Deletion of a Conserved Regulatory Element in the Drosophila Adh Gene Leads to Increased Alcohol Dehydrogenase Activity but Also Delays Development

Corresponding author: John Parsch, Department of Organismic and Evolutionary Biology, Harvard University Biological Laboratories, 16 Divinity Ave., Cambridge, MA 02138-2020., jparsch{at}oeb.harvard.edu (E-mail)

Communicating editor: S. YOKOYAMA

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In vivo levels of enzymatic activity may be increased through either structural or regulatory changes. Here we use Drosophila melanogaster alcohol dehydrogenase (ADH) in an experimental test for selective differences between these two mechanisms. The well-known ADH-Slow (S)/Fast (F) amino acid replacement leads to a twofold increase in activity by increasing the catalytic efficiency of the enzyme. Disruption of a highly conserved, negative regulatory element in the Adh 3' UTR also leads to a twofold increase in activity, although this is achieved by increasing in vivo Adh mRNA and protein concentrations. These two changes appear to be under different types of selection, with positive selection favoring the amino acid replacement and purifying selection maintaining the 3' UTR sequence. Using transgenic experiments we show that deletion of the conserved 3' UTR element increases adult and larval Adh expression in both the ADH-F and ADH-S genetic backgrounds. However, the 3' UTR deletion also leads to a significant increase in developmental time in both backgrounds. ADH allozyme type has no detectable effect on development. These results demonstrate a negative fitness effect associated with Adh overexpression. This provides a mechanism whereby natural selection can discriminate between alternative pathways of increasing enzymatic activity.

THE amount of activity of a given enzyme in an organism may be increased by either of two evolutionary mechanisms: a structural change that alters the catalytic efficiency of the enzyme or a regulatory change that alters the amount of enzyme produced. These two mechanisms are often considered equivalent in models of enzyme evolution because they result in the same phenotype when measured as in vivo enzymatic activity. The phenotypes induced by these alternative mechanisms may differ greatly, however, when measured as in vivo mRNA or protein concentration. Thus, it is possible that natural selection may be able to discriminate between these two routes of increase in enzymatic activity. The Drosophila alcohol dehydrogenase (ADH; EC 1.1.1.1) enzyme provides an excellent opportunity to investigate this possibility. ADH is one of the most abundant proteins in Drosophila melanogaster (BENYAJATI et al. 1980 ) and is encoded by a single gene, Adh. ADH functions primarily in the detoxification of ingested alcohols, and it is thought that the evolution of this enzyme, accompanied by adaptation to survival in alcohol-rich environments such as rotting fruits, has played an important role in the evolution of the family Drosophilidae, particularly the subfamily Drosophilinae (ASHBURNER 1998 ).

ADH protein and DNA sequence variation have been studied extensively in D. melanogaster. Electrophoretic studies of allozyme polymorphism have identified two forms of the ADH enzyme that are present at high frequency in populations worldwide (OAKESHOTT et al. 1982 ). The two forms, designated ADH-Slow (S) and ADH-Fast (F), differ by a single lysine threonine amino acid replacement (FLETCHER et al. 1978 ). The amino acid replacement leads to an approximately twofold increase in catalytic activity in the ADH-F form (CHOUDHARY and LAURIE 1991 ). The amino acid replacement, however, does not significantly affect the total amount of enzyme or Adh mRNA in the fly (CHOUDHARY and LAURIE 1991 ). Several lines of evidence suggest that natural selection favors increased ADH activity and/or the S F replacement in at least some environments: (1) ADH activity is positively correlated with ethanol tolerance in laboratory experiments, and is also positively correlated with ethanol concentration in natural environments (CHAKIR et al. 1993 ; MERCOT et al. 1994 ); (2) ADH-S and ADH-F alleles show similar latitudinal clinal distributions on different continents (OAKESHOTT et al. 1982 ; BERRY and KREITMAN 1993 ), suggesting a common selective gradient; (3) intra- and interspecific DNA sequence comparisons indicate that ADH-S is the ancestral form of the enzyme and that ADH-F has risen to high frequency worldwide in the recent past (KREITMAN 1983 ; AQUADRO et al. 1985 ; SULLIVAN et al. 1990 ); and (4) statistical tests based on levels of nucleotide polymorphism and divergence reject a pattern of neutral molecular evolution at the Adh locus in D. melanogaster (HUDSON et al. 1987 ; KREITMAN and HUDSON 1991 ; BEGUN et al. 1999 ).

