrate the corresponding apoptotic cell and nuclear changes. The most common class of antibodies, class I, identified apoptotic cells and apoptotic cellular fragments (bodies) that are in close association with condensed and fragmented nuclei, which stain brightly with Hoechst dye (Fig. 1A). Unidentified epitopes, on dying or dead cells, recognized by this class of antibodies were named ‘apogens’. Propidium iodide staining revealed that small to large apoptotic cell DNA aggregates were more brightly stained than the noncondensed DNA of healthy cells (Fig. 1B). Class I antibodies appear to recognize condensed and nonuni-form epitopes associated with apoptotic cells and bodies. Class I immunoreactivity can be observed in and/or on individual closely associated apoptotic cells and/or bodies, but was never observed on cells with apparently normal nuclei. In addition, immunoreactivity is not detectable in areas of the limb where programmed cell death is absent, e.g. the digit proper.

Table 1.

Results from the mouse monoclonal antibody fusions A, B and W, and areas of immunoreactivity in the hindlimb foot plate

Results from the mouse monoclonal antibody fusions A, B and W, and areas of immunoreactivity in the hindlimb foot plate
Results from the mouse monoclonal antibody fusions A, B and W, and areas of immunoreactivity in the hindlimb foot plate
Table 2.

Monoclonal antibodies with immunoreactivity for apoptotic cells and bodies and engulfment cells during interdigit programmed cell death

Monoclonal antibodies with immunoreactivity for apoptotic cells and bodies and engulfment cells during interdigit programmed cell death
Monoclonal antibodies with immunoreactivity for apoptotic cells and bodies and engulfment cells during interdigit programmed cell death
Fig. 1.

Apoptotic cell death antibody (anti-apogen) and engulfment (anti-engulfen) immunostaining in interdigit spaces of stage 31 chick foot plates. Frozen sections stained with Hoechst dye (A,D) demonstrating apoptotic nuclear changes (solid arrow). The condensed chromatin of apoptotic cells is more brightly stained with this dye. These sections were also stained with propidium iodide (B,E) which labels more of the apoptotic cell and body features (corresponding arrow). The same sections immunostained with class I, R12 (C) and class II R15F (F) antibodies. The arrows in C and F identify the cells stained in A,B and D,E, respectively. Scale bar, 25 μm.

Fig. 1.

Apoptotic cell death antibody (anti-apogen) and engulfment (anti-engulfen) immunostaining in interdigit spaces of stage 31 chick foot plates. Frozen sections stained with Hoechst dye (A,D) demonstrating apoptotic nuclear changes (solid arrow). The condensed chromatin of apoptotic cells is more brightly stained with this dye. These sections were also stained with propidium iodide (B,E) which labels more of the apoptotic cell and body features (corresponding arrow). The same sections immunostained with class I, R12 (C) and class II R15F (F) antibodies. The arrows in C and F identify the cells stained in A,B and D,E, respectively. Scale bar, 25 μm.

