Sequencing ESTs made possible the generation of intron-exon maps for achi/vis and revealed alternative spliced products from the two transcription units. The major difference between splicing products is the presence or absence of exon6, which is 387 bp in length and encodes an additional 129 aa. The expression profiles of these transcripts were investigated by RT-PCR. The achi/vis transcripts are present from embryogenesis through to adulthood. Interestingly, adult males predominantly expressed the larger (class 2) splice variant and females predominantly the smaller (class 1). Using RNA derived from gonadectomised males compared to RNA derived from testes it was determined that the larger splice isoform was testis specific, whereas the shorter was present in both the carcass and the testis. Sequencing of subclones of this testis-specific large isoform confirm that both achi and vis class 2 transcripts are expressed in the testis (Ayyar, 2003).
In keeping with the meiotic arrest phenotype, RNA in situ analysis of achi/vis expression shows strong expression in the primary spermatocyte stage, which decays as the cells progress through meiosis. There is no detectable expression in the tip of the testis where the stem cells and spermatogonial cells reside (Ayyar, 2003).
To analyze the vis and achi expression patterns a polyclonal antibody was generated against AchiS. Using this antibody, nuclear expression is detected in nearly all cells and stages of development, including cells in the testes. The specificity of this antibody was confirmed by observing only background staining in the pingpong deficiency (deficient for both vis and achi). Moreover, a wild-type testes staining pattern is generated by either the P{vis} or P{achi} transgenes demonstrating that this antibody recognizes both Vis and Achi proteins (Wang, 2003).
Because of the male sterile phenotype of homozygous pingpong flies the Vis/Achi expression pattern was examined in more detail in wild-type testes and its pattern was compared to that of Aly, a chromatin-associated protein that is required for male meiosis in Drosophila (White-Cooper, 2000). At the most apical tip of the testes are somatic 'hub' cells, germline stem cells, and mitotically dividing spermatogonia. Further from the tip is a zone of larger primary spermatocytes organized into 16-cell cysts. These cells, which grow in volume ~25-fold over a period of ~3.5 days, remain in meiotic prophase, a specialized stage of the meiotic cell cycle that precedes the meiotic divisions. Like Aly (White-Cooper, 2000), Vis/Achi is most strongly observed in the nuclei of primary spermatocytes. In addition, weaker nuclear Vis/Achi staining is also observed in the mitotically dividing cells at the apical tip of the testes. This is not a background staining because it is not observed in the pingpong deficiency, which still have these cells. In addition, weak Aly staining is also observed at this stage (Wang, 2003).
The various stages of meiosis can be identified by the state of the chromosomes, as revealed by a fluorescent DNA dye such as propidium iodide. At the large primary spermatocyte stage, when both Aly and Vis/Achi are nuclear, the three major chromosome pairs can be seen as three separate but diffuse signals, indicating that the chromosomes are partially condensed. At metaphase of meiosis I the chromosomes are fully condensed and appear as nuclear dots. At this stage, Aly appears to associate with the spindle and Vis/Achi is predominantly in the cytoplasm. A similar pattern is seen for both proteins at metaphase of meiosis II. Following the second meiotic division spermatid differentiation begins with the onion stage, in which each cell has a characteristically round and specialized mitochondrion adjacent to its haploid nucleus. At this stage, the anti-Aly antibody labels a single dot per nucleus that co-localizes with the DNA. In contrast, Vis/Achi is observed throughout these cells except that it is excluded from the large and specialized mitochondrion next to each nucleus. Finally, at an even later stage of spermatid differentiation, Aly is barely detected above background and Vis/Achi partially co-localizes with the DNA. Vis/Achi is not detected in the somatic cyst cells that surround the primary spermatocytes (Wang, 2003).
To examine the in vivo role of Drosophila TGIF deletions were generated by imprecise excision of a P-element that mapped within the 5'-UTR of the achi transcription unit. Several small deletions were obtained together with a larger deletion, Df(2R)achi1, that removes the whole of the achi transcription unit and part of the vis unit. In addition, using P-element induced male recombination, several large deletions were produced that removed achi and vis together with some neighboring genes. The largest of these deletions Df(2R)achi4 is homozygous lethal. In contrast the smaller deletions Df(2R)achi2 and Df(2R)achi3 have the same phenotype as Df(2R)achi1: they are homozygous viable, the males are completely sterile and females crossed to wild-type males exhibit a delay in egg laying. All three deletions failed to complement Df(2R)BSC3 with respect to the male-sterile phenotype (Ayyar, 2003).
Examination of the testes from homozygous mutants revealed complete absence of mature sperm. Developmental stages up to late primary spermatocyte were present but no meiotic stages could be seen. A mutant with the same phenotype, Z3922 was identified in a screen of EMS mutants. The mutation in Z3922 was mapped by meiotic recombination to 63.3±1.8 mu. The Z3922 chromosome failed to complement Df(2R)BSC3, placing the mutation in the 48F-49A chromosomal region. Fine scale deficiency and recombination mapping relative to P-element insertions in this region located the Z3922 locus to within 7 genes immediately distal to the P-element insertion P{w+}l(2)k17040. PCR sequencing of these candidate genes from Z3922 revealed mutations in both achi and vis; there is a premature stop in exon 5 of achi together with a 56 bp deletion just 5' of the vis homeodomain. The Z3922 chromosome (achiZ3922 visZ3922) fails to complement the Df(2R)achi1 allele (Ayyar, 2003).
The effects of the achi/vis mutations on spermatogenesis were examined in more detail. Labelling the DNA with DAPI provides an easy way to visualize the stages of spermatogenesis. In wild-type whole mounts, cells at the apical tip of the testis label strongly, but during the primary spermatocyte stage the intensity decreases correlating with chromosomal reorganization and increasing cell size. An expanded zone of high DAPI labelling was observed in Df(2R)achi1 mutant testes. This experiment was repeated using confocal microscopy with anti-histone labelling to study the chromatin morphology. The expanded zone of cells with higher DAPI labelling corresponds to an expanded population of small primary spermatocytes with diffuse chromatin surrounding a prominent nucleolus (Ayyar, 2003).
Squashed preparations of cells from the testes were used to study the defects in greater detail using phase contrast microscopy and DAPI labelling. Spermatocytes from stages S1 to S6 were identified in Df(2R)achi1 mutant testes. However, a marked defect was observed in the chromatin of the mature primary spermatocytes (stages S5 and S6). While these cells were large and had prominent nucleoli characteristic of this stage, the chromatin was condensed into tight central blobs normally observed only in later meiotic cells. Hence, in these mutants there appears to be an uncoupling between the nuclear/nucleolar differentiation and the chromatin condensation events of meiosis. Similar results were observed comparing wild-type and mutant histone-labelled preparations (Ayyar, 2003).
TGIF has been extensively characterized as a transcriptional repressor in vertebrates, yet in Drosophila the predominant effect seen in achi/vis mutants is failure to activate a testis-specific developmentally regulated transcriptional program. In Drosophila, achi/vis must be acting either directly as a transcriptional activator, or indirectly, as a repressor of a repressor. To investigate the transcriptional effects of lack of achi/vis in more detail, RNA expression profiling was undertaken, looking particularly for genes whose expression was increased in the mutants. This system is well suited to expression analysis since a viable infertile phenotype allows the easy generation of mutant tissue for comparison with wild type. Preliminary microarray analysis, comparing RNA from Df(2R)achi1 mutant testes to wild-type, has demonstrated, as expected, a large number of transcripts (including the spermiogenesis genes) showing decreased expression relative to wild-type. However, many transcripts were also observed with increased levels in the mutant. The strongly increased transcripts of extra sex combs (esc), in the Df(2R)achi1 mutant testis, was particularly intriguing and esc was further investigated as a possible candidate for repression by achi/vis. The microarray result was supported by RT-PCR comparison of esc transcript in wild type and Df(2R)achi1. Esc is a component of the transcription silencing machinery and hence provided a possible link between Drosophila TGIF and gene repression. If TGIF normally represses esc transcription in primary spermatocytes this might allow the activation of the spermatogenesis transcription program. According to this model TGIF mutants would overexpress esc and this would prevent the program activation (Ayyar, 2003).
The interpretation of altered levels of transcripts in the mutant RNA population is complicated by the gross disruption of the RNA and cellular content of the mutant testes due to the failure of spermatogenesis at the primary spermatocyte stage. The expression of esc was therefore investigated in the testes at a cellular level using an antibody against Esc protein. This analysis did not support the model proposing repression of esc by TGIF. In wild-type testis, Esc is expressed in the early mitotic cells and was also robustly expressed in early primary spermatocytes; clearly esc expression is not normally switched off in primary spermatocytes by achi/vis. Esc immunolocalization also revealed the unexpected observation that as the primary spermatocytes mature, the Esc labelling, at first rather evenly distributed in the nucleus (but excluded from the nucleolus) progressively accumulates in nuclear spots. These nuclear spots are highly reminiscent of the labelling of a variety of functional silencing complexes in various cell types. Levels of Esc protein diminish as wild-type primary spermatocytes mature (Ayyar, 2003).
In the achi/vis mutant testes the overall expression levels of Esc protein are similar to wild type. How then can the increase in esc transcript level be explained? Esc protein level appears highest in early primary spermatocytes and this cell population is markedly expanded in achi/vis mutants. It is thought that expansion of cells with the highest level of esc is the likely cause of the observed increase in esc transcript abundance (Ayyar, 2003).
Despite no change in the apparent level of Esc protein in mutant cells, the Esc localization was strikingly altered in Df(2R)achi1 mutant testes. Although in achi/vis mutants the primary spermatocytes apparently differentiate to the final primary spermatocyte stage, the concomitant accumulation of Esc in nuclear spots fails to occur. It appears achi/vis function is required for assembly of the Esc complexes (Ayyar, 2003).
vis and achi are located on the right arm of chromosome 2 in the Drosophila genome, in cytological position 49A. Starting with a P element [EP(2)2107] inserted close to the 5' end of achi, P element-mediated male recombination was used to generate a ~40 kilobase deletion that removes both vis and achi. Because this deficiency deletes two nearly identical genes and results in male sterility, it was named Df(2R)pingpong and is referred to simply as pingpong. In addition to deleting vis and achi, pingpong also removes four additional genes. However, homozygous flies carrying this deficiency are viable, suggesting that none of these genes are required for embryonic or larval development. Moreover, pingpong homozygous females are fertile and can give rise to viable pingpong progeny, eliminating the possibility that any of these genes have an essential maternal function. The only highly penetrant phenotype observed in pingpong homozygous flies is male sterility, which can be rescued by transgenes carrying genomic regions for either vis (P{vis}) or achi (P{achi}. Thus, it is concluded that vis and achi encode redundant functions required for male fertility (Wang, 2003).
The role of vis and achi in male fertility was quantitated by comparing the fertility of pingpong males with pingpong; P{vis} and pingpong; P{achi} males. Individual males were crossed to wild-type females and the number of progeny after 15 days was counted. In this assay wild-type (yw) males yielded an average of 132 progeny/male. pingpong males were completely infertile. In contrast, pingpong; P{achi} males were nearly as fertile as wild type and pingpong; P{vis} males were partially rescued. Although both P{vis} and P{achi} appear to generate a wild-type Vis/Achi expression pattern in the male germline these results suggest that achi is better able to rescue the mutant phenotype than vis (Wang, 2003).
The male sterile phenotype, together with its strong nuclear expression in primary spermatocytes, suggested that vis or achi may be required for meiosis. Therefore pingpong and pingpong; P{achi} testes were examined by phase contrast microscopy, staining the chromosomes with propidium iodide. In contrast to wild type, pingpong mutant testes are much smaller and are blocked before the first meiotic division. Specifically, no evidence of any elongated or onion stage spermatids is observed in the pingpong mutant. Instead, the testes are filled with cells that appear to remain at the primary spermatocyte stage. 16-cell cysts are still present, suggesting that the four mitotic divisions proceed normally in this mutant. However, the cells are smaller and not as round as in wild type. Far from the apical tip of the testes cells appear to degenerate. In addition, some of the chromosomes fail to fully condense in the absence of Vis and Achi. Typically, three spots of DNA are observed per cell, but the appearance of these spots ranges from diffuse (partially decondensed) to fully condensed. This phenotype indicates that the normally synchronized events leading to chromosome condensation fail in the pingpong mutant. Furthermore, these results suggest that although the pingpong mutant initiates meiosis in the male, meiosis is blocked before the first meiotic division, probably prior to the G2 to M transition. In addition to being blocked in meiosis, pingpong testes do not show any signs of spermatid differentiation, such as the distinctive onion stage cysts or spermatid elongation (Wang, 2003).
Bromodeoxyuridine (BrdU) labeling was used to determine if the 16-cell cysts in pingpong testes undergo DNA synthesis before arresting. BrdU-labeled 16-cell cysts were observed in both wild-type and pingpong testes. Thus, the block in meiosis occurs after DNA synthesis but before the first meiotic division (Wang, 2003).
The genetic analysis of spermatogenesis in Drosophila has identified several genes that are required for the normal progression through meiosis in males. Two of these genes, aly and cookie monster (comr) can be distinguished from the others because they are required for the expression of twine, which encodes a cdc25-like phosphatase, and boule, an ortholog of the human gene Deleted in Azoospermia. In contrast, the gene cannonball (can) is also required for meiosis in males, but is not required for the expression of twine or boule mRNAs. To gain additional insight into the pingpong mutant phenotype mutant testes were stained for several markers known to be expressed in testes, including twine and boule. Like aly and comr, but unlike can mutant testes, the pingpong mutant does not express twine or boule mRNAs. Also like aly and comr mutants, pingpong testes do not express mst87F, a gene that is required for spermatid differentiation. These results suggest that vis and achi function at a similar step as aly and comr (Wang, 2003).
To determine if vis and achi act upstream of or in parallel with aly and comr, pingpong mutant testes were stained with antibodies for the aly and comr gene products, which are also highly expressed in primary spermatocytes (Jiang, 2003; White-Cooper, 2000). Strikingly, both Aly and Comr are observed in the nuclei of most cells in pingpong mutant testes. Thus, vis and achi are not required for the expression or nuclear localization of these genes. Conversely, Vis/Achi proteins are still observed in aly mutant testes (as is Comr). Consistent with these results, Vis and Achi are also expressed and are nuclear in testes mutant for spermatocytes arrest (sa) and can, genes that appear to act downstream of aly and comr. Thus, neither sa, can, or aly are required for the expression or nuclear localization of Vis or Achi. Taken together, these data suggest that Vis and Achi function in spermatogenesis in parallel with Aly and Comr (Wang, 2003).
The lack of twine expression in pingpong mutant testes is consistent with it causing a meiotic arrest phenotype. In addition to twine, meiosis is also controlled by the availability of Cyclins A and B, which are required to activate Cyclin dependent kinase 1 (Cdk1). Both Cyclins A and B are normally expressed during male spermatogenesis, at high levels in the mitotically dividing cells at the apical tip of the testes, at lower levels during the primary spermatocyte stage, and at higher levels in 16-cell cysts just prior to the G2 to M transition. The cyclins are rapidly degraded at the end of metaphase. Cyclin levels were examined in pingpong mutant testes. Although there is some variation between individual testes, in general, intermediate levels of both Cyclin A and Cyclin B are observed throughout pingpong testes. Cyclin A and B levels persist, although at lower levels, up to the point when the cells appear to degenerate. In addition, although there is a transient nuclear localization of both Cyclins prior to cell division in the wild type, both Cyclins are observed predominantly in the cytoplasm in the pingpong mutant. Therefore, unlike in the wild type, Cyclin levels are not modulated in the pingpong mutant, consistent with a block prior to the G2/M transition. In addition, the levels of Polo, a protein kinase required for cytokinesis during meiosis, indicate that in pingpong testes the block occurs prior to the first meiotic division (Wang, 2003).
The characterization of vis and achi cDNAs suggests the existence of two isoforms that differ because of the presence or absence of an alternatively spliced exon. To analyze the protein products derived from vis and achi the anti-Achi antibody was used in immunoblot experiments. Two bands with apparent molecular masses of ~60 kDa and ~80 kDa are detected in whole lysates of wild-type testes. In contrast, only a ~60 kDa band is detected in wild-type ovaries. Neither band is detected in the testes or ovaries from the pingpong mutant, suggesting that both are derived from vis and achi. Furthermore, based on the sizes of proteins observed in pingpong; P{vis} or pingpong; P{achi} flies, it is deduced that VisL and AchiL both migrate at ~80 kDa and both VisS and AchiS migrate at ~60 kDa. This assignment was confirmed by expressing the AchiL or AchiS cDNAs under the control of the alpha1-tubulin promoter in Df(2R)pingpong flies. Furthermore, in wild-type flies the slower migrating forms (AchiL and VisL) are testes specific because only the ~60 kDa species is observed in male or female somatic tissues (Wang, 2003).
Tests were performed to see if either AchiL or AchiS2 is sufficient to rescue the pingpong male infertility phenotype by ubiquitously expressing these isoforms under the control of the tub promoter. Flies containing either tub-AchiL or tub-AchiS in an otherwise wild-type background appear normal. In addition, although the long isoforms are testes specific, both Df(2R)pingpong; tub-AchiL and Df(2R)pingpong; tub-AchiS2 males have normal appearing testes, as demonstrated by both phase contrast microscopy and immunostaining with several markers. However, in both cases these males are only weakly fertile Df(2R)pingpong; tub-AchiL males give rise to an average of 10 progeny/male. One explanation for this result is that the tub promoter fails to provide accurate levels or timing of Achi expression to fully rescue pingpong sterility. However, males in which both long and short forms of Achi are co-expressed exhibit better fertility, suggesting that expression of both isoforms is required for complete rescue (Wang, 2003).
Aoyagi, N. and Wassarman, D. A. (2000). Genes encoding Drosophila melanogaster RNA polymerase II general transcription factors: diversity in TFIIA and TFIID components contributes to gene-specific transcriptional regulation. J. Cell Biol. 150: F45-F49. 10908585
Aoyagi, N. and Wassarman, D. A. (2001). Developmental and transcriptional consequences of mutations in Drosophila TAF(II)60. Mol. Cell Biol. 21: 6808-6819. 11564865
Ayyar, S. and White, R. A. H. (2001). achintya and vismay: two novel, tandem-repeated TALE-class homeobox genes. A. Dros. Res. Conf. 42: 504C
Ayyar, S., Jiang, J., Collu, A., White-Cooper, H. and White, R. A. H. (2003). Drosophila TGIF is essential for developmentally regulated transcription in spermatogenesis. Development 130: 2841-2852. 12756169
Beitel, G. J., Lambie, E. J. and Horvitz, H. R. (2000). The C. elegans gene lin-9, which acts in an Rb-related pathway, is required for gonadal sheath cell development and encodes a novel protein. Gene 254: 253-263. 10974557
Bertolino, E., Reimund, B., Wildt-Perinic, D. and Clerc, R. G. (1995). A novel homeobox protein which recognizes a TGT core and functionally interferes with a retinoid-responsive motif. J. Biol. Chem. 270: 31178-31188. 8537382
Bertolino, E., Wildt, S., Richards, G. and Clerc, R. G. (1996). Expression of a novel murine homeobox gene in the developing cerebellar external granular layer during its proliferation. Dev. Dyn. 205: 410-420. 8901052
Blanco-Arias, P., Sargent, C. A. and Affara, N. A. (2002). The human-specific Yp11.2/Xq21.3 homology block encodes a potentially functional testis-specific TGIF-like retroposon. Mamm. Genome 13: 463-468. 12226713
Burglin, T. R. (1997). Analysis of TALE superclass homeobox genes (MEIS, PBC, KNOX, Iroquois, TGIF) reveals a novel domain conserved between plants and animals. Nucleic Acids Res. 25: 4173-4180. 9336443
Chen, F., Ogawa, K., Nagarajan, R. P., Zhang, M., Kuang, C. and Chen, Y. (2003). Regulation of TG-interacting factor by transforming growth factor-beta. Biochem J. 371(Pt 2): 257-63. 12593671
Gripp, K. W., Wotton, D., Edwards, M. C., Roessler, E., Ades, L., Meinecke, P., Richieri-Costa, A., Zackai, E. H., Massague, J. and Muenke, M. (2000). Mutations in TGIF cause holoprosencephaly and link NODAL signalling to human neural axis determination. Nat. Genet. 25: 205-208. 10835638
Hiller, M. A., Lin, T. Y., Wood, C. and Fuller, M. T. (2001). Developmental regulation of transcription by a tissue-specific TAF homolog. Genes Dev. 15: 1021-1030. 11316795
Imoto, I., Pimkhaokham, A., Watanabe, T., Saito-Ohara, F., Soeda, E. and Inazawa, J. (2000). Amplification and overexpression of TGIF2, a novel homeobox gene of the TALE superclass, in ovarian cancer cell lines. Biochem. Biophys. Res. Commun. 276: 264-270. 11006116
Jiang, J. and White-Cooper, H. (2003). Transcriptional activation in Drosophila spermatogenesis involves the mutually dependent function of aly and a novel meiotic arrest gene cookie monster. Development 130: 563-573. 12490562
Lai, Y. L., Li, H., Chiang, H. S. and Hsieh-Li, H. M. (2002). Expression of a novel TGIF subclass homeobox gene, Tex1, in the spermatids of mouse testis during spermatogenesis. Mech. Dev. 113: 185-187. 11960710
Lin, T. Y., Viswanathan, S., Wood, C., Wilson, P. G., Wolf, N. and Fuller, M. T. (1996). Coordinate developmental control of the meiotic cell cycle and spermatid differentiation in Drosophila males. Development 122: 1331-1341. 8620860
Lo, R. S., Wotton, D. and Massague, J. (2001). Epidermal growth factor signaling via Ras controls the Smad transcriptional co-repressor TGIF. EMBO J. 20: 128-136. 11226163
Melhuish, T. A. and Wotton, D. (2000). The interaction of C-terminal binding protein with the Smad corepressor TG-interacting factor is disrupted by a holoprosencephaly mutation in TGIF. J. Biol. Chem. 275: 39762-39766. 10995736
Melhuish, T. A., Gallo, C. M. and Wotton, D. (2001). TGIF2 interacts with histone deacetylase 1 and represses transcription. J. Biol. Chem. 276: 32109-32114. 11427533
Phillips, C. L., Stark, M. R., Johnson, A. D. and Dahlquist, F. W. (1994). Heterodimerization of the yeast homeodomain transcriptional regulators alpha 2 and a1 induces an interfacial helix in alpha 2. Biochemistry 33: 9294-9302. 8049230
Sharma, M. and Sun, Z. (2001). 5'TG3' interacting factor interacts with Sin3A and represses AR-mediated transcription. Mol. Endocrinology 15 (11): 1918-1928. 11682623
Wang, Z. and Mann, R. S. (2003). Requirement for two nearly identical TGIF-related homeobox genes in Drosophila spermatogenesis. Development 130: 2853-2865. 12756170
White-Cooper, H., Schafer, M. A., Alphey, L. S. and Fuller, M. T. (1998). Transcriptional and post-transcriptional control mechanisms coordinate the onset of spermatid differentiation with meiosis I in Drosophila. Development 125: 125-134. 9389670
Wotton, D., Lo, R. S., Lee, S. and Massague, J. (1999a). A Smad transcriptional corepressor. Cell 97: 29-39.
Wotton, D., Lo, R. S., Swaby, L. A. and Massague, J. (1999b). Multiple modes of repression by the Smad transcriptional corepressor TGIF. J. Biol. Chem. 274: 37105-37110. 10601270
Wotton, D., Knoepfler, P. S., Laherty, C. D., Eisenman, R. N. and Massague, J. (2001a). The Smad transcriptional corepressor TGIF recruits mSin3. Cell Growth Differ. 12: 4574-4563. 11571228
Wotton, D. and Massague, J. (2001b). Smad transcriptional corepressors in TGF beta family signaling. Curr. Top. Microbiol. Immunol. 254: 145-164. 11190572
Yang, Y., Hwang, C. K., D'Souza, U. M., Lee, S. H., Junn, E. and Mouradian, M. M. (2000). Three-amino acid extension loop homeodomain proteins Meis2 and TGIF differentially regulate transcription. J. Biol. Chem. 275: 20734-20741. 10764806
date revised: 12 June 2003
Home page: The Interactive Fly © 2003 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.