A complex polymorphism, designated as 1, in the first (adult) Adh intron that is tightly linked to the S/F replacement site has also been shown to affect ADH activity levels (LAURIE and STAM 1994 ). Experimental replacement of the S-linked 1 sequence with that of the F-linked results in a 15–20% increase in ADH activity (LAURIE and STAM 1994 ). The 1 polymorphism, like the ADH allozymes, shows a geographical cline in frequency and may be a target of positive selection (BERRY and KREITMAN 1993 ). Interestingly, the increase in ADH activity associated with the F-linked 1 sequence is not accompanied by an increase in Adh mRNA concentration (LAURIE and STAM 1994 ).

We have identified previously a negative regulatory element in the Adh 3' untranslated region (UTR; PARSCH et al. 1999 ). Deletion of this highly conserved 8-bp sequence in the ADH-F background leads to a twofold increase in in vivo ADH activity. The relative change in activity is thus nearly identical to the one caused by the S F amino acid replacement. The 3' UTR deletion, however, does not affect the amino acid sequence of the ADH enzyme; activity is increased through twofold greater in vivo concentration of Adh mRNA and protein than in nondeletion flies (PARSCH et al. 1999 ). The conservation of this 3' UTR sequence implies strong purifying selection against mutations that disrupt the 8-bp motif. As discussed above, there is no evidence for purifying selection against ADH-F alleles. On the contrary, most evidence indicates positive selection on the amino acid replacement. Why, then, is there selection against disruption of the 3' UTR motif? We have hypothesized that since mutation of the 3' UTR sequence increases ADH activity by increasing in vivo Adh mRNA and protein concentrations, while the S F amino acid replacement increases activity solely by altering the catalytic efficiency of the enzyme, there may be deleterious effects associated with increased Adh mRNA and protein levels (PARSCH et al. 1999 ).

To test this hypothesis, we first demonstrate that deletion of the conserved 3' UTR motif results in twofold greater ADH activity in the ADH-S background. This is necessary because previous experiments used only the ADH-F background (PARSCH et al. 1999 ). We then compare the developmental times (from egg to adult) of transgenic flies carrying wild-type and deletion 3' UTR sequences in both the ADH-F and ADH-S backgrounds. We find that in both backgrounds, flies with the 3' UTR deletion show a significant increase in developmental time relative to wild type. Importantly, wild-type ADH-F flies and 3' UTR-deletion ADH-S flies, which are nearly identical in ADH activity, also show a significant difference in developmental time. Thus the effect cannot be explained by differences in ADH activity, but must be a result of the underlying difference in Adh mRNA and/or protein concentration. These results are consistent with our hypothesis and suggest that the different mechanisms of increasing ADH enzymatic activity are not selectively equivalent.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Site-directed mutagenesis and plasmid construction:
Wild-type and mutant ADH-F constructs were derived from an 8.6-kb SacI-ClaI fragment of the Adh Wa-f allele (KREITMAN 1983 ) and have been described previously (PARSCH et al. 1999 ). For the ADH-S constructs, the 8.6-kb SacI-ClaI fragment of theWa-s allele (KREITMAN 1983 ) was used. A 3.2-kb SalI-ClaI fragment containing the entire Adh mRNA encoding region was subcloned into a pUC18 plasmid and used as a template for site-directed mutagenesis following the QuikChange (Stratagene, La Jolla, CA) procedure. The mutagenesis primers were 5' GATATTCACGCAAGGCAATTTCGATGCACACTCACATTC 3' and its complement. These primers correspond to positions 1743–1790 in KREITMAN's (1983) consensus sequence, with bases 1762–1769 eliminated. Note, however, that the original consensus sequence contains a sequencing error (an extra T at position 1761; LAURIE et al. 1991 ) that has been corrected in our primer design. Thus, there is not perfect correspondence between our sequence and that of KREITMAN 1983 . For consistency, we use KREITMAN's (1983) coordinates to designate the highly conserved 8-bp motif as bases 1762–1769. The mutagenesis procedure resulted in the specific deletion of the conserved 8-bp 3' UTR motif, which was confirmed by direct DNA sequencing. Following mutagenesis, a BamHI-ClaI restriction fragment containing the mutant 3' UTR was used to replace the corresponding fragment in the original 8.6-kb SacI-ClaI clone. The entire BamHI-ClaI region was then sequenced to ensure that the 8-bp motif was deleted and that no other sequence changes had occurred. The constructs were designated as wt-S, 1762–1769-S, wt-F, and 1762–1769-F, where wt and 1762–1769 indicate the wild-type and mutant 3' UTR, and S and F indicate the Wa-s and Wa-f backgrounds (Fig 1).