Class II immunostaining patterns when observed in conjunction with Hoechst and propidium dyes may be present intracellularly or on membranes of phagocytosing cells that are surrounding condensed apoptotic cells and/or bodies (Fig. 1D-F). When class II immunostaining is present amongst condensed DNA fragments of apoptotic cells, these fragments usually number 3-20. These apoptotic nuclear and cellular features observed in conjunc-tion with class II immunostaining in the foot plate, are similar to the morphological descriptions of phagocytes believed to be present initially in foot plate cell death areas (interdigit, PNZ, ANZ) (Hinchliffe and Ede, 1973). In support of these early phagocyte descriptions, class II immunoreactivity is present on cells in other areas of cell death in the embryo. The antibody R15F immunostains the plasma membrane of an engulfment cell, that surrounds a larger number of condensed apoptotic nuclei and/or nuclear fragments (Fig. 1D-F). Numerous condensed apoptotic nuclei and/or fragments are almost always observed within the distinctly stained plasma membrane of R15F-positive cells in the interdigit spaces as well as throughout the embryo (not shown). The antibodies in class II are alike in that they are immunoreactive also for apparently normal looking cells, and are present in areas without noticeable cell death. Intracellular immunostaining is predominantly observed with class II anti-bodies R3E, A7 and R3D, (not shown). One additional feature of class II anti-bodies, revealed by Hoechst and propidium iodide staining, is that these cells contain an apparently normal non-condensed nucleus (engulfing cell nucleus) among apoptotic nuclei and/or fragments thereof (Fig. 1D,E). Cytological descriptions of many dead cells within a ‘phagocytic cell’ have been reported in the opaque patch of early limb development in chick (Dawd and Hinchliffe, 1971; Bellairs, 1961). Class I and II immunoreactivities are detectable in the primitive limb bud at days 3-4 (stage 24) (not shown), when programmed cell death occurs in the anterior and posterior ‘necrotic’ zones (Saunders and Fallon, 1967). We named the unidentified epitopes recognized by class II antibodies as ‘engulfens’. Further proof of our hypothesis is shown in (Fig. 2A-E) where class I antibody R12 and class II antibody R15A, clearly reveal programmed cell death areas in the chick foot plate at stage 32 (day 7.5). A low power micrograph of class I staining (Fig. 2C), is shown relative to the Hoechst and propidium iodide-stained digit and interdigit cellular patterns (Fig. 2A,B). Peroxidase labelled secondary immunoreactivity against R15A can been seen in each of the three interdigit spaces as well as along the borders of the foot plate (Fig. 2D). The association of several condensed darkly counter-stained apoptotic nuclei with class II staining at higher magnification (Fig. 2E) indicates that class II antibodies recognize an engulfment cell.

Fig. 2.

Class I and x cell death in stage 32 chick foot plates. Hoechst (A) and propidium iodide (B) nuclear staining patterns of a frozen section are shown relative to R12 (class I) antibody staining in the third interdigit space and along the posterior foot plate border (C). d, digits; scale bar, 35 μm (C). (D) Low magnification of R15A (class II) immunolabelled cells in the three interdigit spaces and foot plate borders, scale bar, 50 μm. (E) High magnification of immunolabelled engulfment cells with anti-engulfen R15A; scale bar, 20 μm. The paraffin section (D,E) is counterstained with Mayer’s hematoxylin, which illustrates the numerous dark and condensed nuclei associated with the engulfment immunostaining.

Fig. 2.

Class I and x cell death in stage 32 chick foot plates. Hoechst (A) and propidium iodide (B) nuclear staining patterns of a frozen section are shown relative to R12 (class I) antibody staining in the third interdigit space and along the posterior foot plate border (C). d, digits; scale bar, 35 μm (C). (D) Low magnification of R15A (class II) immunolabelled cells in the three interdigit spaces and foot plate borders, scale bar, 50 μm. (E) High magnification of immunolabelled engulfment cells with anti-engulfen R15A; scale bar, 20 μm. The paraffin section (D,E) is counterstained with Mayer’s hematoxylin, which illustrates the numerous dark and condensed nuclei associated with the engulfment immunostaining.

Fig. 3.

Localization of an apoptotic cell within an engulfment cell in the developing nephrogenic mesoderm. Confocal fluorescence micrographs of a frozen section double stained with an apoptotic specific class I antibody, R6 (IgM), detected with FITC-goat anti-mouse IgMμ-specific secondary antibody (A), and an engulfment class II antibody, R15F (IgG2b), detected by Texas red rabbit anti-mouse IgG (heavy chain specific) antibody (B). (C) Merged confocal fluorescence micrographs A and B, which demonstrates an immunostained apoptotic cell within an engulfment cell (open arrow), and an apoptotic cell (solid arrow) immunostained but not engulfed .

Fig. 3.

Localization of an apoptotic cell within an engulfment cell in the developing nephrogenic mesoderm. Confocal fluorescence micrographs of a frozen section double stained with an apoptotic specific class I antibody, R6 (IgM), detected with FITC-goat anti-mouse IgMμ-specific secondary antibody (A), and an engulfment class II antibody, R15F (IgG2b), detected by Texas red rabbit anti-mouse IgG (heavy chain specific) antibody (B). (C) Merged confocal fluorescence micrographs A and B, which demonstrates an immunostained apoptotic cell within an engulfment cell (open arrow), and an apoptotic cell (solid arrow) immunostained but not engulfed .

