SEARCH

SEARCH BY CITATION

Keywords:

  • embryonic cell death;
  • cathepsin D;
  • limb development;
  • heart development

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The implication of lysosomes in the activation of physiological cell death (PCD) was proposed some decades ago. In this work, we show that the expression of the lysosomal enzyme cathepsin D is up-regulated in developing tissues undergoing apoptosis. By comparing vital and terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) labeling patterns with in situ hybridization for this gene in a variety of tissues and organs, we show that this procedure constitutes a reliable technique to map the regions of PCD in the embryo. Using this methodological approach, we report the occurrence of two new areas of PCD in the developing limb. Developmental Dynamics 236:880–885, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Physiological cell death (PCD) is an important developmental mechanism involved in morphogenesis and histogenesis of most embryonic organs. The sculpturing of the shape of growing organs, the elimination of structures with transitory functions in the embryo, or the structural and functional organization of complex tissues, such as the developing nervous system or the immune system, are some of the functions detected for PCD (Glucksmann,1951; Buss et al.,2006). Moreover, cell death is a critical target for many pathological stimuli, and alterations in the physiological patterns of cell death constitute the basis of many congenital malformations.

The evaluation of the function of cell death during development requires precise information concerning its spatial distribution. This information is often difficult to obtain through the examination of tissue sections. The introduction of cell death labeling procedures in whole-mount approaches has been of great help to identify areas of physiological cell death and to unravel their developmental significance (Saunders et al.,1962; Merino et al.,1999). Vital staining with Nile Blue Sulphate, Neutral Red, Acridine Orange, or Cresyl Blue constitutes the classic approach to characterize the areas of PCD in the embryo (Saunders et al.,1962; Hinchliffe and Griffiths,1986). However, in practice, only superficial structures are accessible to vital staining, while organs and tissues located deep into the embryonic body are hardly accessible to the dye. In addition, except for phagosomes of the macrophages, the nature of the structural components of dying cells stained by vital dyes has been questioned (Sanders and Wride,1995). Because most of the physiological cell death processes occur by apoptosis, more recent approaches to map areas of cell death in the embryo are based on the detection of DNA fragmentation (terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling [TUNEL] assay) in whole-mount specimens using confocal microscopy (Hensey and Gautier,1998). Modern fluorescent vital dyes in combination with confocal microscopy have been also introduced to characterize the areas of cell death (Zucker et al.,1999). However, these techniques are also difficult to perform for organs or structures located deep into the embryonic tissues.

In a recent study, we have observed that lysosomal cathepsins cooperate with caspases in the execution of cell death in the developing limb and expression of cathepsins B and D genes appeared as specific markers for PCD in this system (Zuzarte-Luis et al.,2007). Here, we have analyzed the expression of cathepsin genes in chick embryos to check the reliability of this procedure to detect and characterize the spatial distribution of areas of PCD. We show that expression of cathepsin D is a very good marker for areas of embryonic cell death, including those of deep organs and tissues, such as the myocardial and cushion mesenchymal tissue of the developing heart, that are difficult to examine by conventional vital staining approaches. In addition, examination of in situ hybridization samples allowed us to identify new areas of cell death in the developing limb neglected in previous studies.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Here, we have observed that expression of the cathepsin D gene is a reliable marker of the areas of PCD in the developing embryo. We selected a variety of embryonic areas of cell death containing cells of different lineages or located in places of the embryo of difficult accessibility by vital cell death staining and compared its distribution with the expression of cathepsin D.

Cell Death in Superficial Epithelial and Mesodermal Tissues

The use of vital staining approaches provides very accurate images of cell death processes occurring in structures located close to the embryo surface. The invagination of the lens placode is a very illustrative example of morphogenetic cell death in ectodermal tissue easily detected by vital staining (Fig. 1A; Garcia-Porrero et al.,1979). As shown in Figure 1B, local up-regulation of the cathepsin D gene closely preceded the pattern of cell death detected by vital staining. Consistent with this finding, lysosomal activation has been formerly reported to occur in the dying ectodermal cells (Garcia-Porrero et al.,1984).

thumbnail image

Figure 1. Cathepsin D is expressed in cell death areas of the developing eye and somites. A: Neutral Red staining for cell death in the invaginating lens placode (arrow). B: In situ hybridization showing the parallel expression of cathepsin D in the same region of the developing eye (arrow). C,D: Cell death in developing somites of Hamburger and Hamilton stage (HH) 21 embryo stained with Neutral Red (arrows in C) and its parallel cathepsin D expression (arrows in D).