View larger version (12K): In this window In a new window Download PPT slide   Figure 1. Schematic representation of the Adh mRNA encoding region. Exons are represented as boxes; introns as lines. The protein encoding regions are shown as solid boxes. The F/S amino acid replacement site is indicated, as well as the highly conserved 8-bp 3' UTR sequence that is deleted by the 1762–1769 mutation. Numbering is from KREITMAN 1983 .

Germline transformation:
Wild-type and mutant Adh fragments were inserted into the polycloning region of the YES transformation vector (PATTON et al. 1992 ). This vector contains the D. melanogaster yellow (y) gene as a phenotypic marker and also contains binding sites for the suppressor of Hairy-wing protein, which flank the target DNA and serve as an insulator of chromosomal position effects (PATTON et al. 1992 ). All constructs were introduced into the y w; Adh^fn6; 2–3, Sb/TM6 genetic background (ADH-null) through germline transformation (RUBIN and SPRADLING 1982 ; SPRADLING and RUBIN 1982 ). Some insertions on the X and third chromosomes were mobilized to new locations through genetic crosses using the 2–3 P element as a source of transposase (ROBERTSON et al. 1988 ; PARSCH et al. 1997 ). Following transformation or mobilization, all lines were crossed to a y w; Adh^fn6 stock to remove the source of transposase. To avoid possible effects of dosage compensation on Adh expression (LAURIE-AHLBERG and STAM 1987 ; PARSCH et al. 1997 ), only autosomal-insertion lines were used for further analysis. In addition, we limited our analysis to transformed lines containing only a single Adh insertion, as determined by Southern blotting (PARSCH et al. 1997 ).

Measurement of ADH activity:
For measurements of adult activity, we used a total of 32 transformed lines: 8 wt-S, 7 1762–1769-S, 10 wt-F, and 7 1762–1769-F. Transformed males were crossed to y w; Adh^fn6 females to produce offspring heterozygous for the Adh insertion. For each cross, five males and five females were placed in each of two vials and the pooled progeny were used for ADH assays. Total soluble protein was extracted from five 6- to 8-day-old heterozygous males and ADH activity was determined spectrophotometrically (MARONI 1978 ) using isopropanol as the substrate. The total protein concentration of each preparation was estimated using the method of LOWRY et al. 1951 . Units of ADH activity were measured as micromoles of NAD reduced per minute per milligram of total protein. Two separate ADH assays were performed on each protein preparation, and two separate protein preparations were made from each line. Thus, a total of four ADH assays were performed for each transformed line. Activity differences between the different transformant classes were tested by analysis of variance (ANOVA), using a model that accounts for intraclass position-effect variation (LAURIE-AHLBERG and STAM 1987 ). Statistical analyses were performed with the MacAnova software package (OEHLERT and BINGHAM 1998 ).

For measurements of larval activity, we used homozygous lines established from transformants showing adult ADH activities typical of their respective transformant classes. Larvae were collected at each instar stage from two independent homozygous lines within each transformant class. Soluble proteins were extracted from either 20 (first instar), 10 (second instar), or 5 (third instar) larvae and ADH activity was assayed as described above. ADH histochemical staining was performed on hand-dissected third instar larvae using standard techniques (ASHBURNER 1989 ).