We next examined double staining with appropriate isotype primary antibodies from class I (R6) and II (R15F), along with specific secondary labelled antibodies. As predicted, class I immunoreactive cells (apoptotic cells) could be detected within the cytoplasm of class II-positive cells, (Fig. 3A-C). These results demonstrate that antigens recognized by class I anti-bodies are expressed by dying cells and class II antigens independently by engulfing cells (Fig. 3C). More importantly, class I immunopositive cells can be observed individually without a surrounding engulfment cell (Fig. 3A) suggesting that the antigen recognized by R6 is probably not induced by engulfment cells.

Immunostaining of a common leukocyte antigen in interdigit spaces

Monoclonal antibodies have been isolated that recognize various antigens on cells of the chicken immune system; on in particular reacts with a common leukocyte antigen, HIS-C7, another with mature macrophages, ChNL-68.1 (Janse and Jeurissen, 1991). We suspected that cells of the immune system, e.g., cells of leukocyte and/or macrophage origin may play a role in engulfment and apoptotic cell removal. We evaluated the immunoreactivity of the two antibodies, HIS-C7 and ChNL-68.1, in foot plate sections in relation to apoptosis and to the immunostaining observed with our antibodies. One class II antibody, R15F, closely resembles the membranestaining detected with the common leukocyte antibody HIS-C7 (Janse and Jeurissen, 1991) (Fig. 4B,D). Similar to R15F, HIS-C7 stains the outer membranes of cells which contain numerous condensed nuclei within their cytoplasm. This immunoreactivity is also present in other areas of cell death in the chick embryo. Similar to class II antibodies, HIS-C7 is also detected on certain living cells and in areas without associated apoptotic cellular changes. Thus, it is possible that R15F and HIS-C7 immunostain the same type of engulfment cells. Immunostaining with a monoclonal antibody marker of chicken mononuclear phagocytes ChNL-68.1 (Jeurissen et al., 1988) was absent during interdigit cell death, as well as in other cell death areas of chick embryos. During the period of inter-digit cell death, days 6.5-8.5, chicken embryos do not appear to have mature T and B cells (Chen, H. and Cooper, M. D., 1987; Chen et al., 1986). ChNL-68.1 as well as class II staining was detectable in the bursa of Fabricius in newly hatched chickens that possessed numerous mature macrophages and exhibited cell death (not shown). The described leukocyte immunostaining results suggest that cells of the immune system are involved in engulfing apoptotic cells and bodies in the interdigit and in other areas of the developing embryo.

Fig. 4.

Comparison of class II engulfment immunostaining with chicken common leukocyte immunoreactivty in the interdigit of stage 31 chick foot plates. Hoechst stained foot plate (A) and class II R15F immunostaining (B). A different foot plate section representing a similar interdigit area to that in A is shown in C stained with Hoechst dye and in D, one stained with a common leukocyte marker HIS-C7. R15F and HIS-C7 are immunoreactive for the surface of cells that contain condensed and fragmented nuclei of apoptotic cells. The arrows in A-D mark Hoechst-stained and corresponding immunostained cells. Scale bar, 10 μm.

Fig. 4.

Comparison of class II engulfment immunostaining with chicken common leukocyte immunoreactivty in the interdigit of stage 31 chick foot plates. Hoechst stained foot plate (A) and class II R15F immunostaining (B). A different foot plate section representing a similar interdigit area to that in A is shown in C stained with Hoechst dye and in D, one stained with a common leukocyte marker HIS-C7. R15F and HIS-C7 are immunoreactive for the surface of cells that contain condensed and fragmented nuclei of apoptotic cells. The arrows in A-D mark Hoechst-stained and corresponding immunostained cells. Scale bar, 10 μm.