Download figure to PowerPoint

Vital staining of chick embryos during Hamburger and Hamilton stages (HH) 17 and HH23 reveals clumps of dying cells within the paraxial mesoderm, which are grouped in characteristic segmental patterns described in detail by Jeffs and Osmond (1992; Fig. 1C). These areas of cell death have been mentioned in the earliest review of embryonic cell death (Glucksmann,1951) and identified as corresponding to the developing sclerotome by a variety of labeling procedures (Sanders,1997). To our knowledge, there is not detailed cytological or histochemical analysis testing the implication of lysosomes in this apoptotic process. Here, we observed that cathepsin D gene expression exhibits a dotted domain of expression with a segmental pattern of distribution (Fig. 1D), closely correlated with the pattern of cell death detected with vital staining.

Cell Death in the Developing Heart

The developing heart was selected in this study to illustrate cell death in mesenchymal and myocardial tissues located deep into the embryonic body, which are difficult to illustrate in whole-mount samples. In previous studies using mainly tissue sections, we and others have reported that, in chick and mouse embryos during the stages of heart septation, areas of cell death are present in the mesenchymal tissue of the outflow tract and atrioventricular cushions (Fig. 2A–D; Hurle and Ojeda,1979; Sharma et al.,2004). In addition, the myocardial layer of the proximal segment of the outflow tract also undergoes a dramatic cell death process (Fig. 2A–C; Hurle et al.,1977; Sugishita et al.,2004; Barbosky et al.,2006). All the mentioned areas of cell death appeared clearly outlined by the expression domain of the cathepsin D gene in our study (Fig. 2E–G). The cathepsin expression domains are easily appreciated through the surface of the heart in specimens cleared with 50% of glycerol (Fig. 2E,F), but domains of expression in cushion tissue mesenchyme are better appreciated in microdissected heart fragments (Fig. 2G).

thumbnail image

Figure 2. Cathepsin D is expressed in cell death areas of the developing heart. A: Schematic diagram of a transverse heart section illustrating cell death in the outflow tract (red dotted areas). Note distinct areas of cell death in the myocardial layer (m) and mesenchymal cushions (c). B: Frontal section of the outflow tract of the embryonic heart at Hamburger and Hamilton stage (HH) 30 labeled with terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) to show apoptotic cells in the myocardial layer (m) and in the cushion tissue mesenchyme (c). C: Detailed view of cell death in the myocardial tissue of the outflow tract after double labeling for sarcomeric myosin (red) and TUNEL (green). D: Schematic diagram of a frontal section of the heart illustrating cell death (red dotted area) in the atrioventricular cushion tissue at stage HH31. E: Glycerol-cleared HH30 heart specimen showing the domains of cathepsin D expression in the outflow tract cushion tissue (arrows), which correlate with the regions of cell death formerly detected in tissue sections (see c in A). F: Ventral view of un-cleared embryonic hearts at stage HH31 showing the expression of cathepsin D in the proximal segment of the outflow tract. This domain of cathepsin D expression correlates with the area of myocardial cell death (see m in A). G: Embryonic HH31 heart microdissection to expose the atrioventricular cushion and valves, showing the expression of cathepsin D paralleling the area of cell death present in this structure (see D). H: Embryonic HH31 heart microdissection to expose the atrioventricular cushion and valves, showing the strong expression of cathepsin B in the atrioventricular valves leaflets (arrows).

Download figure to PowerPoint

Because in the developing heart cathepsin proteases are involved in tissue remodeling processes lacking cell death (Lange and Yutzey,2006; Petermann et al.,2006), to check the specificity of cathepsin D in marking apoptotic events, we compared its expression with other cathepsins. We observed that cathepsin B, which is a good marker for areas of cell death in the developing limb (Zuzarte-Luis et al.,2007), is expressed in the developing valves (Fig. 2H) in a pattern coincident with that previously observed for cathepsin K (Lange and Yutzey,2006). Cathepsin L is also a cysteine protease present in the advanced stages of the areas of cell death in the developing limb (Zuzarte-Luis et al.,2007), which has been implicated in the remodeling of the heart extracellular matrix (Petermann et al.,2006). In the embryo, we have appreciated a weak basal expression of this enzyme without specific domains associated with the areas of cell death (not shown). Together these findings suggest that cathepsin D plays a distinctive role in cell death with respect to other members of this protease family.

