Cytochrome C has two apparently separable cellular functions: respiration and caspase activation during apoptosis. While a role of the mitochondria and cytochrome C in the assembly of the apoptosome and caspase activation has been established for mammalian cells, the existence of a comparable function for cytochrome C in invertebrates remains controversial. Drosophila possesses two cytochrome c genes, cyt-c-d and cyt-c-p. Only cyt-c-d is required for caspase activation in an apoptosis-like process during spermatid differentiation, whereas cyt-c-p is required for respiration in the soma. However, both cytochrome C proteins can function interchangeably in respiration and caspase activation, and the difference in their genetic requirements can be attributed to differential expression in the soma and testes. Furthermore, orthologues of the apoptosome components, Ark (Apaf-1) and Dronc (caspase-9), are also required for the proper removal of bulk cytoplasm during spermatogenesis. Finally, several mutants that block caspase activation during spermatogenesis were isolated in a genetic screen, including mutants with defects in spermatid mitochondrial organization. These observations establish a role for the mitochondria in caspase activation during spermatogenesis (Arama, 2006).

Apoptosis is a morphologically distinct form of active cellular suicide that serves to eliminate unwanted and potentially dangerous cells. The key enzymes responsible for the execution of apoptosis are an evolutionarily conserved family of cysteine proteases known as caspases. Caspases are present in an inactive or weakly active state in virtually all cells of higher metazoans, and their activity is carefully regulated by both activators and inhibitors. In vertebrates, the mitochondria play an important role in the control of apoptosis: they release cytochrome C and other pro-apoptotic proteins in response to various death signals. In the cytosol, cytochrome C binds to Apaf-1 (Zou, 1997) which in turn promotes the assembly of a multiprotein complex, termed the 'apoptosome', and caspase-9 activation (Rodriguez, 1999; Adams, 2002; Cain, 2002; Salvesen, 2002). In the ensuing 'caspase cascade', many intracellular substrates are cleaved and apoptosis is executed. However, the exact physiological role of cytochrome C for caspase activation remains to be determined, and a recent report on a mutant cytochrome c that fails to activate Apaf-1 in the mouse (Hao, 2005) suggests that cytochrome C is required for caspase activation in only some mammalian cell types (Arama, 2006).

In invertebrates, any role of cytochrome C for the activation of caspases has remained highly controversial. Whereas RNAi experiments in Drosophila S2 cells have failed to reveal a role for cytochrome C in apoptosis, other reports suggest that cytochrome C may promote caspase activation (Dorstyn, 2002, 2004; Zimmermann, 2002). Drosophila contains two Apaf-1 isoforms: one with a WD40 repeat domain, the target for cytochrome C binding, and another lacking this domain, similar to Caenorhabditis elegans Ced-4. The large isoform can directly bind cytochrome C in vitro and promote cytochrome C-dependent caspase activation in lysates from developing embryos (Kanuka, 1999). Furthermore, an overt alteration in the cytochrome C immuno-staining can be detected in doomed cells in some Drosophila tissues, and the mitochondria from apoptotic cells can activate cytosolic caspases (Varkey, 1999). Finally, disruption of one of the two Drosophila cytochrome c genes, cyt-c-d, is associated with a failure to activate caspases in an apoptosis-like process during sperm terminal differentiation in Drosophila (Arama, 2003). In this process, also known as spermatid individualization, the majority of cytoplasm and cellular organelles are eliminated from the developing spermatids in an apoptosis-like process that requires caspase activity (Arama, 2003). However, it was suggested that the mutants used in s previous study (Huh, 2004) may also affect other genes located in the vicinity of the cyt-c-d locus (Arama, 2006).

In order to rigorously address this issue, a series of genetic and transgenic rescue experiments were conducted that unequivocally establish a role of cytochrome C for caspase activation during Drosophila spermatogenesis. First, a point mutation was isolated in cyt-c-d that is defective in caspase activation. Next, it was demonstrated that transgenic expression of cyt-c-d restores effector caspase activation and rescues all the sterility phenotypes associated with various cyt-c-d mutant alleles. The possibility that cyt-c-p functions specifically in respiration was investigated, whereas cyt-c-d plays a role in caspase regulation. Surprisingly, it was found that expression of either cyt-c-d or cyt-c-p can restore caspase activation in cyt-c-d-deficient spermatids, demonstrating that both proteins are functionally equivalent. Other apoptosome proteins in Drosophila, Ark (Apaf-1) and Dronc (caspase-9) are also required for spermatid individualization, and their mutant phenotypes are similar to spermatids with a block in caspase activity. Surprisingly, however, some active caspase-3 staining can still be detected in these mutant testes, suggesting that cytochrome-C-d may function in yet other unknown pathways to promote caspase-3 activation. Finally, several mutants affecting spermatid mitochondria were identifed that provide a strong link between mitochondrial organization and caspase activation during sperm development (Arama, 2006).