Immunostaining of apoptotic cellular changes in embryonic tissues

To determine if the anti-apogens and anti-engulfens described in the interdigit are also expressed in other areas of programmed cell death in the embryo, we examined all class I antibodies for immunoreactivity with apoptotic cells in stage 18-27 chick embryos. Cell death in chick embryos between stages 18-27 (days 3-5) were evaluated in detail with antibody W4. This antibody detected apoptotic changes in association with typical apoptotic nuclear morphology as revealed by Hoechst staining in almost all tissues of the developing embryo that undergo programmed cell death. W4 stained apoptotic cells in the posterior and ventral part of the mesonephros (Fig. 5A,B) and in the lateral boundary of the dermomyotome (Fig. 5 C,D), in the liver and mandibular arches (not shown). In the cardiovascular system, W4 stained the walls of the sinus venosus, aortic arches (Fig. 5 E,F), mesenteric vessels, anterior entrance of the atrium, subvalvular apparatus of the atrio-venticular valva, bulbus cordis and dorsal aorta (not shown). In neuroectoderm-derived tissues, W4 immunoreactivity was intense in the optic fissure formed by the ventral invagination of the optic cups and stalks. This area of cell death allows optic nerve fibers from the retina to reach their proper postsynaptic target within the diencephalon and also allows the blood vessels to enter the optic cup. W4 stained the optic stalk as well. At day 5, the staining was intense in the posterior part of the mesencephalon isthmus (Fig. 5G,H), and in the posterior mesenchyme at the level of the diencephalon.

Fig. 5.

Fluorescence micrographs showing Hoechst labelling (A,C,E,G) and immunostaining (B,D,F,H) of apoptotic cells in various tissues of stage 25 chick embryos with class I antibody W4. (A) parasagittal section, mid trunk level, showing Hoechst staining of nuclei in the developing kidney. Lumen of one mesonephric tubule (t). (B) Intense immunoreactivity in the superior mesoderm surrounding the mesonephric tubules. (C) Parasagittal section showing Hoechst staining, and (D) immunostaining of apoptotic cells at the dermomyotome boundaries. (E,F) Parasagittal sections showing numerous positive cells surrounding the aortic arch vessel. aa, lumen of aortic arch. (G) Hoechst staining and (H) immunoreactivity on apoptotic cells in the posterior part of the mesencephalon isthmus (arrow). In each photograph, dorsal is at the top and ventral is bottom. Scale bar, 25 μm.

Fig. 5.

Fluorescence micrographs showing Hoechst labelling (A,C,E,G) and immunostaining (B,D,F,H) of apoptotic cells in various tissues of stage 25 chick embryos with class I antibody W4. (A) parasagittal section, mid trunk level, showing Hoechst staining of nuclei in the developing kidney. Lumen of one mesonephric tubule (t). (B) Intense immunoreactivity in the superior mesoderm surrounding the mesonephric tubules. (C) Parasagittal section showing Hoechst staining, and (D) immunostaining of apoptotic cells at the dermomyotome boundaries. (E,F) Parasagittal sections showing numerous positive cells surrounding the aortic arch vessel. aa, lumen of aortic arch. (G) Hoechst staining and (H) immunoreactivity on apoptotic cells in the posterior part of the mesencephalon isthmus (arrow). In each photograph, dorsal is at the top and ventral is bottom. Scale bar, 25 μm.

Using immunosuppression and neonatal tolerization protocols, we have isolated 16 monoclonal antibodies that specifically recognize apoptotic or engulfment cells during chicken embryonic development. These antibodies can be classified into 2 categories. The class I antibodies recognize antigens unique to apoptotic cells in the developing chick foot plate as well as elsewhere in the chicken embryos. The immunostaining colocalizes to nuclear changes in dying or dead cells as revealed by propidium iodide and Hoechst dye staining. Immature or mature chicken erythrocytes (permanently nucleated cells) which may possess nuclear features resembling apoptotic cells were not stained by these antibodies (data not shown). The cell death antigens recognized by class I antibodies are consistently detected at the predicted times of interdigital cell death, as well as in dying cells in the anterior and posterior ‘necrotic zones’ and the apical ectodermal ridge. In these areas virtually all of the condensed apoptotic nuclei detected by Hoechst dye are observed in conjunction with the anti-apogen immunostaining, suggesting that these cell deaths share a common program. To examine if class I immunostaining is a secondary event induced by engulfment, double labelling experiments with appropriate isotype controlled class II antibodies revealed that class I immunostaining can be clearly observed outside engulfing cells. Thus, it appears that at least one apogen (epitope) detected by class I antibody R6, is not induced by the presence and/or contact of an engulfment cell immunopositive for an engulfen epitope.