Developmental time-course assays:
Three separate experiments (hereafter referred to as E1, E2, and E3) were performed to assess the rate of development of wild-type and 3' UTR-deletion transformants. Experiments were performed at two different locations and by three different investigators: E1 by J. A. Russell (in Rochester, NY); E2 and E3 by J. Parsch and I. Beerman, respectively (in Cambridge, MA). E1 compared wt-F and 1762–1769-F transformants, with all flies maintained at 22°. E2 and E3 compared wt-S, 1762–1769-S, wt-F, and 1762–1769-F transformants, with all flies maintained at 25°. Otherwise, the experimental protocol was identical for each of the three experiments except where noted. Flies were reared in 25 x 95-mm vials containing 7–8 ml of standard cornmeal-molasses medium. Males and virgin females were collected from homozygous lines and aged 2–5 days. Within each transformant class, a single male from one homozygous line was placed in a vial with a female from a different homozygous line and mating was allowed to occur over a 12-hr period. This mating scheme results in progeny carrying two nonallelic copies of the same Adh transgene and was employed to eliminate the possibility that homozygous deleterious mutations caused by the random insertion of our transposable element vectors into the genome might affect the rate of development. For each experiment, three to five homozygous lines from within each transformant class were used for developmental assays. In addition, three to six replicates of each cross (and of the reciprocal cross) were performed. After mating, the male was discarded and the female was placed in a fresh vial, which became the day 1 vial. Females were allowed to lay eggs in the vial for a period of 24 hr, and then were transferred to another fresh vial. This process was repeated for the following 4 days, when the female was discarded. Each female thus produced five vials of eggs that were allowed to develop in uncrowded conditions (<50 eggs per vial). Vials in which the female did not lay eggs were discarded and not considered in further analyses. In total, the number of crosses producing fertile eggs in each experiment ranged from 31 to 71. The number of pupae and the number of emerging adults in each vial were recorded every 24 hr. In E1 and E2, the number of eggs and the number of hatched larvae were recorded for a subset of the vials (the day 1 and day 5 vials) over the first 5 days of development.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Effect of the 3' UTR deletion in ADH-F/S backgrounds:
We have demonstrated previously that deletion of the conserved 8-bp 3' UTR motif leads to a significant (approximately twofold) increase in ADH activity in the ADH-F background (PARSCH et al. 1999 ). Although this region of the 3' UTR is identical in Adh-f and Adh-s alleles (KREITMAN 1983 ; LAURIE et al. 1991 ), it was not clear whether or not deletion of the 8-bp sequence would lead to a similar increase in activity in the ADH-S background. Flies homozygous for ADH-F typically have two to three times the in vivo ADH activity of those homozygous for ADH-S (LAURIE et al. 1991 ). A large portion of this difference can be attributed to the S/F amino acid replacement (CHOUDHARY and LAURIE 1991 ). Other nonreplacement differences (including 1) between Adh-f and Adh-s alleles, however, have also been shown to affect Adh expression (LAURIE and STAM 1994 ; STAM and LAURIE 1996 ). It is thus possible that interactions between the S/F site (or other linked nonreplacement sites) and the conserved 3' UTR sequence may affect the overall level of Adh expression. To test this, we specifically deleted the 8-bp 3' UTR motif in the Adh-s background and compared the ADH activities of wt-S and 1762–1769-S transformed lines (Fig 2). The activities were also compared to those of transformed lines containing the wt-F and 1762–1769-F Adh constructs (Fig 2; PARSCH et al. 1999 ). Our results indicate that the 8-bp 3' UTR deletion leads to a highly significant increase in ADH activity in both the ADH-S (P < 0.001) and ADH-F (P < 0.001) backgrounds. Note, however, that there is no difference in activity between 1762–1769-S and wt-F transformants (P = 0.414). This indicates that disruption of the conserved 3' UTR motif has the same effect on in vivo ADH activity as the S F amino acid replacement.

View larger version (14K): In this window In a new window Download PPT slide   Figure 2. Average adult ADH activity of the four different transformant types. Wild-type and 1762–1769 3' UTRs are represented as wt and , respectively. ADH allozyme type is indicated as either F or S. ADH activity is given in units of micromoles of NAD reduced per minute per milligram of total protein (multiplied by 100). Error bars represent ±1 SE.

ADH activity at different larval stages:
In wild-type flies, Adh gene expression is known to vary over the course of development, beginning with low levels at the first instar larval stage and increasing through the larval stages until the highest levels are reached in the adult (CORBIN and MANIATIS 1989 ). To determine if the 8-bp 3' UTR deletion had any effect on the developmental pattern of Adh expression in either the ADH-S or ADH-F background, we measured ADH enzymatic activity of wt-S, 1762–1769-S, wt-F, and 1762–1769-F flies at each larval instar stage (Fig 3). The pattern of ADH activity at each larval stage is very similar to that seen in adults: the 3' UTR deletion leads to a large increase in activity at each stage in both backgrounds. The activities of 1762–1769-S and wt-F are indistinguishable at each stage (Fig 3).

View larger version (26K): In this window In a new window Download PPT slide   Figure 3. Average ADH activity of wt-S, 1762–1769-S, wt-F, and 1762–1769-F transformants at different larval stages. 1762–1769 is indicated by . Units of ADH activity are the same as in Fig 2. Error bars represent the range of observed activities.