In mammals, mitochondria are important for the regulation of apoptosis, and it has been shown that they can release several proapoptotic proteins into the cytosol in response to apoptotic stimuli. The best-studied case is the release of cytochrome C, an essential component of the respiratory chain. Cytosolic cytochrome C can bind to and activate Apaf-1, which in turn leads to the activation of caspase-9. However, no comparable role of mitochondrial factors for caspase activation has yet been established in invertebrates. The elimination of cytoplasm during terminal differentiation of spermatids in Drosophila involves an apoptosis-like process that requires caspase activity; a P-element insertion (bln¹) in one of the two Drosophila cytochrome c genes, cyt-c-d, has been shown to be associated with male-sterility and loss of effector caspase activation during spermatid individualization. This study demonstrates that the defects in caspase activation and spermatid individualization of bln¹ mutant males can be rescued by transgenic expression of the ORF of cyt-c-d. Furthermore, from screening more than a thousand male-sterile lines with defects in sperm individualization for defects in active-caspase (CM1) staining, a nonsense point mutation was identified in cyt-c-d, that recapitulates all the phenotypes observed for bln¹. Taken together, these results unequivocally demonstrate that cyt-c-d is necessary for effector caspase activation and sperm terminal differentiation in Drosophila (Arama, 2006).

Two decades ago, the mouse cytochrome c gene was used as a probe for screening a Drosophila genomic library and a fragment was isolated that carried two distinct cytochrome c genes. Northern blot analyses indicated high levels of cyt-c-p expression, while cyt-c-d was reported to be expressed at much lower levels in all stages of development. However, neither the exon/intron organization nor the boundaries of the 5' and 3' UTRs of these genes were determined at the time. As a result, the original Northern analyses were performed with a probe corresponding to the untranscribed genomic region between the two cytochrome c genes that was not suitable to properly assess the size and distribution of cytochrome c transcripts. Unfortunately, this has caused considerable confusion in the field from the start, as even the original report noted that the size of the observed cyt-c-d transcript differed more than two-fold from the predicted size. More recently, relying on the incorrect assumption that cyt-c-d is ubiquitously expressed in the fly, it has been suggested that a loss-of-function mutation in cyt-c-d should lead to severe developmental defects and lethality rather than merely male sterility. However, using a specific cyt-c-d 3' UTR probe reveals a transcript of the predicted size that is absent in cyt-c-d^bln1 mutants. Furthermore, the RT–PCR and immunofluorescence analyses presented in this study indicate that cyt-c-d is mainly expressed in the male germ line and is completely absent during embryonic and larval development, while cyt-c-p is expressed in the soma during all stages of development. In light of these findings, it is not surprising that loss-of-function mutations in cyt-c-d cause male sterility, whereas cyt-c-p mutations lead to embryonic lethality. RT–PCR results suggest that cyt-c-p is also expressed in the testis, although to a much lower extent than cyt-c-d. This expression is attributed primarily to the somatic cells of the testis, since no cytochrome C protein is detected in cyt-c-d^bln1 elongating spermatids, while cyt-c-p RNA is expressed in cyt-c-d^bln1 mutant flies. However, the very low cyt-c-d expression detected in the soma of adult females leaves room for the possibility that cyt-c-d might function in caspase activation in some somatic cells as well (Arama, 2006).