We were interested to see if our class I antibodies recognized the death of motorneurons or sensory neurons that were described by Oppenheim and others (Oppenheim, 1991; Okado and Oppenheim 1984). We observed immunoreactivity in neuronal tissues such as the dorsal and ventral horns and spinal ganglia during the period of extensive cell death with each class of antibody, but the numbers of cells immunostained were far less than the quantitative data reported by Chu-Wang and Oppenheim (1978a,b). The staining we have observed with class I antibodies in neuronal tissue may correspond to the sporadic cell deaths described by Pittman and Oppenheim (1979) that regularly occur during the neuronal development. Alternatively, dying motorneurons may be rapidly engulfed and cleared, because it is difficult to find pyknotic figures in these areas, or the differences may be attributed to the techniques used to evaluate cell death, i.e., immunocytochemistry versus histological staining.

What is the nature of the antigens recognized by class I antibodies? Since the apoptotic cells are almost always condensed and fragmented, it is difficult to determine the location of the antigen at the light microscopic level. Antibody reactivity will need to be evaluated at the electron microscopy level, or subcellular fractionation analyses will be needed to determine the exact location of the apogen in the dying cells. Our hypothesis is that some of the antigens recognized by these antibodies are molecules present on a dead cell, which allows it to be recognized by the appropriate engulfment cells (Morris et al., 1984; Duvall et al.,1985). In the nematode C. elegans, seven genes, ced-1, ced-2, ced-5, ced-6, ced-7, ced-8 and ced-10, have been identified as responsible for efficient removal of dead cells (Ellis et al., 1991; Hedgecock et al., 1983). It is possible that our antibodies recognize the components of a similar engulfment system in vertebrates (Lang and Bishop 1993). Presently it is not known whether class I and II antibodies crossreact with dying or engulfing cells in C. elegans. Having developed these monoclonal antibodies, we are in a position to isolate and characterize these antigens, and eventually clone the genes coding for these molecules. From our panel of antibodies, 6 distinct proteins are detected on immunoblots, which suggests that different epitopes are recognized on different molecules, rather than similar epitopes on different molecules (data not shown).

Dawd and Hinchliffe (1971) described several classes of engulfment cells at the electron microscope level, according to their sizes, the approximate number of dead cells they contain and the level of acid phosphatase. Strikingly, our class II antibodies almost always stain cells in cell death areas that contain the numerous fragmented and apoptotic nuclei (brightly stained propidium iodide positive material, Fig. 1E). In these instances virtually all of the apoptotic cells are found within the engulfing cells. This suggests that engulfment occurs very quickly after cell death, during chick embryonic development, which is very similar to cell death and engulfment in C. elegans (Robertson and Thomson, 1982; Sulston and Horvitz, 1977). In summary, the class I antibodies that we have developed are the first specific markers of programmed cell death that have been identified in vertebrates. These antibodies will be valuable tools for us to analyze the programmed cell death in detail during vertebrate embryonic development. We intend to use these antibodies to isolate the cell death-specific antigens, and to elucidate the role of the corresponding genes during programmed cell death. Such studies will provide us with the tools to identify the genetic pathways of programmed cell death in vertebrates.

R. R. and P.-A. F. contributed equally to this work. We thank S. Hockfield for NS-1 cells and for comments on monoclonal antibody production. We thank Kurt Christianson, Jane Dodd, William Matthew and Ed Harlow for helpful suggestions on hybridomas. We are grateful to Susan Jeurissen for leukocyte and macrophage antibodies. We thank Johannes Drexler, Masayuki Miura and Louise Bergeron for constructive suggestions throughout the course of this work and critical review of the manuscript. This work was supported in part by grants from National Institute of Aging (AG11017) and from Bristol-Myers/Squibb to J. Y.