New Areas of Cell Death in the Developing Limb

The limb bud was chosen because it has been one of the embryonic structures studied in more detail in relation with PCD (see Zuzarte-Luis and Hurle,2002; Fernandez-Teran et al.,2006). We previously reported that cathepsin B and D expression appeared very precise markers for the anterior and posterior areas of cell death (ANZ and PNZ) and also for the areas of interdigital cell death (INZ; Zuzarte-Luis et al.,2007). However, in this study cathepsin D expression drew our attention toward two regions of embryonic limb, where physiological cell death has not been previously noticed. One of these areas occurred in the mesoderm occupying the interappendicular level of the flank of stage 18–20 embryos (Fig. 3A). The increased accumulation of cells with intense positivity for cathepsin D was consistent with the presence in tissue sections of TUNEL-positive cells scattered under the ectoderm (Fig. 3D). It is remarkable that, before stage HH17, this region of the embryonic flank is able to form an extra limb upon implantation of fibroblast growth factor beads (Cohn et al.,1995). Hence, the presence of the apoptotic area may account for the loss of such a regenerative potential.

thumbnail image

Figure 3. Two new areas of cell death in the developing limb are revealed by the expression of cathepsin D. A: Intense expression of cathepsin D in the embryonic flank between both limb primordia (arrows), which correlates with an area of cell death in the interappendicular mesodermal tissue (see D). B: Ventral view of the chick leg bud at Hamburger and Hamilton stage (HH)31 showing a lineal domain of cathepsin D expression along each lateral margin of the autopod (white arrows). C: A detailed view of the zone of the posterior margin of the bud indicated by an asterisk in B showing that, in this region, a well-defined ectodermal domain of expression is clearly distinguished from the subjacent mesodermal domain (asterisk in C). D: Terminal deoxynucleotidyl transferase–mediated deoxyuridinetriphosphate nick end-labeling (TUNEL) staining of a transversal section of the embryo at the level of the interappendicular region where abundance of apoptotic cells is detected at the level of the flank mesenchyme (arrows). E,F: TUNEL-stained transverse (E) and longitudinal (F) sections of the leg bud through the ectodermal domain of expression of cathepsin D indicated by an arrowhead in B, showing a well-defined region of massive ectodermal apoptosis. G: Two consecutive transverse Vibratome sections in situ hybridized for cathepsin D (upper) and for the dorsal marker Lmx1 (lower) showing the precise location of the cathepsin domain in the boundary between the dorsal Lmx1-expressing mesenchyme and the ventral region of the limb (negative for Lmx-1).

Download figure to PowerPoint

The second area that we analyzed was located in the margins of the developing limb. In situ hybridization studies revealed a well-defined line of cathepsin D expression marking the lateral margins of the limb bud (arrows in Fig. 3B,C). Again by TUNEL assay in tissue sections, we confirmed that this domain corresponded with a precise area of apoptotic cells located mainly in the ectodermal tissue (arrowhead in Fig. 3E,F). By comparing the expression of cathepsin D and Lmx1b, a transcription factor whose expression delimits the dorsal territory of the limb (Riddle et al.,1995), we observed that the apoptotic cells were precisely located in the boundary between the dorsal and ventral compartments of the limb bud (Fig. 3G). At early stages of development (23 to 26), this ectodermal cell death was closely associated with mesodermal cell death of ANZ and PNZ, but from stages 29–30, cell death was restricted to the margins of the proximal portion of the autopod. Whether this apoptotic area is related to a specific manner of growth and renewal of the limb, ectodermal tissue remains to be analyzed. However, the precise location of this area in the borderline between tissue domains with opposite dorsal and ventral molecular identities emphasizes the potential impact of dorsoventral molecular determinant factors on cell survival.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

This study shows that the cathepsin D gene exhibits specific domains of expression in the embryo coincident with the areas of PCD. Positivity was observed both in mesodermal and epithelial tissues. Hence, whole-mount in situ hybridization for this gene constitutes a useful procedure to evaluate the spatial distribution of areas of cell death difficult to recognize by other approaches. In this regard, the identification by this procedure of two new areas of cell death in the developing limb bud, which escaped detection in numerous extensive studies using vital staining or TUNEL assay in tissue sections (Zuzarte-Luis and Hurle,2002), is remarkable. The usefulness of the in situ hybridization approach to detect cell death was also remarkable in the developing heart. The original description of PCD in the embryonic heart required the careful analysis of serial sections and subsequent spatial reconstructions of the areas of cell death (Hurle and Ojeda,1979).

