Publications

An important goal in molecular genetics has been to identify a transposable element that might serve as an efficient transformation vector in diverse species of insects. The transposable element mariner occurs naturally in a wide variety of insects. Although virtually all mariner elements are nonfunctional, the Mos1 element isolated from Drosophila mauritiana is functional. Mos1 was injected into the pole-cell region of embryos of D. virilis, which last shared a common ancestor with D. mauritiana 40 million years ago. Mos1 PCR fragments were detected in several pools of DNA from progeny of injected animals, and backcross lines were established. Because G0 lines were pooled, possibly only one transformation event was actually obtained, yielding a minimum frequency of 4%. Mos1 segregated in a Mendelian fashion, demonstrating chromosomal integration. The copy number increased by spontaneous mobilization. In situ hybridization confirmed multiple polymorphic locations of Mos1. Integration results in a characteristic 2-bp TA duplication. One Mos1 element integrated into a tandem array of 370-bp repeats. Some copies may have integrated into heterochromatin, as evidenced by their ability to support PCR amplification despite absence of a signal in Southern and in situ hybridization.

Pseudogenes are common in mammals but virtually absent in Drosophila. All putative Drosophila pseudogenes show patterns of molecular evolution that are inconsistent with the lack of functional constraints. The absence of bona fide pseudogenes is not only puzzling, it also hampers attempts to estimate rates and patterns of neutral DNA change. The estimation problem is especially acute in the case of deletions and insertions, which are likely to have large effects when they occur in functional genes and are therefore subject to strong purifying selection. We propose a solution to this problem by taking advantage of the propensity of retrotransposable elements without long terminal repeats (non-LTR) to create non-functional, 'dead-on-arrival' copies of themselves as a common by-product of their transpositional cycle. Phylogenetic analysis of a non-LTR element, Helena, demonstrates that copies lose DNA at an unusually high rate, suggesting that lack of pseudogenes in Drosophila is the product of rampant deletion of DNA in unconstrained regions. This finding has important implications for the study of genome evolution in general and the 'C-value paradox' in particular.

Drosophila virilis is a prominent reference species for comparison with Drosophila melanogaster in regard to patterns and mechanisms of molecular and genomic evolution. Sequences were determined for 11 Adh genes from 8 species of the D. virilis species group, including species from both the virilis phylad and the montana subphylad. The genome of D. virilis contains a 6-kb duplication that includes the entire Adh coding region. The pattern of sequence identity within the duplication strongly suggests a recent gene-conversion event bordered by 36-bp indels. As in other Drosophila, the amino-acid coding region of Adh is encoded by three exons interrupted by two short introns. The promoter region includes 16 blocks of sequence that are well conserved in D. virilis, Drosophila hydei, and D. melanogaster. The developmental profile of Adh transcription suggests a distal/proximal promoter switch analogous to that in D. melanogaster. Duplicate Adh genes were also found in Drosophila montana and Drosophila lacicola, which apparently originated independently of that in D. virilis. The Adh genes in all species of the D. virilis group have among the lowest codon bias of any Adh genes so far reported in the genus Drosophila. Taking the low codon bias into account, we estimate the time of divergence between the virilis and montana clades as 9.0 +/- 0.7 Mya and the approximate time of divergence of D. virilis from other members of the virilis phylad as 2.6 +/- 0.4 Mya. The region of the D. virilis genome containing Adh, as well as the chromosome as a whole, gives evidence of extensive rearrangements relative to the genome of D. melanogaster.

Despite the prominence of the Azteca-Cecropia interaction as the focus of extensive ecological investigation, a reliable phylogeny of the Azteca ants has been lacking, primarily because many of the morphological and behavioral characters are phylogenetically uninformative or conflicting. A phylogenetic analysis of a select set of Azteca ants, including six Cecropia inhabitants and two non-Cecropia inhabitants, plus an outgroup taxon, is presented on the basis of mitochondrial DNA sequences. The evolutionary relationships deduced from the molecular data are analyzed with reference to ecological and morphological studies, specifically addressing the phylogenetic relationship of structurally an behaviorally ambiguous taxa, species complex groupings, and the colonization of Cecropia trees. According to the molecular phylogeny, the Cecropia-inhabiting Azteca do not form a monophyletic clade, indicating multiple independent colonization or abandonment of Cecropia trees by the Azteca.

The distribution of plasmids related to the fertility factor F was examined in the ECOR reference collection of Escherichia coli. Probes specific for four F-related genes were isolated and used to survey the collection by DNA hybridization. To estimate the genetic diversity of genes in F-like plasmids, DNA sequences were obtained for four plasmid genes. The phylogenetic relationships among the plasmids in the ECOR strains is very different from that of the strains themselves. This finding supports the view that plasmid transfer has been frequent within and between the major groups of ECOR. Furthermore, the sequences indicate that recombination between genes in plasmids takes place at a considerably higher frequency than that observed for chromosomal genes. The plasmid genes, and by inference the plasmids themselves, are mosaic in structure with different regions acquired from different sources. Comparison of gene sequences from a variety of naturally occurring plasmids suggested a plausible donor of some of the recombinant regions as well as implicating a chi site in the mechanism of genetic exchange. The relatively high rate of recombination in F-plasmid genes suggests that conjugational gene transfer may play a greater role in bacterial population structure than previously appreciated.

Germline mobilization of the transposable element mariner is severely inhibited by the insertion of a 4.5- to 11.9-kb fragment of exogenous DNA into a unique SacI site approximately in the middle of the 1286-bp element. In the presence of transposase driven by the germline-specific hsp26-sgs3 promoter, mobilization of the MlwB construct (containing a 11.9-kb insertion) is detected at low frequency. Analysis of a mobilized MlwB element indicated that mobilization is mediated by the mariner transposase. However, transposed MlwB elements are also defective in germline mobilization. Rare, transposase-induced germline excision events were also recovered for such vectors. The estimated rate of excision is < 0.1% per chromosome per generation. Excision appears to be accompanied by gap repair if a suitable template is available. The data imply that the reduced mobility of mariner vectors with exogenous DNA in the SacI site results from disruption of sequences necessary for efficient mobilization. The relative stability may be a valuable property in the uses of mariner-like elements in genetic engineering of insects of economic importance.

A strategy of "sequence scanning" is proposed for rapid acquisition of sequence from clones such as bacteriophage P1 clones, cosmids, or yeast artificial chromosomes. The approach makes use of a special vector, called LambdaScan, that reliably yields subclones with inserts in the size range 8-12 kb. A number of subclones, typically 96 or 192, are chosen at random, and the ends of the inserts are sequenced using vector-specific primers. Then long-range spectrum PCR is used to order and orient the clones. This combination of shotgun and directed sequencing results in a high-resolution physical map suitable for the identification of coding regions or for comparison of sequence organization among genomes. Computer simulations indicate that, for a target clone of 100 kb, the scanning of 192 subclones with sequencing reads as short as 350 bp results in an approximate ratio of 1:2:1 of regions of double-stranded sequence, single-stranded sequence, and gaps. Longer sequencing reads tip the ratio strongly toward increased double-stranded sequence.

We have studied the Mos1 transposase encoded by the transposable element mariner. This-transposase is a member of the "D,D(35)E" superfamily of proteins exhibiting the motif D,D(34)D. It is not known whether this transposase, or other eukaryote transposases manifesting the D,D(35)E domain, functions in a multimeric form. Evidence for oligomerization was found in the negative complementation of Mos1 by an EMS-induced transposase mutation in the catalytic domain. The transposase produced by this mutation has a glycine-to-arginine replacement at position 292. The G292R mutation strongly interferes with the ability of wild-type transposase to catalyze excision of a target element. Negative complementation was also observed for two other EMS mutations, although the effect was weaker than observed with G292R. Results from the yeast two-hybrid system also imply that Mos1 subunits interact, suggesting the possibility of subunit oligomerization in the transposition reaction. Overproduction of Mos1 subunits through an hsp70 promoter also inhibits excision of the target element, possibly through autoregulatory feedback on transcription or through formation of inactive or less active oligomers. The effects of both negative complementation and overproduction may contribute to the regulation of mariner transposition.

Studies of gene function and regulation in transgenic Drosophila are often compromised by the possibility of genomic position effects on gene expression. We have developed a method called transgene coplacement, in which any two sequences can be positioned at exactly the same site and orientation in the genome. Transgene coplacement makes use of the bacteriophage P1 system of Cre/loxP site-specific recombination, which we have introduced into Drosophila. In the presence of a cre transgene driven by a dual hsp70-Mos1 promoter, a white reporter gene flanked by loxP sites is excised with virtually 100% efficiency both in somatic cells and in germ cells. A strong maternal effect, resulting from Cre recombinase present in the oocyte, is observed as white or mosaic eye color in F1 progeny. Excision in germ cells of the F1 yields a strong grand-maternal effect, observed as a highly skewed ratio of eye-color phenotypes in the F2 generation. The excision reactions of Cre/loxP and the related FLP/FRT system are used to create Drosophila lines in which transgenes are at exactly allelic sites in homologous chromosomes.

We analyzed the geographic distribution of the Ixodes ricinus-like ticks in eastern North America by comparing the mitochondrial 16S rDNA sequences of specimens sampled directly from the field during the 1990s. Two distinct lineages are evident. The southern clade includes ticks from the southeastern and middle-eastern regions of the United States. The range of the northern clade, which appears to have been restricted to the northeastern region until the mid-1900s, now extends throughout the northeastern and middle-eastern regions. These phyletic units correspond to northern and southern taxa that have previously been assigned specific status as Ixodes dammini and Ixodes scapularis, respectively. The expanding range of I. dammini appears to drive the present outbreaks of zoonotic disease in eastern North America that include Lyme disease and human babesiosis.

We describe a system of hybrid dysgenesis in Drosophila virilis in which at least four unrelated transposable elements are all mobilized following a dysgenic cross. The data are largely consistent with the superposition of at least three different systems of hybrid dysgenesis, each repressing a different transposable element, which break down following the hybrid cross, possibly because they share a common pathway in the host. The data are also consistent with a mechanism in which mobilization of a single element triggers that of others, perhaps through chromosome breakage. The mobilization of multiple, unrelated elements in hybrid dysgenesis is reminiscent of McClintock's evidence [McClintock, B. (1955) Brookhaven Symp. Biol. 8, 58-74] for simultaneous mobilization of different transposable elements in maize.

Transposable elements are a major source of genetic change, including the creation of novel genes, the alteration of gene expression in development, and the genesis of major genomic rearrangements. They are ubiquitous among contemporary organisms and probably as old as life itself. The long coexistence of transposable elements in the genome would be expected to be accompanied by host-element coevolution. Indeed, the important role of host factors in the regulation of transposable elements has been illuminated by recent studies of several systems in Drosophila. These include host factors that regulate the P element, a host mutation that renders the genome permissive for gypsy mobilization and infection, and newly induced mutations that affect the expression of transposon insertion mutations. The finding of a type of hybrid dysgenesis in D. virilis, in which multiple unrelated transposable elements are mobilized simultaneously, may also be relevant to host-factor regulation of transposition.

The baseline rate of spontaneous integration of the autonomous mariner element Mos1 into the germline of Drosophila melanogaster is estimated as 16 +/- 5% (mean +/- SE) among fertile G0 flies. However, the transformation rate is reduced approximately 20-fold in Mos1 constructs with exogenous DNA in the size range 5-12 kb inserted into the SacI site. To provide alternative Mos1 helper plasmids for transformation experiments, two types of Mos1-promoter fusions were constructed: hsp-70:Mos1 and hsp26-Sgs3:Mos1. The former has the Mos1 coding region driven by the hsp70 heat-shock promoter; the latter has it driven by the basal Sgs3 promoter under the control of the hsp26 female-germline specific transcriptional regulator. When introduced into D. melanogaster by P-element-mediated germline transformation, these elements are unable to transpose or excise in the presence of autonomous Mos1-related elements (they are "marooned") because the 5' inverted repeat of Mos1 is missing. As expected, the hsp26-Sgs3:Mos1 fusions exhibit a significantly greater rate of germline excision of a target mariner element than do the hsp70:Mos1 fusions. Unexpectedly, the rate of excision of target mariner elements induced by hsp26-Sgs3:Mos1 is the same in the male germline as in the female germline. Both hsp:Mos1 fusions show strong germline expression and a maternal effect of the mariner transposase. A significant grand-maternal effect of the hsp:Mos1 fusions was also detected as a result of a maternal effect on the germline of the F1 progeny. Among flies carrying the promoter fusions inherited maternally, about three-quarters of the overall rate of germline excision derives from the direct genotypic effect and about one-quarter results from the grand-maternal effect. Despite the strong somatic expression of the hsp:Mos1 fusions, mariner transformants carrying a white+ reporter gene at the SacI site remained stable in the soma.

Horizontal transmission has been well documented as a major mechanism for the dissemination of mariner-like elements (MLEs) among species. Less well understood are mechanisms that limit vertical transmission of MLEs resulting in the "spotty" or discontinuous distribution observed in closely related species. In this article we present evidence that the genome of the common ancestor of the melanogaster species subgroup of Drosophila contained an MLE related to the mellifera (honey bee) subfamily. Horizontal transmission, approximately 3-10 MYA, is strongly suggested by the observation that the sequence of the MLE in Drosophila erecta is 97% identical in nucleotide sequence with that of an MLE in the cat flea, Ctenocephalides felis. The D. erecta MLE has a spotty distribution among species in the melanogaster subgroup. The element has a high copy number in D. erecta and D. orena, a moderate copy number in D. teissieri and D. yakuba, and was apparently lost ("stochastic loss") in the lineage leading to D. melanogaster, D. simulans, D. mauritiana, and D. sechellia. In D. erecta, most copies are concentrated in the heterochromatin. Two copies from D. erecta, denoted De12 and De19, were cloned and sequenced, and they appear to be nonfunctional ("vertical inactivation"). It therefore appears that the predominant mode of MLE evolution is vertical inactivation and stochastic loss balanced against occasional reinvasion of lineages by horizontal transmission.

The vast diversity in spectral sensitivities in the vision of many organisms is mediated mostly (although not entirely) through variation in the photosensitive visual pigments (opsins) of the eye. Specifically, shifts in absorption maxima of visual pigments are thought to be a result of interactions within the binding pocket of the opsin, between amino acid side chains and the retinal chromophore, However, it has proven difficult to identify specific amino acid residues important in determining wavelength absorption maxima, especially for some of the short wavelength (blue) opsins. In this paper, a comparative phylogenetic approach was applied to opsin protein sequence data to identify residues important in opsin wavelength regulation. In essence, this approach consisted of interpreting evolutionary history as a series of experiments in which natural selection has repeatedly favored amino acid replacements of certain residues to shift the opsin absorption spectra to either shorter or longer wavelengths. Opsin protein sequences were obtained from GenBank, aligned, and used to reconstruct a phylogenetic tree. Amino acid replacements were traced along the branches of this opsin tree, focusing only on residues likely to reside within the chromophore-binding pocket. A number of functionally convergent, nonconservative amino acid replacements in independently evolved opsins with similar shifts in spectral properties were identified. In short, reconstruction of the phylogeny of the opsin molecule allowed us to track amino acid substitutions in specific sites within the opsin and to target those particular substitutions that are repeatedly associated with marked changes in peak absorbance, shifting the spectral sensitivity of the opsin toward shorter or longer wavelengths. Based on these results, we propose a model for blue shifts of opsin absorption spectra. Amino acid replacements of four polar and charged residues near the protonated Schiff base (SBH+) end of the chromophore are proposed to result in blue shifts of the opsin absorption spectra. This model may explain some of the diversity of blue opsins apparent in both vertebrates and invertebrates.