In mammalian cells, release of cytochrome C into the cytosol in response to proapoptotic stimuli can be readily demonstrated. However, previous attempts to detect a similar phenomenon in Drosophila have been unsuccessful. In contrast, apoptotic stimuli can lead to increased cytochrome C immuno-reactivity. A possible limitation is that all these studies were conducted using mammalian antibodies with questionable specificity and sensitivity, and only in a small number of cell types and paradigms. Using an antibody that was raised against Drosophila cytochrome C-d, an increase in a 'grainy signal' was detected upon the onset of individualization, with the highest staining observed in the vicinity of the individualization comple (IC). Since it is highly unlikely that additional cytochrome C-d is being transcribed and imported to the mitochondria at this late stage, the explanation is favored that a conformational change or an exposure of a hidden epitope causes the increase in the intensity of the signal. The activation of Dronc, the Drosophila caspase-9 orthologue, also occurs in association with the IC and depends on the presence of the Drosophila Apaf-1 orthologue, Ark. Moreover, the proapoptotic Hid protein is localized in a similar fashion. What are these structures then, which accumulate apoptotic factors in the vicinity of the IC? One plausible suggestion from the literature is that these structures correspond to 'mitochondrial whorls', which result from the extrusion of material from the minor mitochondrial derivative and constitute the leading component of the IC. These 'whorls' can be labeled using a testes-specific mitochondrial-expressed GFP line. Using this GFP marker, it was found that cytochrome C-d is indeed closely associated with mitochondrial whorls. Therefore, it is possible that an active apoptosome forms in the vicinity of the IC in response to dramatic changes in the mitochondrial architecture that occur at this stage of spermatid differentiation. Similarly, studying the response of Drosophila flight muscle cells to oxygen stress, have recently reported that the cristae within individual mitochondria become locally rearranged in a pattern that they termed a 'swirl'. This process was associated with widespread apoptotic cell death in the flight muscle, which was correlated with a conformational change of cytochrome C manifested by the display of an otherwise hidden epitope. Collectively, these observations suggest that apoptosome-like complexes composed of cytochrome C-d, Ark, and Dronc might be associated with unique mitochondrial swirl-like structures. Consistent with this idea, it was found that the long isoform of Ark that contains the WD40 repeats, the target for cytochrome C binding to mammalian Apaf-1, is the major form detectably expressed in testes (Arama, 2006).

The fact that cytochrome C-d immunoreactivity increases in the vicinity of the IC suggests that the extensive mitochondrial organizations preceding individualization may be partially required for caspase activation. Consistent with this idea, several mutants, such as pln^Z2-0516, which display defects in Nebenkern differentiation and caspase activation. However, not all mitochondrial differentiation events are required for caspase activation. For example, CM1 staining is seen in fuzzy onions, a mutant defective in the mitochondrial fusion event that generates the Nebenkern. In contrast, analysis of the pln mutant indicates that proper elongation of the Nebenkern is essential for caspase activation. Therefore, characterization of other mitochondrial mutants may shed light on the connection between mitochondrial organization and caspase activation during sperm differentiation (Arama, 2006).

What are the mechanisms by which cytochrome C-d activates caspases during late spermatogenesis? In vertebrate cells, following its release into the cytosol, cytochrome C binds to the WD40 domain of the adaptor molecule Apaf-1, which in turn multimerizes and recruits the initiator caspase, caspase-9 via interaction of their CARD domains. This complex, known as the apoptosome, further cleaves and activates effector caspases like caspase-3. Although this model has become the prevailing dogma in the field, the phenotype of mice mutant for a Cyt c with drastically reduced apoptogenic function ('KA allele') suggests that the mechanisms for caspase activation may be more complex than previously thought. In particular, this study suggests that cytochrome C-independent mechanisms for the activation of Apaf-1 and caspase-9 exist, as well as cytochrome C-dependent but Apaf-1-independent mechanisms for apoptosis. These analyses of ark (Apaf-1) and dronc (caspase-9) loss-of-function mutants demonstrate that both genes are required for spermatid individualization, and that their phenotypes, in particular their failure to properly remove the spermatid cytoplasm into the WB, resemble cyt-c-d mutant spermatids and expression of the caspase inhibitor p35 in the testes. However, some caspase-3-like activity could still be detected in these mutant testes. This may suggest that either the ark and dronc alleles are not null, or that cytochrome C-d also functions in an apoptosome-independent pathway to promote caspase-3 activation. Therefore, the regulation of caspase activation and apoptosis may be more similar between insects and mammals than has been previously appreciated. Further genetic analysis of this pathway in Drosophila may provide general insights into diverse mechanisms of apoptosis activation (Arama, 2006).

Previous observations raised the possibility that the two distinct cytochrome c genes may have evolved to serve distinct functions in respiration and caspase regulation. In order to address this hypothesis, it was asked whether expression of one protein might rescue mutations in the other cytochrome c gene. Surprisingly, it was found that transgenic expression of the cyt-c-p ORF in germ cells rescues caspase activation, spermatid individualization, and sterility of cyt-c-d^-/- flies. Therefore, the ability to activate caspases is not restricted to the cytochrome C-d protein, and it is possible that cytochrome C-p functions in apoptosis in at least some somatic cells (Arama, 2006).

Although cyt-c-d is almost exclusively expressed in the male germ cells, ectopic expression of this protein in the soma can rescue the respiration defect and lethality of cyt-c-p^-/- mutant flies, demonstrating that cytochrome C-d can function in energy metabolism. This raises the question whether the lack of caspase activation could be due to reduced ATP-levels. Although this is a formal possibility, this explanation is considered very unlikely since mutant spermatids complete many other energy-intensive cellular processes. These include the extensive transformation from round spermatids to 1.8 mm long elongated spermatids, a process that involves extensive remodeling and movement of actin filaments, generation of the axonemal tail, mitochondrial reorganization, plasma/axonemal membranes reorganization, and nuclear condensation and elongation. Since all of these processes can occur in the absence of cytochrome C-d, there is no overt shortage of ATP in cyt-c-d mutants. It is therefore considered very unlikely that ATP has become limiting in these mutant cells. Since earlier stage spermatids express cytochrome C-p, sufficient ATP seems to persist to late developmental stages. In mammalian cells, cellular ATP concentration is sufficiently high (around 2 mM) to keep cultured cell alive for several days upon ATP synthase inhibition. Furthermore, cells in which cytochrome c expression is decreased by RNAi still undergo apoptosis in response to various stimuli. Likewise, it appears that cytochrome C is not essential for the function of mature murine sperm, since mice deficient for the testis specific form of cytochrome C, Cyt c_T, are fertile. Taken together, all these observations argue strongly against the possibility that ATP levels in cyt-c-d^-/- mutant spermatids would be insufficient for caspase activation (Arama, 2006).

In conclusion, the results presented in this study definitively demonstrate that cytochrome C-d is essential for caspase activation and spermatid individualization. Both cytochrome C proteins of Drosophila are, at least to some extent, functionally interchangeable. The results also indicate that cytochrome C can promote caspase activation in the absence of a functional apoptosome. Given the powerful genetic techniques available, late spermatogenesis of Drosophila promises to be a powerful system to identify novel pathways for mitochondrial regulation of caspase activation (Arama, 2006).

GENE STRUCTURE

cDNA clone length - 754 bp (Cyt-c-p) and 877 bp (Cyt-c-d)

Bases in 5' UTR - 155 (Cyt-c-p) and 369 (Cyt-c-d)

Exons - 2 (Cyt-c-p) and 2 (Cyt-c-d)

Bases in 3' UTR - 283 (Cyt-c-p) and 190 (Cyt-c-d)

PROTEIN STRUCTURE

Amino Acids - 108 (Cyt-c-p) and 195 (Cyt-c-d)

Structural Domains

Analysis of total Drosophila melanogaster DNA by genomic blot hybridization indicates that two cytochrome c-like sequences exist in the Drosophila genome. These two sequences, DC3 and DC4, have been isolated from a Charon 4A-D. melanogaster genomic library. DC3 and DC4 are located within a 4 kb region of DNA, at position 36A 10-11, on the left arm of chromosome 2. The nucleotide sequence of these two clones has been determined. Both DC3 and DC4 can encode functional cytochrome c proteins. The polypeptide sequences predicted by these two genes, however, differ at 32 amino acid residues. DC4 is expressed at varying, but relatively high levels throughout Drosophila development. In contrast, DC3 is expressed at constant, but relatively low levels throughout development (Limbach, 1985: full text of article).

The amino acid sequences of cytochromes c purified from the fruit fly Drosophila melanogaster and the flesh fly Boettcherisca peregrina were determined. In contrast with the case of the housefly, isocytochromes c were not detected in these flies at any developmental stage. The sequence of fruit fly cytochrome c differed from that reported previously but was identical with that predicted from the nucleotide sequence of the fruit fly cytochrome c gene (DC4). Isocytochrome c of the fruit fly, reported to be encoded by the DC3 gene, was not detected as a functional cytochrome c molecule (Inoue, 1986).

date revised: 15 March 2007

Home page: The Interactive Fly © 2006 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.