Vital staining with dyes such as Neutral Red and Nile Blue has been widely used to study PCD. These dyes are acidophilic and are concentrated in acidic membrane-bound cytoplasmic compartments. Early studies described the staining pattern of the areas of cell death as constituted by small spots corresponding with the dye concentrated in the autophagic vacuoles of dying cells and larger spots corresponding with the heterophagic vacuoles of macrophages engaged in the cleaning up of the dead and dying cells (Hinchliffe and Griffiths,1986). However, the staining specificity for dying cells has been questioned (Sanders and Wride,1995), because it was believed that autophagy and lysosomal activation were absent in apoptotic cell death (Wyllie et al.,1980). The expression pattern of cathepsin D observed here indicates that lysosomes are intensely activated in most areas of PCD, suggesting they are responsible for the high affinity of the apoptotic areas to vital staining. Consistent with this interpretation, the lysosomal marker LysoTracker Red has been successfully introduced as vital staining marker to map areas of cell death (Zucker el al.,1999; Barbosky et al.,2006).

A central question in PCD is the molecular mechanism responsible for cell killing (Yuan,2006). Studies in a large variety of organisms have established that PCD is a suicide cell behavior. This means that cells have, and are able to self-activate, the molecular machinery that cause their destruction. Earliest studies of embryonic cell death proposed the lysosomes as central elements in the killing pathway (Hurle and Hinchliffe,1978). However, the introduction of the term apoptosis to define cell death resulting from the activation of a cell suicide program discarded the implication of lysosomes (Wyllie et al.,1980). In the past decade, a considerable number of studies have rescued lysosomes and lysosomal enzymes as potential players in a variety of apoptotic processes (Leist and Jaattela,2001a,b; Kroemer and Jaatela,2005), including the areas of interdigital cell death of mouse and chick embryos (Zuzarte-Luis et al.,2007). The up-regulation of cathepsin D gene observed here in areas of embryonic cell death suggests a wide implication of lysosomes in PCD. In support of this interpretation, there is now compiling evidence showing that lysosomal cathepsins are able to mediate cell death by apoptosis in a variety of systems (Leist and Jaattela,2001a,b). Cathepsins may activate caspases at the initial stages of the apoptotic processes (Ishizaki et al.,1998; Schotte et al.,1998; Vancompernolle et al.,1998). However, it has been also shown that capthesins are able to promote the mitochondrial release of apoptotic inducing factor (AIF) even when caspase function is impaired (Bidere et al.,2003).

Here, we cannot discard that, in some areas of cell death, lysosomal activity may reflect the phagocytic removal of resulting cell debris rather than being a primary apoptotic event (Wyllie et al.,1980). This interpretation is more likely to occur in areas of cell death, such as those of the developing heart in which phagocytosis of cell debris appears to be carried out mainly by local cells (Hurle et al.,1978; Sharma et al.,2004). However, even in this case, the precise association of cathepsin D, but not cathepsin B, L, or K, with the areas of cell death points to a specific role of this protease rather than a general unspecific involvement of lysosomes in cell death.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Animal Model

Rhode Island chicken embryos ranging from 2 to 9 days of incubation (stages 15–35 of Hamburger and Hamilton,1951) were used.

In Situ Hybridization

For in situ hybridization, antisense probes for cathepsin D, B, and L (Zuzarte-Luis et al.,2007) or Lmx1b were synthesized. Embryos were dissected free, fixed in paraformaldehyde, and dehydrated in methanol. In some experiments, 100-μm Vibratome sections of the limb were taken. Specimens were stored in 100% methanol at −20°C until use. In situ hybridization was performed in whole-mount specimens treated with 10 mg/ml of proteinase K for 20–30 min at 20°C. Hybridization with digoxigenin-labeled antisense RNA probes was performed at 68°C. Reactions were developed with 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium substrate or with purple AP substrate (Roche). Control hybridizations were performed with sense probes of Cathepsin D to test the specificity of the signal.

Neutral Red Staining

For vital staining, embryos or embryonic tissue fragments were dissected free and immersed in Neutral Red diluted in phosphate buffered saline at 0.01% at 37°C for 15 to 20 min until a positive staining of the areas of cell death was obtained.

TUNEL and Immunolabeling

In these experiments, embryos were fixed in paraformaldehyde. TUNEL assay and antibody staining using F59 (Hybridoma Bank) to muscle sarcomeric myosin were performed either in 10-μm-thick sections of paraffin wax embedded tissue or in 75-μm-thick Vibratome sections from un-embedded tissue.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Montse Fernandez Calderon and Sonia Perez Mantecon for excellent technical assistance. Antibody F59 was obtained from the Developmental Studies Hybridoma Bank, University of Iowa. J.M.H. and J.A.M. were supported by grants from the Spanish Education and Sciences Ministry, and J.A.M. is supported by the Ramon y Cajal program from the Spanish Education and Sciences Ministry.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES