On the Genomic Location of the exuperantia1 Gene in Drosophila miranda: The Limits of in Situ Hybridization Experiments

Corresponding author: Doris Bachtrog, Animal and Population Biology, University of Edinburgh, King's Bldgs., West Mains Rd., Edinburgh EH9 3JT, United Kingdom., doris.bachtrog{at}ed.ac.uk (E-mail)

Communicating editor: M. A. F. NOOR

	ABSTRACT

TOP ABSTRACT LITERATURE CITED

In situ hybridization to Drosophila polytene chromosomes is a powerful tool for determining the chromosomal location of genes. Using in situ hybridization experiments, Yi and Charlesworth recently reported the transposition of the exuperantia1 gene (exu1) from a neo-sex chromosome to the ancestral X chromosome of Drosophila miranda, close to exuperantia2 (exu2). By characterizing sequences flanking exu1, however, we found the position of exu1 to be conserved on the neo-sex chromosome. Further, the exu2 gene was found to be tandemly duplicated on the X chromosome of D. miranda. The misleading hybridization signal of exu1 may be caused by multiple copies of exu2, which interfere with the hybridization of the exu1 probe to its genomic position on the neo-X chromosome. This suggests that flanking DNA should be used to confirm the positions of members of gene families.

CHROMOSOMAL arm homology is well established in Drosophila, with the position of genes being well conserved on chromosomes (so-called Muller's elements A–F; MULLER 1940 ). As an illustration, >160 independent DNA markers located on chromosome 3R of Drosophila melanogaster (element E) mapped, without exception, to the homologous salivary gland chromosome 2 of D. repleta (RANZ et al. 2001 ). D. repleta belongs to the Drosophila subgenus, whereas D. melanogaster is a member of the Sophophora subgenus (POWELL 1997 ); thus, chromosomal gene content has been largely preserved during a total time span of 80–120 MY of independent evolution.

Recently, an exception to this rule was reported, namely the transposition of the exuperantia1 gene (exu1) from the neo-sex chromosomes (element C) in D. miranda to the ancestral X chromosome (element A; YI and CHARLESWORTH 2000 ). In situ hybridization using the exu1 gene as a probe detected only a signal close to the exu2 gene on the X chromosome (YI and CHARLESWORTH 2000 ). exu2 is an ancient duplication of the exuperantia gene in the D. pseudoobscura lineage (LUK et al. 1994 ), which shows 85% identity to the exu1 locus at the nucleotide level. No sign of hybridization of an exu1 probe was detected on the neo-X (YI and CHARLESWORTH 2000 ). In the close relatives of D. miranda, however, the position of exu1 is conserved on Muller's element C (YI and CHARLESWORTH 2000 ). This observation was of particular interest, since it not only violated the rule of gene content conservation in the genus Drosophila, but also suggested a novel means of dosage compensation, as the locus involved was located on the newly evolving neo-sex chromosomes of D. miranda. The transposition of exu1 was interpreted as an advantageous event that conferred dosage compensation in response to the degeneration of the copy of exu1 on the neo-Y chromosome (YI and CHARLESWORTH 2000 ).

To further characterize this translocation, a genomic library of D. miranda was screened using the exu1 gene as a probe (BACHTROG 2003A ). Of 12 positive clones that were isolated, 2 corresponded to the neo-Y linked copy of exu1 (BACHTROG 2003A ), 1 corresponded to the exu2 gene, and the remaining 9 copies contained an exu1 sequence that was identical to the exu1 gene described in YI and CHARLESWORTH 2000 . Sequence analysis revealed that a clone containing exu1 also contained four genes that in the D. melanogaster genome surround the exuperantia gene (Fig 1). Unexpectedly, in situ hybridization using the entire -clone containing exu1 as a probe clearly demonstrated that the sequence is derived from the neo-X chromosome, and only weak staining is observed on the X (Fig 2A). Several experiments were conducted to further investigate the discrepancy between these results and those expected from the results of YI and CHARLESWORTH 2000 . First, we repeated the in situ hybridization described in Yi and Charlesworth, using only the exu1 gene as a probe. When exu1 amplified from D. pseudoobscura was used as a probe (as done by YI and CHARLESWORTH 2000 ), a strong band on the X chromosome (element A) was detected, but some very weak staining on the neo-X could be observed in some nuclei (Fig 2B). The same pattern was found if exu1 amplified from D. miranda was used as a probe (Fig 2C). Probes immediately flanking exu1 on both sides (see Fig 1), however, showed clear and exclusive staining to the neo-X chromosome (Fig 2D and Fig E). Thus, while exu1 clearly hybridizes most strongly to the X chromosome, adjacent genes from the same library clone hybridize only to the neo-X chromosome.

View larger version (5K): In this window In a new window Download PPT slide   Figure 1. Genomic organization of the -clone containing exu1 that was analyzed. A 16.5-kb fragment, which contains the exu1 gene as well as four adjacent genes, was isolated and sequenced. The solid bars indicate probes used for the in situ hybridization experiments (probe 1 and probe 3 correspond to PCR amplification products, and probe 2 is a SacI restriction fragment; see Fig 2).

View larger version (208K): In this window In a new window Download PPT slide   Figure 2. In situ hybridization of probes to the polytene chromosomes of D. miranda. Arrows mark hybridization signals. (A) Hybridization to the neo-X chromosome using the -clone containing exu1 as a probe; weak staining was also detected on the X chromosome. (B) Hybridization of exu1 amplified from D. pseudoobscura to the X chromosome of D. miranda (probe 1). (C) Hybridization of exu1 amplified from D. miranda to the X chromosome (probe 1); weak staining to the neo-X chromosome was detected in some nuclei. (D) Hybridization of flanking probes immediately upstream of exu1 to the neo-X chromosome (probe 2). (E) Hybridization of flanking probes immediately downstream of exu1 to the neo-X chromosome (probe 3). (F) Hybridization of the -clone containing exu2 to the X chromosome in D. miranda.

A possible explanation for this puzzle is provided by data on exu2. exu2 is a duplication of the exuperantia gene onto the ancestral X chromosome in the D. pseudoobscura group (LUK et al. 1994 ). Sequence analysis of a -clone containing exu2 revealed that exu2 is tandemly duplicated on the X chromosome; in the clone analyzed, three copies of exu2 were found (Fig 3). The three copies exhibit a very high level of sequence similarity; two copies can be translated into a functional protein, whereas the last copy contains a premature stop codon and is truncated at the 3' end (Fig 3). Further, the copy of exu2 sequenced by YI and CHARLESWORTH 2000 differs slightly from those in the clone recovered, implying that there are at least four copies of exu2 in the genome of D. miranda. The presence of multiple exu2 copies was also confirmed by long PCR and Southern blotting experiments (results not shown). Thus, despite exu1 and exu2 being 15% divergent at the nucleotide level (LUK et al. 1994 ), the occurrence of multiple copies of exu2 on the X chromosome of D. miranda might out-compete hybridization with the exu1 copy on the neo-X. In situ hybridization experiments using exu2 as a probe showed that exu1 and exu2 indeed hybridize to the same pair of chromosomal bands in D. miranda (Fig 2F), as reported by YI and CHARLESWORTH 2000 . Together, these data suggest that exu1 is still located on the neo-X chromosome and that the hybridization signal on the X chromosome (as detected by Yi and Charlesworth and reconfirmed here) was caused by the presence of multiple copies of exu2 on the X. We also performed double hybridization experiments with a labeled exu1 probe and unlabeled exu2 PCR product to investigate whether we could block the hybridization signal of exu1 to the X chromosome. Hybridization to both locations (i.e., strong staining with the X and weak staining with the neo-X, as with the exu1 probe) or no staining at all (i.e., the unlabeled exu2 supposedly dilutes the exu1 probe too much) was observed. This, however, is not surprising, since exu1 and exu2 also cross-hybridize in Southern blotting experiments, even at very high stringency (results not shown). However, since our library screen data suggest that there is only one copy of exu1 in the genome of D. miranda (see below), it seems very unlikely that we would observe the hybridization signal of exu1 at the same location as exu2 purely by coincidence. In D. pseudoobscura, exu1 does hybridize to element C (YI and CHARLESWORTH 2000 ); thus, we expect exu2 to be a single-copy gene in this species. Indeed, a 12-kb clone analyzed (from the D. pseudoobscura Genome Project; http://hgsc.bcm.tmc.edu/blast/?organism=Dpseudoobscura) contains only a single copy of the exu2 gene and is flanked by X-derived sequence in D. melanogaster (Fig 3). Southern blotting experiments using a conserved region of the exu gene as a probe always detected only two hybridization signals in D. pseudoobscura (i.e., one corresponding to exu1 and the other to exu2; LUK et al. 1994 ; YI and CHARLESWORTH 2000 ; results not shown), strongly suggesting that this is the only copy of exu2 in the D. pseudoobscura genome. The observation of multiple copies of exu2 in D. miranda and only a single copy of exu2 in D. pseudoobscura is consistent with our hypothesis that multiple copies of exu2 interfere with the in situ localization of the exu1 gene specifically in D. miranda.

View larger version (12K): In this window In a new window Download PPT slide   Figure 3. Genomic organization of the -clone containing exu2 in D. miranda. A 12.2-kb clone was analyzed, which was found to contain three tandem copies of exu2. The D. pseudoobscura genomic region (http://hgsc.bcm.tmc.edu/blast/?organism=Dpseudoobscura) contains only one copy of exu2; the shaded area indicates DNA sequence homologous to the X chromosome of D. melanogaster and the solid area indicates no homology to the D. melanogaster X chromosome.

We have demonstrated that there is a copy of exu1 on the neo-X and that this copy is not easily detected by in situ hybridization. However, there is still the possibility that there was a duplication event of exu1 close to exu2, which was so recent that there has not been sufficient time for new mutations to occur. Several lines of evidence argue against this possibility. First, all nine exu1 copies isolated from the -library screen were found to contain flanking sequence that does not hybridize to the X chromosome (i.e., the 5' flanking region of the CG9025 gene was sequenced in each clone; see Fig 1). This means that we picked up the neo-X linked copy nine times but never picked up the putatively transposed one. This is highly improbable (P < 0.01), assuming that we are equally likely to isolate the two copies with the library screen (i.e., their sequence is highly conserved). Second, polymorphism at exu1 was found to be reduced (YI and CHARLESWORTH 2000 ), which was interpreted as the signature of a selective sweep associated with the translocation of the exu1 gene. In 12 D. miranda lines studied, only two segregating variants in the exu1 gene close to the 5' end of the locus were observed in the D. miranda MA32 line (YI and CHARLESWORTH 2000 ). We performed a PCR experiment using one primer located within the exu1 gene and one primer located in flanking DNA that hybridizes only to the neo-X (i.e., in the 5' region of the CG30152 gene; see Fig 1). Sequence analysis of the same 12 D. miranda lines confirms that this amplification product contains the same two polymorphic sites observed by YI and CHARLESWORTH 2000 in the MA32 line. This demonstrates that the copy of exu1 investigated by YI and CHARLESWORTH 2000 is physically linked to the neo-X chromosome. Finally, polymorphism data of sequences adjacent to exu1, which hybridize only to the neo-X, also show a reduction in variability, indicating that the postulated selective sweep actually happened on the neo-X chromosome (BACHTROG 2003B ).

Together, these data suggest that exu1 did not translocate in D. miranda. The misleading hybridization result of exu1 may be caused by the existence of multiple copies of exu2 on the X chromosome of D. miranda. Only localization of flanking sequences allowed the determination of the correct position of exu1 on the neo-X of D. miranda. Our results imply that care must be taken in interpreting in situ hybridization data in the presence of gene families; specifically, they suggest that flanking DNA should be used to confirm the positions of members of gene families.

In addition, our observation that the exu1 gene has not moved to a different chromosome supports the traditional view of the conservation of gene content in the Drosophila genus. It also implies that there is no evidence for an alternative dosage compensation mechanism by the means of gene transpositions in the genus Drosophila. This is supported by in situ hybridization data on D. pseudoobscura, a closely related species that also has a neo-X chromosome (formed by the ancient fusion of Muller's element D with the X). In situ hybridization data of 18 element D loci in D. pseudoobscura did not detect any transposition events—only hybridization to the putative homologous chromosomes—apart from a movement of genes from A to D, probably due to a pericentric inversion (SEGARRA et al. 1996 ; B. CHARLESWORTH, unpublished results). Thus, while chromosomal rearrangements like inversions or chromosomal fusions are common in the genus Drosophila, the position of genes is well conserved among chromosomal arms.

	ACKNOWLEDGMENTS

We are very grateful to P. Andolfatto and S. Yi for helpful discussions and comments on the manuscript.

Manuscript received February 7, 2003; Accepted for publication March 7, 2003.

	LITERATURE CITED

TOP ABSTRACT LITERATURE CITED

BACHTROG, D., 2003a  Accumulation of spock and worf, two novel retrotransposons, on the neo-Y chromosome of Drosophila miranda.. Mol. Biol. Evol. 20:173-181.[Abstract/Free Full Text]

BACHTROG, D., 2003b  Adaptation shapes patterns of genome evolution on sexual and asexual chromosomes in Drosophila. Nat. Genet. 34:215-219.[Medline]

LUK, S. K., M. KILPATRICK, K. KERR, and P. M. MACDONALD, 1994  Components acting in localization of bicoid mRNA are conserved among Drosophila species. Genetics 137:521-530.[Abstract]

MULLER, H. J., 1940 Bearings of the Drosophila work on systematics, pp. 185–268 in The New Systematics, edited by J. S. HUXLEY. Oxford University Press, London.

POWELL, J. R., 1997 Progress and Prospects in Evolutionary Biology: The Drosophila Model. Oxford University Press, New York.

RANZ, J. M., F. CASALS, and A. RUIZ, 2001  How malleable is the eukaryotic genome? Extreme rate of chromosomal rearrangement in the genus Drosophila. Genome Res. 11:230-239.[Abstract/Free Full Text]

SEGARRA, C., G. RIBO, and M. AGUADE, 1996  Differentiation of Muller's chromosomal elements D and E in the obscura group of Drosophila. Genetics 144:139-146.[Abstract]

YI, S. and B. CHARLESWORTH, 2000  A selective sweep associated with a recent gene transposition in Drosophila miranda.. Genetics 156:1753-1763.[Abstract/Free Full Text]

This article has been cited by other articles:

				C. Bartolome and B. Charlesworth Rates and Patterns of Chromosomal Evolution in Drosophila pseudoobscura and D. miranda Genetics, June 1, 2006; 173(2): 779 - 791. [Abstract] [Full Text] [PDF]

				J. Gonzalez, F. Casals, and A. Ruiz Duplicative and Conservative Transpositions of Larval serum protein 1 Genes in the Genus Drosophila Genetics, September 1, 2004; 168(1): 253 - 264. [Abstract] [Full Text] [PDF]