Bellairs
,
R.
(
1961
).
Cell death in chick embryos as studied by electron microscopy
,
J. Anat. (Lond)
.
95
,
54
60
.
Bradford
,
M. M.
(
1976
).
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding
.
Analyt. Biochem
.
72
,
248
254
.
Boise
,
L. H.
,
Gonzalez-Garcia
,
M.
,
Postema
,
C. E.
,
Ding
,
L.
,
Lindsten
,
T.
,
Turka
,
L. A.
,
Mao
,
X.
,
Nunez
,
G.
and
Thompson
G. B.
(
1993
).
bcl-x, a bcl-2-related gene that functions as a dominaant regulator of apoptotic cell death
.
Cell
74
,
597
608
.
Bowen
,
I. D.
(
1981
).
Techniques for demonstrating cell death. In: Cell Death in Biology and Pathology
. (eds
Bowen
I. D.
and
Lockshin
R. A.
), pp.
379
444
.
London
:
Chapman & Hall
.
Chen
,
C. H.
,
Ager
,
L. L.
,
Gartland
G. L.
and
Cooper
,
M. D.
(
1986
).
Identification of a T3/T cell receptor complex in chickens
.
J. Exp. Med
.
164
,
375
380
.
Chu-Wang
,
I.
and
Oppenheim
,
R. W.
(
1978a
).
Cell death of motoneurons in the chick embryo spinal cord. I. A light and electron microscopic study of naturally-occurring and induced cell loss during development
.
J. Comp. Neurol
.
177
,
33
58
.
Chu-Wang
,
I.
and
Oppenheim
,
R. W.
(
1978b
).
Cell death of motorneurons in the chick embryo spinal cord. II. A quantitative and qualitative analysis of degeneration in the ventral root, including evidence for axon outgrowth and limb innervation prior to cell death
.
J. Comp. Neur
.
177
,
59
86
.
Clarke
,
P. G. H.
(
1990
).
Developmental cell death: morphological diversity and multiple mechanisms
.
Anat. Embryol
.
181
,
195
213
.
Dawd
,
D. S.
and
Hinchliffe
,
J. R.
(
1971
).
Cell death in the ‘opaque patch’ in the central mesenchyme of the developing chick limb: a cytological, cytochemical and electron microscope analysis
.
J. Embryol. Exp. Morphol
.
26
,
401
424
.
Duvall
,
E.
,
Wyllie
,
A. H.
and
Morris
,
R. G.
(
1985
).
Macrophage recognition of cell undergoing programmed cell death (apoptosis)
.
Immunology
56
,
351
358
.
Ellis
,
R. E.
,
Jacobson
,
D.
and
Horvitz
,
H. R.
(
1991
).
Genes required for the engulfment of cell corpses during programmed cell death in C. elegans Genetics
112
,
591
603
.
Fallon
,
J. F.
and
Saunders
,
J. W.
, Jr
. (
1968
).
In vitro analysis of the control of cell death in a zone of prospective necrosis from the chick wing bud
.
Dev. Biol
.
18
,
553
570
.
Hamburger
,
V.
(
1975
).
Cell death in the development of the lateral motor column of the chick embryo
.
J. Comp. Neurol
.
160
,
535
546
.
Hamburger
,
V.
and
Hamilton
,
H. L.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morphol
.
88
,
49
92
.
Hammar
,
S. P.
and
Mottet
,
N. K.
(
1971
).
Tetrazolium salt and electron microscopic studies of cellular degeneration and necrosis in the interdigital areas of the developing chick limb
.
J. Cell Sci
.
8
,
229
251
.
Hedgecock
,
E. M.
Sulston
,
J. E.
and
Thomson
,
J. N.
(
1983
).
Mutations affecting programmed cell death in the nematode Caenorhabditis elegans