Tissue-specific expression patterns:
SIEGAL and HARTL 1999 observed tissue-specific silencing of some Adh transgenes due to generalized position effects. In these cases, Adh expression was abolished specifically in larval gastric ceca and the transformed lines showed ADH activity levels lower than average for their specific transformant class (SIEGAL and HARTL 1999 ). To test for such effects in our transformed lines, we performed histochemical staining of ADH activity in dissected larval tissues. We did not observe differences in the ADH staining pattern among any of our transformant lines, particularly in the gut or gastric ceca (Fig 4). These results indicate that the qualitative pattern of Adh expression is not altered by the 3' UTR deletion. Differences in the overall amount of ADH activity must, therefore, be due to quantitative differences in Adh expression in the tissues where it is normally expressed.

View larger version (73K): In this window In a new window Download PPT slide   Figure 4. Histochemical staining of ADH activity in the digestive system of third instar larvae. (A) Adhfn6 (ADH-null), (B) wild-type Adh (Wa-s allele), (C) wt-S transformant, (D) 1762–1769-S transformant, (E) wt-F transformant, (F) 1762–1769-F transformant. Staining in the gastric ceca is indicated by an arrow in B.

Developmental time-course results:
The results of the three developmental time-course experiments (E1, E2, and E3) are summarized in Table 1. Despite being performed at two different locations, by three different investigators, and in three different time blocks, the results are remarkably consistent. The most notable difference among experiments is that the developmental times are overall slower in E1 than in E2 or E3 (Table 1). This difference can be attributed to the experiments being performed at different temperatures (22° for E1, 25° for E2 and E3), as development is known to be slower at lower temperatures (ASHBURNER 1989 ). The effects of other environmental differences among the three experiments, however, appear to be small, and the relation of developmental times is in complete agreement across experiments. 1762–1769-S and 1762–1769-F flies show an increase in the time required to reach both the pupal and adult stages relative to their wild-type counterparts (Table 1). Nested ANOVA was performed separately for each experiment to test for significant effects of either allozyme (ADH-S/F) or UTR (wt/1762–1769) on the time required to reach the larval, pupal, or adult stage (Table 2). In all three experiments, we detect a highly significant effect of UTR on the time to reach both the pupal and adult stages (Table 2). There is, however, no significant effect of allozyme on developmental time at any stage (Table 2). Under the ANOVA model, UTR was nested within allozyme to test for significant effects of UTR in both the ADH-S and ADH-F backgrounds. A nonhierarchical model produced essentially the same results (not shown). Note that the difference in time to reach the adult stage between wild-type and 1762–1769 flies in each experiment is nearly identical to the difference in time to reach the pupal stage. This indicates that the slowdown in development in 1762–1769-F and 1762–1769-S flies can be completely explained by an increase in the time to reach the pupal stage. Fig 5 shows the cumulative fraction of flies reaching the pupal or adult stage on each day of the three experiments. Again, the results are consistent across experiments. Overall, flies with the wild-type 3' UTR reach the pupal and adult stages faster than their 1762–1769 counterparts in both the ADH-S and ADH-F backgrounds. The effect is strongest on the 3–4 days following the first appearance of either pupae or adults (Fig 5).

View larger version (63K): In this window In a new window Download PPT slide   Figure 5. Cumulative fraction of pupae and adults on each day of the developmental time-course experiments. (A) E1 pupae, (B) E1 adults, (C) E2 pupae, (D) E2 adults, (E) E3 pupae, (F) E3 adults.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A number of 3' UTR sequence motifs that affect gene expression are known, the majority of which act post-transcriptionally by altering mRNA stability, localization, or translation initiation (reviewed by CURTIS et al. 1995 ; ST. JOHNSTON 1995 ; ROSS 1996 ). In Drosophila, three 3' UTR-specific sequence elements involved in the post-transcriptional regulation of genes in the Bearded and Enhancer of split complexes have been described (LAI and POSAKONY 1997 ; LEVITEN et al. 1997 ; LAI et al. 1998 ). These sequences, the Bearded (Brd) box (AGCTTTA), GY box (GTCTTCC), and K box (TGTGAT), are highly conserved within paralogous gene families in D. melanogaster, and also in orthologous genes in D. virilis and D. hydei (LAI and POSAKONY 1997 ; LAI et al. 1998 ). Experimental deletion of these motifs leads to overexpression of their respective genes and results in defects in neural development (LAI and POSAKONY 1997 ; LAI et al. 1998 ). Similarly, the Drosophila Dlmo gene contains several Brd-like motifs in its 3' UTR (SHORESH et al. 1998 ). Beadex mutations, which disrupt the Dlmo 3' UTR, cause Dlmo overexpression and result in a characteristic scalloping of the wings (SHORESH et al. 1998 ). These examples demonstrate that there may be deleterious fitness effects associated with disruption of conserved 3' UTR motifs and the resulting gene overexpression.

Deletion of the Adh 3' UTR motif (AAGGCTGA) results in a twofold increase in Adh expression in both the ADH-S and ADH-F backgrounds (Fig 2 and Fig 3). This overexpression, however, does not result in any obvious phenotypic abnormalities, and flies overexpressing Adh appear normal with respect to fertility and viability. The strong conservation of the 8-bp 3' UTR motif, however, suggests that there are deleterious fitness effects associated with mutations in this region (PARSCH et al. 1999 ). This is somewhat unexpected given that there is apparently no deleterious effect of increased ADH activity caused by the S F amino acid replacement. On the contrary, most evidence indicates positive selection for increased ADH activity. Why, then, is a negative regulatory element so highly conserved in Adh? Our developmental time-course results provide an answer to this question. The deleterious effect is not due to increased ADH activity per se, but is instead due to the increased level of Adh expression caused by disruption of the 8-bp 3' UTR motif. In both the ADH-S and ADH-F backgrounds, 1762–1769 flies show a highly significant increase in developmental time relative to flies with the wild-type 3' UTR (Table 2). It is possible that the excess metabolic energy spent producing ADH is in part responsible for the developmental slowdown in 1762–1769 flies. Also, it is generally assumed that the presence of free ribosomes is the limiting factor in translation. Thus, the increased levels of Adh mRNA in the 1762–1769 mutants may occupy a substantial fraction of free ribosomes, leaving them unavailable for the translation of other proteins. The effects may be most dramatic during development, when a large number of proteins need to be synthesized during a relatively short period of time, and may be particularly strong in the case of Adh, because it is expressed at very high levels, accounting for 1–2% of the total translational activity in wild-type adult flies (BENYAJATI et al. 1980 ).

Consistent with the above interpretation is the observation that ADH allozyme type does not significantly affect developmental rate (Table 2). Thus, the developmental slowdown cannot be attributed to differences in in vivo ADH activity. Furthermore, 1762–1769-S and wt-F transformants, which are nearly identical in their levels of ADH activity (Fig 2 and Fig 3), show a highly significant difference in developmental rate. This indicates that the developmental slowdown can only be caused by differences in Adh mRNA or protein concentration, not by differences in ADH activity. Also consistent with our hypothesis is the observation that the developmental slowdown occurs almost entirely between hatching and pupation. Adh is not expressed in the developing embryo, and we find no difference in the hatching times among our four different transformant types (Table 1 and Table 2). Adh expression begins during the first instar larval stage and increases throughout larval development (Fig 3; CORBIN and MANIATIS 1989 ). This is the time when we most expect to see a developmental slowdown in 1762–1769 flies relative to wild type, and this is indeed where the slowdown occurs. We did not observe a difference in the length of pupation among any of our transformed lines. This is expected, however, because Adh is not expressed in pupae (CORBIN and MANIATIS 1989 ). The highest levels of Adh expression are seen in adult flies (Fig 2). As stated above, we have not observed any negative effects of increased Adh expression in adult flies. However, in experiments where females are placed in vials with males immediately after emerging, not aged for 2–5 days as in our experimental protocol, there is a significant reduction in the number of eggs laid by 1762–1769-F females relative to wt-F females over the first 1–2 days of egg laying (J. PARSCH, unpublished results).

Our results indicate that the underlying mechanisms for increases in enzymatic activity need to be considered in models of enzyme evolution. For example, simple models based on metabolic control theory, such as that of HARTL et al. 1985 , consider saturation kinetics in unbranched metabolic pathways in which fitness is proportional to flux through the pathway. In such a situation, the fitness of an organism is a function only of the enzymatic activity, whether the result of structural or regulatory mutations. But when deleterious pleiotropic effects associated with the metabolic cost of transcription/translation occur, these must also be taken into account. In this case, structural and regulatory changes are not equivalent. HARTL et al. 1985 also present rough estimates of selection coefficients (s) associated with changes in Adh activity. For a twofold reduction in activity, such as that observed in ADH-null heterozygotes, they estimate s = 4.52 x 10^-3 against null heterozygotes relative to wild-type homozygotes. Due to the saturation curve associated with enzyme kinetics, a twofold increase in activity would result in a smaller, positive increase in the selection coefficient [(s = 9.28 x 10^-4 under the model of HARTL et al. 1985 ]. Since the Adh 3' UTR motif is conserved even among species, we conclude that selection pressure against increased levels of Adh expression through alteration in the 3' UTR motif must be even greater (s > 9.28 x 10^-4). This is because the selective disadvantage of the twofold increase in Adh expression must outweigh any selective advantage associated with the twofold increase in ADH activity. Thus in the case of Drosophila ADH, expression levels may have a greater impact on fitness than do activity levels.

Another arena in which structural and regulatory changes are often regarded as functionally equivalent is in general discussions of the role of the two types of mutation in adaptive evolution. In bacteria, the importance of regulatory mutations in the adaptation to new carbon sources has been demonstrated experimentally (reviewed by WILSON 1977 ). Similarly, large interspecific differences in intestinal lysozyme concentration in mice are determined by a single regulatory locus (HAMMER and WILSON 1987 ). Even in Drosophila Adh, the 1 intronic polymorphism is associated with an 15–20% difference in ADH activity (LAURIE and STAM 1994 ) and shows a geographical cline in frequency stronger than that of the ADH allozymes (BERRY and KREITMAN 1993 ). These examples demonstrate that adaptive changes in enzymatic activity may be the result of regulatory mutations. To account for the lack of correlation between molecular and morphological evolution in some lineages, WILSON 1977 proposed that regulatory mutations play a primary role in adaptation. More recently, HALL 1999 has argued that selective constraints are too poorly understood to predict which genes, or which mutations in those genes, are most likely to evolve a particular novel function. Our example of the 3' UTR motif in Drosophila Adh is a case in point: it is an example of a regulatory sequence that is more constrained in evolution than the coding sequence it regulates.

	FOOTNOTES

¹ Present address: Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ 85721.

	ACKNOWLEDGMENTS

We thank Carlos Bustamante and Jeff Townsend for helpful discussions and advice on statistical analysis. This research was supported in part by National Institutes of Health grants GM02150 to D.L.H. and GM58405 to W.S.

Manuscript received February 18, 2000; Accepted for publication May 15, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AQUADRO, C. F., S. F. DESSE, M. M. BLAND, C. H. LANGLEY, and C. C. LAURIE-AHLBERG, 1985  Molecular population genetics of the alcohol dehydrogenase gene region of Drosophila melanogaster.. Genetics 114:1165-1190[Abstract/Free Full Text].

ASHBURNER, M., 1989 Drosophila: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

ASHBURNER, M., 1998  Speculations on the subject of alcohol dehydrogenase and its properties in Drosophila and other flies. BioEssays 20:949-954[Medline].

BEGUN, D. J., A. J. BETANCOURT, C. H. LANGLEY, and W. STEPHAN, 1999  Is the fast/slow allozyme variation at the Adh locus of Drosophila melanogaster an ancient balanced polymorphism? Mol. Biol. Evol. 16:1816-1819[Medline].

BENYAJATI, C., N. WANG, A. REDDY, E. WEINBERG, and W. SOFER, 1980  Alcohol dehydrogenase in Drosophila: isolation and characterization of messenger RNA and cDNA clone. Nucleic Acids Res. 23:5649-5667.

BERRY, A. and M. KREITMAN, 1993  Molecular analysis of an allozyme cline: alcohol dehydrogenase in Drosophila melanogaster on the east coast of North America. Genetics 134:869-893[Abstract].

CHAKIR, M., O. PERIDY, P. CAPY, E. PLA, and J. DAVID, 1993  Adaptation to alcoholic fermentation in Drosophila: a parallel selection imposed by environmental ethanol and acetic acid. Proc. Natl. Acad. Sci. USA 90:3621-3625[Abstract/Free Full Text].

CHOUDHARY, M. and C. C. LAURIE, 1991  Use of in vitro mutagenesis to analyze the difference in Adh expression associated with the allozyme polymorphism in Drosophila melanogaster.. Genetics 129:481-488[Abstract].

CORBIN, V. and T. MANIATIS, 1989  The role of specific enhancer-promoter interactions in the Drosophila Adh promoter switch. Genes Dev. 3:2191-2200[Abstract/Free Full Text].

CURTIS, D., R. LEHMANN, and P. D. ZAMORE, 1995  Translational regulation in development. Cell 81:171-178[Medline].

