The Interactive Fly
Genes involved in tissue and organ development
A role for actin dynamics in individualization during spermatogenesis
Cytoskeletal dynamics in male meiosis
The search for Y-linked genes: Y chromosome fertility factors encode dynein heavy chain polypeptides
During late stages of spermatogenesis in Drosophila, a cyst of 64 syncytial spermatids elongates as the sperm axonemes are formed inside it. Then this elongated cyst is remodeled into individual sperm by a process called individualization. At the start of individualization actin cones assemble around the spermatid nuclei and then synchronously move from the heads to the tips of the tails. As the actin cones move, a large accumulation of cytoplasm and vesicles, called the cystic bulge, forms around them. In the cystic bulge, the membrane of the cyst is remodeled to enclose each sperm axoneme. Individualization is especially interesting as a cell biological process because it requires an unusual amount of membrane remodeling using a well-defined actin structure. The fully elongated cyst can be up to 1800 microm long; therefore, this process requires the actin structures important for the process to move unidirectionally over a significant length. During the process, the bulk of the cytoplasm is discarded from the cell body. However, there is little information about the mechanism of this process (Noguchi, 2003).
In order to better understand the mechanism of sperm individualization, an in vitro culture system was developed in which live observation of individualization can be performed in isolated cysts. The whole process of individualization, during which a bundle of 64 syncytial spermatids is separated into individual sperm, takes place in these cultures. The speed of cystic bulge movement is fairly constant along the length of the cyst. Actin drugs, but not microtubule drugs inhibit cystic bulge movement, suggesting that the movement requires proper actin dynamics but not microtubules. GFP-tagged actin was expressed in the cyst and fluorescence recovery after photobleaching was monitored using confocal microscopy to analyze actin dynamics in cones. Actin turns over throughout the cone, with that at the leading edge of the cones turning over with slightly faster kinetics. Actin does not treadmill from the front to the back of the cone. Actin in moving actin cones turns over in about 12 minutes, although prior to onset of movement, turnover is much slower. Visualization of membrane using the dye FM1-43 reveals that the cystic bulge has an extremely complicated series of membrane invaginations and the transition from syncytial to individualized spermatids occurs at the front of the actin cones. It is also suggested that endocytosis and exocytosis might not be important for membrane remodeling. This system should be suitable for analysis of defects in male sterile mutants and for investigating other steps of spermatogenesis (Noguchi, 2003).
These data are most consistent with the idea that actin cone movement is driven by actin polymerization, similar to lamellipodia extension and Listeria motility. The speed of the cystic bulge movement is similar to the speed of movement of the leading edge of lamellipodia. Cystic bulge movement is altered very quickly after inhibiting either assembly or disassembly of actin, consistent with the requirement for active actin assembly and disassembly for movement. In fluorescence recovery after photobleaching (FRAP) experiments, a slightly faster rate of turnover at the front was detected than in the rear of the cone. It is likely that the faster dynamics of actin at the front is important for movement. In addition, the observed acceleration of actin dynamics after the onset of movement supports this idea. Arp 2/3 complex, which is the key factor involved in promoting actin polymerization at the leading edge, is enriched at the front of actin cones, suggesting that this site is important for force generation. All these data support the idea that the driving force is actin polymerization (Noguchi, 2003).
However, some puzzling differences in actin behavior in this structure when compared with leading edge protrusion make it difficult to explain how assembly drives movement in this case. (1) It had been expected that actin would treadmill through the actin cone from front to back, because of assembly at the front, i.e. in the direction of movement. This has been observed in other actin motility processes. However, this is not the case in actin cones. The filaments in the cone move forward relative to the substrate. (2) Actin turns over at a rate that is much slower than that of actin in lamellipodia and Listeria comet tails. In both of these structures, filaments turn over in 1-2 minutes, but in case of actin cones, turnover takes 12 minutes. (3). Another puzzle is the stability of actin cones to depolymerization by the actin depolymerizing drug latrunculin A (LTA). Actin completely turns over in 12 minutes in moving cones, so it might be expected that LTA would cause depolymerization in that time frame. However, even after 2 hours of LTA treatment, cones remain. It is likely that actin in cones is stabilized by binding of cross linkers or other proteins, but an understanding is not yet clear of the mechanism that regulates stability to permit turnover as the cones move, but prevents depolymerization when assembly stops (Noguchi, 2003).
Despite these differences from other motility processes, the favored model involves only actin assembly as the driving force for motility. In order to explain the dynamics of actin in the cone and results of pharmacological experiments, it is suggested that there are two actin structural components in a moving actin cone. The first actin structure is the actin cone itself. Three characteristics (stability, filament translocation and slow turnover) suggest that the actin cone is a highly organized and stable structure compared to the actin network in lamellipodia. It is likely that each actin cone moves forward as one unit. The second actin structure component is an actin network near the membrane that pushes the actin cone forward by force of polymerization. Actin filaments elongate near the membrane, similar to the leading edge, but the membrane is held rigid, rather than protruding as it does at the leading edge. In this case, a photobleached GFP-actin in a filament would be pushed away from the membrane, i.e. `forward' relative to the membrane. Eventually, this filament would be crosslinked into the actin cone as new actin filaments assembled. This model is consistent with the data, no information is available about the orientation of actin filaments and the sites of actin monomer incorporation that would provide additional support for such a model. In addition, the molecules that might be important to keep the membrane rigid and prevent its protrusion are as yet unknown (Noguchi, 2003).
An alternative model is that myosin based motility contributes to movement by generating force using cortical actin and/or actin cones as a substrate for movement. Using a motor protein to provide force is compatible with the idea that the actin cone moves forward as a unit. However, it is not obvious why motor-driven movement would be coupled with actin dynamics. Perhaps movement requires a dynamic cortical actin network around the actin cone, to provide tracks for myosin movement. This track might need to be continually assembled at new sites during movement. Since the actin cone is a very large and intensely labeled structure, it might be hard to detect the different dynamic behavior in a less prominent, thin cortical structure in the same region. Myosin VI is present on actin cones and essential for individualization, making it tempting to speculate that myosin VI might provide the driving force. However, cystic bulges of myosin VI mutants can move partway along the cyst, indicating its function is not required for cone movement. Instead, myosin VI is important for regulating actin dynamics during movement. Other myosins may be important for some aspects of cystic bulge movement, but this remains to be demonstrated. BDM, an inhibitor of myosin ATPases, did not block movement. This inhibitor has been demonstrated to block activity of myosin I, II and V, suggesting it is a general inhibitor. The lack of effect of BDM on cone movement makes a myosin-based motility model less likely: myosins cannot be ruled out as force generators in this process at this time. Mutant alleles of all the predicted myosins in the Drosophila are not available and the effect of BDM on myosins in many classes is unknown (Noguchi, 2003).
Microtubule-based motility is not likely to be involved in cone movement. There are no cytoplasmic microtubules, which might participate in generating force in cooperation with microtubule motors and inhibitors of microtubule dynamics, and motors do not stop movement. In addition, when actin dynamics are altered, the cystic bulge stops immediately. If the movement was microtubule based, it is not clear why actin dynamics would be important (Noguchi, 2003).
Further studies are required to provide support for this model of actin cone motility. Additional studies examining membrane dynamics, effect of disruption of actin polymerization regulators and ultrastructure of the actin cones will be needed for more insight into the similarities and differences in the mechanism of actin cone movement and lamellipodia extension (Noguchi, 2003).
Although the mechanism of actin cone movement is not fully understood, it is possible to speculate about the role of actin cones during individualization. It is suggested that the actin cones have three roles: (1) the actin cones have the ability to push the cystic bulge forward, using actin polymerization; (2) the actin cones sweep the cytoplasm and organelles out of the sperm flagella, acting as a sieve; (3) the actin cones must bind the cell membrane around them and shape it into the observed thin tubular structure. Eventually, as the actin depolymerizes at the cone tip, the membrane must attach to the axoneme (Noguchi, 2003).
There is an interesting transition that occurs as individualization begins. Microtubule staining disappears during a very short period around the onset of actin cone movement. The data suggests that this disappearance is due to tubulin degradation as movement begins. This idea is supported by observations that the amount of tubulin present in individualized spermatids is much less than in cysts prior to individualization, and that cytoplasmic microtubules disappear during individualization (Noguchi, 2003).
This transition temporally coincides with the onset of actin cone movement, rather than sperm nucleus DNA condensation. FRAP experiments demonstrate that actin dynamics also accelerate after the onset of movement. Therefore, it is suspected that a global signal orchestrates these events to trigger the onset of individualization (Noguchi, 2003).
Membrane remodeling does not require endocytosis or exocytosis. Conventional endocytosis may not be important for movement of the cystic bulge, because FM1-43 staining of cell membrane demonstrates that membrane uptake does not take place around the actin cones, and blocking endocytosis using temperature shift of the shibire (dynamin) mutant does not affect cystic bulge movement. In addition, no concentration of alpha-adaptin has been observed in the region around the actin cones, suggesting that no coated pit formation occurs there. Conversely, clathrin mutants have defects in individualization, but the reason that individualization fails has not been well studied. The discrepancies in these data will only be resolved by further analysis of the clathrin mutant phenotype and studies of the effects of loss of function in other proteins in the endocytosis pathway (Noguchi, 2003).
Likewise, exocytosis may not play a major role in the membrane remodeling process, because membrane staining with the dye FM1-43 suggests that there is not a significant amount of membrane insertion at the sites around the actin cones, and treatment with the the exocytosis inhibitor BFA does not affect the movement of cystic bulge. These data do not completely exclude the possibility that exocytic events participate in remodeling, since the exocytosis could not be directly measured. However, it seems more likely that the large number of membrane invaginations that are present in the cystic bulge is a sufficient source of membrane to accomplish remodeling. The plasma membrane seems to be smoothly reorganized into thin tubular structures around the actin cones. Furthermore, ultrastructural observations have shown that the membrane around actin cones is flat, without any invaginating or docking membrane vesicles. These data support the idea that the cell membrane in the cystic bulge is directly deformed into a thin tubular structure (Noguchi, 2003).
In animal cells, cytokinesis is accomplished by the contractile ring, a transient structure containing actin and myosin II filaments (Zipper) that is anchored to the equatorial cortex. Interactions between these filaments lead to the constriction of a ring that pinches the dividing cell in the middle like an ever tightening purse string until cleavage is completed. Male meiosis was examined in mutants of the chickadee (chic) locus, a Drosophila gene that encodes profilin, a low molecular weight actin-binding protein that modulates F-actin polymerization. These mutants are severely defective in meiotic cytokinesis. Difficulties in meiotic cytokinesis are immediately obvious because of the characteristic appearance of spermatids directly after their formation at the so-called onion stage. Wild-type onion stage spermatids contain a single phase-light nucleus and a similarly sized phase-dark Nebenkern (a mitochondrial derivative). Failures in cytokinesis result in abnormally large Nebenkern associated with multiple normal-sized nuclei. The resulting phenotypes fall into multiple groups: in testes of males homozygous for chic a large fraction of onion-stage spermatids contain a single Nebenkern of larger than normal size, associated with two or more normal-sized nuclei. A substantial proportion have two nuclei with an intermediate-size Nebekern, but most frequently, these aberrant spermatids contain four nuclei and a very large Nebenkern. These phenotypes reflect failures of cytokinesis at either one or the other or both meiotic divisions, respectively, which would prevent proper subdivision of mitochondria and nuclei into daughter spermatids (Giansanti, 1998).
In wild-type testes, phalloidin staining reveals an F-actin-enriched contractile ring that encircles the spindle midzone (the bundle of interdigitated microtubules between the separating chromosome complements during anaphase and telophase) from late anaphase through telophase of both meiotic divisions. In contrast, in many meiotic divisions of chic mutants, no actin staining is visible at the cell equator. In most mutant ana-telophases with equatorial actin staining, only irregular patches of F-actin are observed. Chic protein is found in clear concentrations near the cell cortex, particularly in the equatorial zone. In chic mutants abnormal aggregations of F-actin are observed primarily in premeiotic mature spermatocytes at the S5 stage. These aggregates are enriched in alpha-spectrin and are almost invariably associated with ring canals, suggesting that these actin aggregates are in fact relatively undegraded remnants of the male fusome (see Drosophila Spectrin for more information on fusomes). It is suggested that in the absence of chic the disintegration of the fusome is partially blocked (Giansanti, 1998).
In addition to the absence of the contractile ring, the ana-telophases of all the chic mutants exhibit another obvious abnormality: a defect in central spindle morphology. In wild-type ana-telophases of both meiotic divisions, the two daughter nuclei are connected by a prominent bundle of interdigitating microtubules. In chic mutants, this central spindle structure is considerably less dense than in wild type, and microtubules show very little or no interdigitation. In wild type, the central spindle is already evident by mid-anaphase, before the actin ring constriction has caused substantial equatorial pinching. In the strongest chic mutants, these early stages of central spindle formation are never seen, indicating that the chic phenotype reflects a failure of central spindle assembly rather than a degradation of this structure in the absence of the actin ring. This result is surprising because the central spindle is a tubulin based cytoskeletal structure. There is substantial evidence that the central spindle is required for structuring of the actin based contractile ring but not similar evidence that the contractile ring is required for building the central spindle. Together, these observations indicate that chic mutations disrupt two major cytokinetic structures: the microtubule-based central spindle and the actomyosin contractile ring (Giansanti, 1998).
In wild-type primary spermatocytes during the prophase-prometaphase transition of the first meiotic division (stage M1), centrosomes migrate from a position just under the plasma membrane to the nuclear envelope, where they nucleate prominent asters. The two asters then separate and move around the periphery of the nuclear membrane, so as to establish a bipolar spindle. Similarly, during late telophase of the first meiotic division and the short interphase between meiosis I and meiosis II, asters separate and migrate to the opposite poles of secondary spermatocytes. Although the asters in chic mutants are improperly positioned by the start of prometaphase, relatively normal-looking bipolar spindles eventually form by late metaphase. It is remarkable that the function of these spindles, with respect to chromosome segregation, appears to be largely unimpaired. Only a very low frequency of onion-stage spermatids contain irregularly sized nuclei, such as micronuclei (Giansanti, 1998).
Lesions in twinstar (tsr), a gene encoding a Drosophila cofilin (an actin severing and depolymerizing protein), cause a syndrome of phenotypic effects that have both similarities and differences to those described above for chic mutations. In tsr, as in chic spermatocytes, centrosome separation and migration are defective, abnormal accumulations of F-actin are apparent, and cytokinesis often fails after one or both meiotic divisions. tsr and chic mutant phenotypes, however, can be easily distinguished on the basis of several criteria. In tsr mutants, the central spindle is normal, and the contractile ring still forms (though in misshapen and enlarged form), in contrast to the situation for chic. The types of F-actin aggregates formed in tsr spermatocytes are much different from those in chic. It is concluded that tsr and chic mutations differ in how they interfere with meiosis (Giansanti, 1998).
To further investigate the relationships between the central spindle and the contractile ring, meiosis was examined in the cytokinesis-defective mutants KLP3A and diaphanous. The KLP3A gene encodes a kinesin-like protein that accumulates in the central spindle midzone during anaphase and telophase of both meiotic divisions. Accordingly, mutations in this gene disrupt central spindle formation and cause frequent failures in meiotic cytokinesis. To check whether the defect in central spindle integrity observed in KLP3A mutants also affects actin ring assembly, KLP3A mutant testes were stained with rhodamine-labeled phalloidin. The results of this experiment clearly show that most mutant ana-telophases (90%) are completely devoid of actin rings. The rare ana-telophases that exhibit thin and incomplete actin rings also contain more densely packed central spindles than those of cells completely lacking contractile rings. Despite the absence of the contractile ring, KLP3A mutants do not exhibit aberrant actin accumulations or problems in aster migration like those described above for chic and tsr mutants (Giansanti, 1998).
The diaphanous gene encodes a protein that interacts with profilin through its proline-rich domain. All the ana-telophases present in testes homozygous for dia mutants are completely devoid of actin rings. It is of interest that these figures also show severe defects in the central spindle, similar to those observed in chic and KLP3A. The effects on the actomyosin contractile ring and the central spindle observed in chic and dia mutants could be specific consequences of lesions in the corresponding gene products. Alternatively, these effects could result from a more general disruption of the actin cytoskeleton. To discriminate between these possibilities, wild-type testes were treated with cytochalasin B prior to fixation and staining. Cytochalasin B binds the barbed ends of actin filaments and promotes the conversion of ATP-actin monomers to ADP-actin , preventing proper assembly of the contractile ring in most cell types. Remarkably, incubation with this drug produces an almost exact phenocopy of strong chic alleles. No F-actin staining is observed in any contractile ring-like structures at the equator of ana-telophase cells. (Giansanti, 1998).
In all cases examined, the central spindle and the contractile ring in meiotic ana-telophases were simultaneously absent. Together, these results suggest a cooperative interaction between elements of the actin-based contractile ring and the central spindle microtubules: when one of these structures is disrupted, the proper assembly of the other is also affected. In addition to effects on the central spindle and the cytokinetic apparatus, another consequence of chic mutations was observed: A large fraction of chic spermatocytes exhibit abnormal positioning and delayed migration of asters to the cell poles. A similar phenotype was seen in testes treated with cytochalasin B and has been noted previously in mutants at the twinstar locus. These observations all indicate that proper actin assembly is necessary for centrosome separation and migration, and that the central spindle and the contractile ring are interdependent structures (Giansanti, 1998).
The best candidate at present for mediating interactions between the central spindle and cortical actin, at least during male meiosis, is the KLP3A kinesin-like protein. This protein could interact directly with both the central spindle microtubules and components of the contractile ring. Alternatively, KLP3A could transport to the spindle midzone molecules that mediate F actin-microtubule interactions. At the moment, it is not possible to discriminate between these possibilities, nor is there any information on the proteins that bind to or might be transported by KLP3A. It is believed, however, that the isolation and characterization of additional mutations causing cytological phenotypes similar to those of KLP3A, chic, and dia, will eventually provide substantial insight into the mechanisms underlying microtubule-actin interaction during cytokinesis (Giansanti, 1998).
The molecular identity and function of the Drosophila melanogaster Y-linked fertility factors have long eluded researchers. Although the D. melanogaster genome sequence has recently been completed, the fertility factors still are not identified, in part because of low cloning efficiency of heterochromatic Y sequences. A method for iterative BLAST searching has been used to assemble heterochromatic genes from shotgun assemblies, and kl-2 and kl-3 have been identified as 1-beta and gamma-dynein heavy chains, respectively. These conclusions are supported by formal genetics with X-Y translocation lines. Reverse transcription-PCR was successful in linking together unmapped sequence fragments from the whole-genome shotgun assembly, although some sequences were missing altogether from the shotgun effort and had to be generated de novo. A previously undescribed Y gene, polycystine-related (PRY), was also found. The closest paralogs of kl-2, kl-3, and PRY (and also of kl-5) are autosomal and not X-linked, suggesting that the evolution of the Drosophila Y chromosome has been driven by an accumulation of male-related genes arising de novo from the autosomes (Carvalho, 2000).
The discovery that the Y chromosome of Drosophila melanogaster contains genes essential only for male fertility dates back to the birth of Drosophila genetics and the theory of chromosomal inheritance. In 1929, Stern showed that these genes are localized in both the short (YS) and long (YL) arms of the Y chromosome, and in 1960, Brosseau used x-ray-induced mutations to identify seven complementation groups, two in YS (ks-1 and ks-2) and five in YL (kl-1 to kl-5). In 1981, Kennison obtained fertile X-Y translocation lines and used them to construct males with deletions in each of the fertility factors. With these lines, Kennison confirmed six of the seven fertility factors previously identified by Brosseau (kl-4 was not confirmed). The same lines allowed a more precise identification of the defects associated with the lack of each of the fertility factors. In particular, the lack of kl-3 or kl-5 causes the loss of the outer arm of the sperm tail axoneme, a structure known to contain the molecular motor protein dynein in other organisms. Indeed, in 1982, Goldstein showed that sperm from kl-3- and kl-5- (and also kl-2-) males lack three discrete high molecular weight proteins with mobility similar to dynein heavy chains of Chlamydomonas reinhardtii and proposed that these fertility factors are the structural genes of three different dynein heavy chain proteins. In 1993, Gepner and Hays sequenced part of kl-5 and showed that it encodes an axonemal beta-dynein heavy chain that is expressed in the testis (Carvalho, 2000 and references therein).
Axonemal dynein heavy chains are known to be responsible for the beating of flagella and cilia, which explains why kl mutants produce immotile sperm. There are several isoforms of axonemal dynein heavy chains (alpha, beta, gamma, 1beta, 1alpha, etc.) that associate to form the inner and outer arms of the axonemes. D. melanogaster has at least seven other dynein heavy chain genes, scattered in chromosomes X, 2, and 3 (Carvalho, 2000 and references therein).
Another important experimental breakthrough was the development of a method to discern banding patterns in Drosophila heterochromatin, which allowed the first detailed cytogenetic investigation of the Y chromosome. Gatti and Pimpinelli (1983) identified 25 heterochromatic bands on the Y and mapped the fertility factors to these bands. It became clear that some of the fertility factors, including kl-5, are unusually large [~3 megabase (Mb)]. The paradox of a conventional coding gene (e.g., kl-5), spread over a huge amount of DNA was solved by Bünemann and coworkers: in the kl-5 homolog of Drosophila hydei, some of the introns are gigantic (>1 Mb) and most likely account for the unusual size of the gene. These introns are composed of short repetitive sequences and satellite DNA. These key discoveries trace back to the extensive work on lampbrush Y chromosomes initiated by Meyer and coworkers in 1961 (Carvalho, 2000 and references therein).
As can be seen from the above summary, the progress on the identification of Y-linked genes has been very slow. This slow progress is mainly a consequence of the technical difficulties caused by the heterochromatic state of the Y chromosome, and most of the experimental breakthroughs mentioned above actually are ingenious ways to implement standard tools used for euchromatic genes in heterochromatin. The Y chromosome does not recombine during meiosis, preventing classical genetic mapping; this problem was solved by Kennison's lines. It does not undergo polytenization, making cytogenetic studies more difficult [solved by Gatti and Pimpinelli]. P element mutagenesis was also more difficult, because the common markers are often silenced when inserted in the Y, but now there are special P constructs that make it possible to overcome this limitation (Carvalho, 2000 and references therein).
The recent sequencing of the Drosophila genome might have yielded the final solution, but again the heterochromatic nature of the Y chromosome posed special difficulties. Most heterochromatin is composed of short repetitive sequences that are not stable in the vectors used in sequencing projects. Thus, despite comprising nearly 30% of the genome, heterochromatic sequences account for only 2% of the sequence reads. Furthermore, its repetitive nature does not allow the assembly of the individual sequence reads (~500 bp) into larger scaffolds, and these into complete chromosome arms. As a result, only 15 kb (a small portion of the kl-5 gene) have been assigned to the Y chromosome, whereas essentially all of the 120 Mb of the euchromatin have been assembled into chromosomes X, 2, 3, and 4. Besides these mapped sequences, 631 scaffolds (ranging from 1 kb to 64 kb, and totaling ~4 Mb of sequence) remain unmapped. These unmapped scaffolds most likely contain pieces of heterochromatic genes, including Y-linked ones (Carvalho, 2000 and references therein).
Of special interest in the study has been the inventive use of iterative BLAST searching for assembly of heterochromatic sequences. The unmapped Drosophila scaffolds (called "armU" in Celera's CD-ROM release of the Drosophila genome) were downloaded from ftp://ncbi.nlm.nih.gov/genbank/genomes/D_melanogaster/, and then an armU database was built by using the FORMATDB program of the STANDALONE BLAST. In this way BLAST searches could be restricted to the set of unmapped scaffolds. In addition to STANDALONE BLAST, extensive use was made of the programs WWWSTANDALONE BLAST (Linux version), NETBLAST, REPEATMASKER (available at http://repeatmasker.genome.washington.edu/cgi-bin/RepeatMasker), and NAP and GAP2 (available at http://genome.cs.mtu.edu/sas.html). BLAST programs were downloaded from the National Center of Biotechnology Information (http://www.ncbi.nlm.nih.gov/) (Carvalho, 2000 and references therein).
To see how a Y-linked gene appears in the armU sequences, the complete cDNA of kl-5 (Genbank nucleotide record: AF210453) was used as a query sequence in a BLASTN search against the armU database. In addition to the fragment already identified (CG17616 gene in the Genbank AE002688 scaffold), most of the kl-5 gene was retrieved, scattered across five scaffolds. The majority of these scaffolds contain complete exons (the exception is AE003233, which begins in the middle of an exon). Their 3' ends contain the 5' splice junctions and a variable portion of downstream intronic sequences, often ending with simple repetitive sequences. The 5' ends of scaffolds in armU have analogous structures. Some exons are missing altogether in armU. These observations fit well with the expected behavior of a gene like kl-5 in whole genome shotgun (WGS) projects: exons define unique, nonrepetitive sequences that will be cloned regularly and will be assembled into at least a small scaffold in the end of WGS. Normally sized introns will be readily cloned and assembled along with exons. Indeed, most of the kl-5 scaffolds contain several exons and the intervening short introns. However, some introns of kl-5 probably contain Mega base-sized blocks of repetitive DNA that cannot be assembled by WGS or any other available method. These fragments will rarely be cloned and sequenced and, even if sequenced, would not be assembled into a scaffold. In short, during WGS, a gene like kl-5 will be chopped into several pieces, delimited by the unclonable intronic satellite DNA. Most of the time, a gene immersed in heterochromatin will go undetected by the normal 'first pass' annotation procedures (which rely on gene prediction tools and BLASTX with high stringency), because these methods will work poorly with individual exon sequences. However, the whole gene may be retrieved if a suitable query sequence is available (the kl-5 cDNA in this case) to identify and align its pieces. Very small exons embedded in large introns will most likely be lost during the WGS, and this probably explains the lack of some kl-5 exons (Carvalho, 2000).
To identify other fertility factors in armU, protein sequences were used as query sequences. Suitable proteins were chosen as follows. First the 631 scaffolds of armU were filtered with REPEATMASKER and a BLASTX search of each of them was performed against the nr database (all known proteins, including putative ones) with a rather high stringency (e = 10-4). There were proteins that gave hits in hundreds of scaffolds; most of them are reverse transcriptases, copia polyprotein, etc., and most likely are matching transposable elements of armU that 'escaped' REPEATMASKER. Some other proteins have hits in a few scaffolds; these are homologs of prospective Y-linked genes, chopped in pieces as kl-5. Each of these prospective proteins was used as a query sequence (as was done with the cDNA of kl-5), running TBLASTN with a lower stringency (e = 10) against the armU database. A staggered pattern results from the very large introns of Y-linked genes. Two such cases are myosin VII (AAF06035, from Dictyostelium discoideum) and gamma-dynein heavy chain, from C. reinhardtii], and were investigated further. All tested scaffolds relating to myosin VII proved to be not Y-linked (i.e., PCR produces bands when either male or female DNA is used as the template), whereas most of the dynein-related were Y-linked. Several of the Y scaffolds were identified by using gamma-dynein heavy chain (Genbank record: Q39575) as a TBLASTN query sequence with a low stringency (sometimes e = 1,000), in an attempt to retrieve missing exons. Because there is a big overlap among several of the dynein-related scaffolds, most likely two different Y-linked dynein heavy chain genes were found (Carvalho, 2000).
Genomic DNA from Y deficient males (kl-1-, kl-2-, etc.) was used in PCR to map each of the Y-linked scaffolds identified in the previous step. It should be noted that this procedure assigns a given scaffold to a region of the Y chromosome (e.g., the kl-2 region), but it does not necessarily imply that this scaffold belongs to the actual fertility gene. This distinction is important, because a given region may contain more than one gene. For the sake of simplicity the regions of the Y are referred to by the name of the respective fertility factor they carry (Carvalho, 2000).
Several exons of kl-2 and kl-3 genes are missing from the BLAST results. These missing exons may be absent in armU sequences (as happened with kl-5) or may have diverged enough to be no longer identified by these methods. RT-PCR was used to obtain the sequence of these missing exons and to check whether the Y sequences that were detected are expressed. RT-PCR sequences were obtained from all splice junctions between adjacent scaffolds so that they could be precisely identified. The sequencing of the gaps revealed several previously missed armU scaffolds. In kl-2, some 330 codons of the N terminus are still missing. AE003086 filled the gap between AE003157 and AE002962. There is no sequence gap between AE002962 and AE003049. AE002706 filled a small portion of the gap between AE003049 and AE003219; the remaining 2 kb were sequenced and it was found that it is entirely missing in armU. The AE003219 scaffold contains five internal, short introns and extends through the stop codon. Regarding kl-3, AE002577 and AE002776 appear to be spurious matches caused by running TBLASTN with low stringency, because no RT-PCR product could be recovered that includes these sequences. The gap between AE002917 and AE002920 was sequenced; the 948-bp sequence is missing in armU. Finally, some 230 codons in the C terminus seem to be missing. Each of the three big kl-3 scaffolds contains one internal intron (Carvalho, 2000).
The internal introns were identified and localized with the NAP program, which aligns genomic DNA with proteins allowing for GT/AG bounded gaps (in the case presented here, armU scaffolds and the gamma-dynein Genbank sequence Q39575 were aligned). RT-PCR sequences surrounding each putative intron were obtained and aligned with the corresponding armU scaffold with the GAP2 program, which aligns genomic DNA with cDNA, again allowing for GT/AG bounded gaps. Almost all putative introns suggested by NAP were confirmed, although the inferred splice junctions frequently were not precise. Seven frame-shift sequence errors in armU sequences were pinpointed by NAP and BLASTX and were corrected by sequencing (Carvalho, 2000).
The assembled cDNA of kl-2 and kl-3 were deposited in GenBank under the accession numbers AF313479 and AF313480. Polycystine-related (PRY) is a putative, previously unidentified Y-linked gene. During attempts to retrieve missing exons two armU scaffolds were found that map to the kl-5 region but have no similarity with the kl-5 cDNA. AE002774 seems to contain only two short pieces of transposable elements and was not further investigated. AE003011 showed a strong similarity with the product of a putative gene localized in chromosome 2 (AAF44887) and also a weaker similarity with the human polycystine protein (AAD18021). Interestingly, polycystine is similar to the sea urchin sperm receptor for egg jelly (AAB08448). Using the Drosophila hypothetical protein AAF44887 as a query sequence in TBLASTN (against armU sequences) another closely related scaffold, AE003212 was recovered. RT-PCR closed the gap between it and AE003011; thus, they most likely are part of a previously unidentified expressed Y-linked gene, which is currently being sequenced. Surprisingly, AE003212 maps to the kl-3 region. These findings imply that the breakpoint of the V24 translocation (the h4 band) cuts the PRY gene in the middle. Thus, V24 is defective for PRY -- a close examination of this line may give some clue about the function of this gene. Because the kl-3 and kl-5 regions are known to contain factors (other than the dyneins) that cause sterility when present in three copies, it is possible that PRY is responsible for this phenotype (Carvalho, 2000).
A simple method has been described for tailoring BLAST searches in such a way that poorly assembled fragments from WGS projects may reveal genes embedded in heterochromatin. The method relies on TBLASTN searches (instead of the more usual BLASTX) to identify putative heterochromatic genes by the distinct staggered pattern they produce. It was applied and the kl-2 and kl-3 genes on the Y chromosome of D. melanogaster were successfully identified. The sequences that were found are long, transcribed open reading frames that encode dynein heavy chain polypeptides. Thus the quest for the Y dyneins initiated by Hardy is now finished, and Drosophila Y gene hunting may now move to the even more exciting task of identifying the mysterious kl-1, ks-1, and ks-2 (Carvalho, 2000).
Heterochromatic genes are islands of unique sequence and appear in the end of WGS as isolated scaffolds that could not be assembled into chromosomes. If introns are large enough and contain heterochromatic repeat sequences, they will be sufficiently underrepresented in the WGS sequences to disrupt the assembly of flanking scaffolds. As a result, exons of the same gene are scattered in several unmapped scaffolds ('armU'), generating a staggered pattern in TBLASTN and BLASTN searches. This pattern will occur not only in the Y-linked genes but also in autosomal heterochromatin, as in the case of the Drosophila myosin VII homolog. Although heterochromatic genes pose special problems for genome sequencing, it is possible that they have an advantage over euchromatic genes: once the euchromatic sequence of a whole chromosome arm (a Mb-sized, unique sequence scaffold) is obtained, it may be very difficult to detect all of the genes it contains, whereas it is possible that a significant proportion of the unique heterochromatic sequences represents functional genes. It seems that only functional sequences resist the forces that fill heterochromatic regions with short repeats and thus remain clonable, unique sequences. During this project, 38 armU scaffolds were examined, and at least 12 of them (~1/3) seem to be part of genes. Of course, this sample is not random, but it strongly suggests that the small isolated scaffolds remaining at the end of WGS projects may be a good source of interesting genes. Indeed, it is an advantage of the WGS approach over clone-based strategies that, besides the euchromatin, it also retrieves unique sequence heterochromatin, no matter how deeply the sequence is located within the heterochromatin (Carvalho, 2000).
A striking pattern emerges from the phylogeny of the Y dyneins: they all are closely related to other Drosophila genes, but none of these paralogous genes is X-linked. The same pattern occurs with PRY. Furthermore, the Drosophila X chromosome contains only one dynein heavy chain (Dhc 16F), in contrast with the three Y-linked ones. Thus, it seems likely that these genes were acquired from autosomes, rather than being present in the hypothetical chromosome pair that gave rise to the X and Y. This mechanism has been demonstrated for the mammalian Y, but in that case the Y chromosome also exhibits a number of X-derived genes. Another Drosophila Y-linked gene, Su(Ste), has been shown to be recently originated from an autosomal gene. Repetitive sequences also do not show any sign of X-Y homology; it has been proposed that much of the Drosophila Y is virtually a new construct, rather than a degenerated X. The data presented in this study clearly support this hypothesis. It remains to be seen whether any part of the ancestral Y was homologous to the X (as may be the case for rDNA genes, which are present in both X and Y chromosomes) or whether it is a totally new construct. This picture of the Drosophila Y may change if other, yet unidentified Y genes (kl-1, ks-1, ks-2, etc.) turn out to have X homologs. Whatever its origin, the present configuration of the Drosophila Y chromosome seems to be quite old, for at least kl-5 is present also in D. hydei and Drosophila mediopunctata, which diverged from D. melanogaster ~39 million years ago. A few Drosophila species have fertile X0 males; it will be most interesting to study the location of their axonemal dynein heavy chain genes (Carvalho, 2000).
The absence of X homologs and the close similarity between Y and autosomal genes suggest that the former is an agglomeration of autosomal genes. This hypothesis is the most parsimonious and explains well the kl-5, kl-3, and PRY cases. However, it is also possible that Y chromosomal genes have transposed to the autosomes, and this possibility might explain the kl-2 case. The closest paralogs of kl-3 and kl-2 are the CG9492 and CG9068 genes, respectively. Dynein heavy chains have ~4,500 amino acids, whereas CG9492 and CG9068 are shorter (3,508 and 1,227, respectively) and seem to lack the C terminus. The former case results from a misannotation: BLASTX and NAP identified all of the missing ~1,000 amino acids of CG9492 (including the stop codon at position 188,944 in the AE003683 scaffold). However, CG9068 seems to be truncated, for no sign of the 'missing' C terminus could be found. Therefore, the relationship between CG9068 and kl-2 is unclear; it is possible that kl-2 originated from CG9068 and that after this the latter suffered a deletion. However, it is also possible that CG9068 results from a partial transposition (perhaps being a pseudogene) of kl-2 (Carvalho, 2000).
The phylogeny of the dyneins strongly suggests that kl-2 encodes a 1beta-dynein, whereas kl-3 encodes a gamma-dynein. This phylogeny fits well with the known mutant phenotypes of kl genes and with the function of dynein heavy chains; kl-3- mutations (but not kl-2-) disrupt the outer arms of axonemal microtubules, and gamma-dyneins are part of these structures. 1beta-dyneins are part of the inner arms, and it remains to be explained why kl-2- mutants do not show cytological defects (Carvalho, 2000).
It has been noted that the human Y chromosome exhibits a 'functional coherence'; besides housekeeping genes, many Y genes have male-related functions, which contrasts with the random content of the other chromosomes. It is striking that the Drosophila Y has an even stronger coherence, approaching obsession; all known fertility factors (kl-2, kl-3, and kl-5) encode proteins belonging to the same gene family (axonemal dynein heavy chain). This extreme functional coherence, coupled with the lack of X homologs (which might provide an 'historical' cause), begs for an explanation (Carvalho, 2000).
Theoretically, the Y chromosome is expected to accumulate male-related genes; male-female antagonistic effect of genes may hamper the evolution of male-related traits, unless they are located in a male-specific region of the genome. This prediction has been demonstrated experimentally, and the findings presented here support it. Regarding the particular male fitness trait involved, the most likely advantage conferred by sperm axonemal motor proteins is sperm competitive ability. The PRY gene may also be involved in sperm competition if it has a function similar to its homolog in sea urchin. Drosophila females mate several times; thus, there is ample room for sperm competition, and clearly there is genetic variation for this trait. It is proposed that the evolution of the Drosophila Y chromosome has been driven by an accumulation of male-related genes, most likely caused by sperm competition. This hypothesis explains the puzzling finding of a Y chromosome packed with motor proteins which are absent in the X chromosome. The large element of chance involved in the occurrence of the appropriate translocations probably explains the apparent incompleteness of the process, that is, outer arms are composed of alpha-, beta-, and gamma-dyneins, but only beta and gamma got Y counterparts (Carvalho, 2000).
The hypothesis that natural selection has driven an accumulation on the Drosophila Y of genes related to sperm function may be tested in several ways. (1) Studies designed to quantify Y-linked variation in sperm competition are clearly needed. (2) The comparative method of looking for dynein heavy chain genes in other Diptera (including species with fertile X0 males) may reveal the intermediate steps of the birth of dynein-packed Y chromosomes. (3) the identification of the other fertility factors may yield more clues about the forces shaping Y chromosome evolution in Drosophila (Carvalho, 2000).
Carvalho, A. B., Lazzaro, B. P. and Clark, A. G. (2000). Y chromosomal fertility factors kl-2 and kl-3 of Drosophila melanogaster encode dynein heavy chain polypeptides. Proc. Natl. Acad. Sci. 97: 13239-13244. Medline abstract: 20542110
Gatti, M., Pimpinelli, S. (1983). Cytological and genetic analysis of the Y chromosome of Drosophila melanogaster. Chromosoma 88: 349-373
Giansanti, M. G., et al. (1998). Cooperative interactions between the central spindle and the contractile ring during Drosophila cytokinesis. Genes Dev. 12(3): 396-410. Medline abstract: 98119836
Noguchi, T. and Miller, K. G. (2003). A role for actin dynamics in individualization during spermatogenesis in Drosophila melanogaster. Development 130: 1805-1816. Medline abstract: 12642486
Genes Involved in spermatogenesis
Genes involved in organ development
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
Please e-mail comments/corrections to brodyt@codon.nih.gov