,
Science
220
,
1277
1279
.
Hengartner
,
M.
,
Ellis
,
R. E.
and
Horvitz
,
H. R.
(
1992
)
Caenorhabditis elegans gene ced-9 protects cells from programmed cell death
.
Nature
356
,
494
499
.
Hinchliffe
,
J. R.
and
Ede
,
D. A.
(
1973
).
Cell death and the development of limb form and skeletal pattern in normal and wingless (ws) chick embryos
.
J. Embryol. Exp. Morph
.
30
,
753
772
.
Hinchliffe
,
J. R.
(
1981
).
Cell death in embryogenesis
. In
Cell Death in Biology and Pathology
(eds
Bowen
I. D.
and
Lockshin
R. A.
) pp.
35
69
.
London
:
Chapman and Hall
.
Hockfield
,
S.
(
1987
).
A mAb to a unique cerebellar neuron generated by immunosuppression and rapid immunization
.
Science
237
,
67
70
.
Hurle
,
J.
and
Hinchliffe
,
J. R.
(
1978
).
Cell death in the posterior nectrotic zone (PNZ) of the chick wing-bud: a stereoscan and ultrastructural survey of autolysis and cell fragmentation
.
J. Embryol. Exp. Morph
.
43
,
123
136
.
Hurle
,
J. M.
and
Colvee
,
E.
(
1982
).
Surface changes in the embryonic interdigital epithelium during the formation of the free digits: a comparative study in the chick and duck foot
.
J. Embryol. Exp. Morph
.
69
,
251
263
.
Janse
,
E. M.
and
Jeurissen
,
S. H. M.
(
1991
).
Ontogeny and function of two non-lymphoid cell populations in the chicken embryo
.
Immunobiol
.
182
,
472
481
.
Jeurissen
,
S. H. M.
,
Janse
,
E. M.
,
Ekino
,
S.
,
Nieuwenhuis
,
P.
,
Koch
,
G.
and
De Boer
,
G. F.
(
1988
).
Monoclonal antibodies as probes for defining cellular subsets in the bone marrow, thymus, bursa of fabricius, and spleen of the chicken
.
Veter. Immun. Immunopath
.
19
,
225
238
.
Kerr
,
J. F. R.
(
1971
).
Shrinkage necrosis: a distinct mode of cellular death
.
J. Pathol
.
105
,
13
20
.
Kerr
,
J. F. R.
,
Wyllie
,
A. H.
and
Currie
,
A. R.
(
1972
).
Apoptosis: a basic biological phenomenon with wide ranging implications in tissue kinetics
.
Br. J. Cancer
26
,
239
257
.
Lang
,
R. A.
and
Bishop
,
J. M.
(
1993
).
Macrophages are required for cell death and tissue remodelling in the developing mouse eye
.
Cell
74
,
453
462
.
Manasek
,
F. J.
(
1969
).
Myocardial cell death in the embryonic chick ventricle
,
J. Embryol. Exp. Morph
.
21
,
271
284
.
Matthew
,
W. D.
and
Patterson
,
P. H.
(
1983
).
The production of a monoclonal antibody that blocks the action of a neurite outgrowth-promoting factor
.
Cold Spring Harbor Symp. Quant. Biol
.
48
,
625
631
.
Miura
,
M.
,
Zhu
,
H.
,
Rotello
,
R.
,
Hartweig
,
E. A.
and
Yuan
,
J.
(
1993
).
Induction of apoptosis in fibroblasts by IL-1β converting enzyme, a mammalian homolog of the C. elegans cell death gene ced-3
.
Cell
75
,
653
660
.
Morris
,
R. G.
,
Hangreaves
,
A. D.
,
Duvall
,
E.
and
Wyllie
,
A. H.
(
1984
).
Hormone-induced cell death. 2. Surface changes in thymocytes undergoing apoptosis
.
Am. J. Path
.
115
,
426
436
.
Okado
,
N.
and
Oppenheim
. (
1984
).
Cell death of motorneurons in the chick embryo spinal cord. IX. The loss of motorneurons following removal of afferent inputs
.
J. Neurosci
.
4
,
1639
1652
.
Oltavi
,
Z. N.
,
Milliman
,
C. L.
and
Korsmeyer
,
S. J.
(
1993
).
Bcl-2 heterodimerizes in vivo with a conserved homolog, Bax, that accelerates programmed cell death
.
Cell
74
,
609
619
.
Oppenheim
,
R. W.
(
1991
).
Cell death during development of the nervous system
.
Annu. Rev. Neurosci
.
14
,
453
501
.
Pexieder
,
T.
(
1975
).
Cell death in the morphogenesis and teratogenesis of the heart
.
Adv. Anat. Embryol. Cell Biol
.
51
,
1
100
.
Pittman
,
R.
and
Oppenheim
,
R. W.
(
1979
).
Cell death of motoneurons in the chick embryo spinal cord, IV. Evidence that a functional neuromuscular interaction is involved in the regulation of naturally occurring cell death and the stabilization of synapses
.
J. Comp. Neurol
.
187
,
425
446
.
Robertson
,
A. M. G.
and
Thomson
,
J. N.
(
1982
).
Morphology of programmed cell death in the ventral nerve cord of Caenorhabditis elegans larvae
.
J. Embryol. Exp. Morph
.
67
,
89
100
.
Rodriguez-Torchy
,
G.
,
Collins
,
M.
and
Lopez-Rivas
,
A.
(
1990
).
Regulation of apoptosis in IL-3 dependent hemopoetic cells by IL-3 and calcium ionophores
.
EMBO J
.
9
,
2997
3002
.
Salzgeber
,
B.
and
Webver
,
R.
(
1966
).
Le régression du mésonéphros chez l’embryon de poulet. Etude des activités de la phosphatase acide et des cathepsines. Analyse biochimique, histochimique et observations au microscope électronique
.
J. Embryol. Exp. Morph
.
15
,
397
419
.
Saunders
,
J. W.
, Jr
. (
1966
).
Death in embryonic systems, Science
154
,
604
612
.
Saunders
,
J. W.
, Jr.
and
Fallon
,
J. F.
(
1967
).
Cell death in morphogenesis
. In
Major Problems in Developmental Biology
(ed.
M.
Locke
), pp.
289
314
.
New York
:
Academic Press
.
Sulston
,
J. E.
and
Horvitz
,
H. R.
(
1977
).
Post-embryonic cell lineages of the nematode Caenorhabditis elegans
.
Dev. Biol
.
56
,
110
156
.
Sulston
,
J. E.
,
Albertson
,
D. G.
and
Thomson
,
J. N.
(
1980
).
The Caenorhabditis elegans male: Postembryonic development of nongonadal structures
.
Dev. Biol
.
78
,
542
576
.
Thornberry
,
N. A.
,
Bull
,
H. G.
,
Calaycay
,
J. R.
,
Chapman
,
K. T.
,
Howard
,
A. D.
,
Kostura
,
M. J.
,
Miller
,
D. K.
,
Molineaux
,
S. M.
,
Weidner
,
J. R.
,
Aunins
,
J.
,
Elliston
,
K. O.
,
Ayala
,
J. M.
,
Casano
,
F. J.
,
Chin
,
J.
,
Ding
,
G.
,
Egger
,
L. A.
,
Gaffney
,
E. P.
,
Limjuco
,
G.
,
Palyha
,
O. C.
,
Raju
,
S. M.
,
Rolando
,
A. M.
,
Salley
,
J. P.
,
Yamin
,
T. T.
,
Lee
,
T. D.
,
Shively
,
J. E.
,
MacCross
,
M.
,
Mumford
,
R. A.
,
Schmidt
,
J. A.
and
Tocci
,
M. J.
(
1992
).
A novel heterodimeric cysteine protease is required for interleukin-1β processing in monocytes
.
Nature
356
,
768
774
.
Yuan
,
J.
,
Shaham
,
S.
,
Ledoux
,
S.
,
Ellis
,
H. M.
and
Horvitz
,
H. R.
(
1993
).
The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1β converting enzyme
.
Cell
75
,
641
652
.