FLETCHER, T. S., F. J. AYALA, D. R. THATCHER, and G. K. CHAMBERS, 1978  Structural analysis of the ADHs electromorph of Drosophila melanogaster.. Proc. Natl. Acad. Sci. USA 75:5609-5612[Abstract/Free Full Text].

HALL, B. G., 1999  Toward an understanding of evolutionary potential. FEMS Microbiol. Lett. 178:1-6.

HAMMER, M. F. and A. C. WILSON, 1987  Regulatory and structural genes for lysozymes of mice. Genetics 115:521-533[Abstract/Free Full Text].

HARTL, D. L., D. E. DYKHUIZEN, and A. M. DEAN, 1985  Limits of adaptation: the evolution of selective neutrality. Genetics 111:655-674[Abstract/Free Full Text].

HUDSON, R. R., M. KREITMAN, and M. AGUADÉ, 1987  A test of neutral molecular evolution based on nucleotide data. Genetics 116:153-159[Abstract/Free Full Text].

KREITMAN, M., 1983  Nucleotide polymorphism at the alcohol dehydrogenase locus of Drosophila melanogaster.. Nature 304:412-417[Medline].

KREITMAN, M. and R. R. HUDSON, 1991  Inferring the evolutionary histories of the Adh and Adh-dup loci in Drosophila melanogaster from patterns of polymorphism and divergence. Genetics 127:565-582[Abstract].

LAI, E. C. and J. W. POSAKONY, 1997  The Bearded box, a novel 3' UTR sequence motif, mediates negative post-transcriptional regulation of Bearded and Enhancer of split complex gene expression. Development 124:4847-4856[Abstract].

LAI, E. C., C. BURKS, and J. W. POSAKONY, 1998  The K box, a conserved 3' UTR sequence motif, negatively regulates accumulation of Enhancer of split complex transcripts. Development 125:4077-4088[Abstract].

LAURIE, C. C. and L. F. STAM, 1994  The effect of an intronic polymorphism on alcohol dehydrogenase expression in Drosophila melanogaster.. Genetics 138:379-385[Abstract].

LAURIE, C. C., J. T. BRIDGHAM, and M. CHOUDHARY, 1991  Associations between DNA sequence variation and variation in expression of the Adh gene in natural populations of Drosophila melanogaster.. Genetics 129:489-499[Abstract].

LAURIE-AHLBERG, C. C. and L. F. STAM, 1987  Use of P-element-mediated transformation to identify the molecular basis of naturally occurring variants affecting Adh expression in Drosophila melanogaster.. Genetics 115:129-140[Abstract/Free Full Text].

LEVITEN, M., E. C. LAI, and J. W. POSAKONY, 1997  The Drosophila gene Bearded encodes a novel small protein and shares 3' UTR sequence motifs with multiple Enhancer of split complex genes. Development 124:4039-4051[Abstract].

LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR, and R. J. RANDALL, 1951  protein measurements with the Folin phenol reagent. J. Biol. Chem. 193:265-275[Free Full Text].

MARONI, G., 1978  Genetic control of alcohol dehydrogenase levels in Drosophila.. Biochem. Genet. 16:509-523[Medline].

MERCOT, H., D. DEFAYE, P. CAPY, E. PLA, and J. R. DAVID, 1994  Alcohol tolerance, ADH activity, and ecological niche of Drosophila species. Evolution 48:746-757.

OAKESHOTT, J. G., J. B. GIBSON, P. R. ANDERSON, W. R. KNIBB, and D. G. ANDERSON et al., 1982  Alcohol dehydrogenase and glycerol-3-phosphate dehydrogenase clines in Drosophila melanogaster on different continents. Evolution 36:86-96.

OEHLERT, G. W., and C. BINGHAM, 1998 MacAnova: an interactive program for statistical analysis and matrix algebra, v. 4.06. Available at http://www.stat.umn.edu/~gary/macanova/macanova. home.html

PARSCH, J., S. TANDA, and W. STEPHAN, 1997  Site-directed mutations reveal long-range compensatory interactions in the Adh gene of Drosophila melanogaster.. Proc. Natl. Acad. Sci. USA 94:928-933[Abstract/Free Full Text].

PARSCH, J., W. STEPHAN, and S. TANDA, 1999  A highly conserved sequence in the 3'-untranslated region of the Drosophila Adh gene plays a functional role in Adh expression. Genetics 151:667-674[Abstract/Free Full Text].

PATTON, J. S., X. V. GOMES, and P. K. GEYER, 1992  Position-independent germline transformation in Drosophila using a cuticle pigmentation gene as a selectable marker. Nucleic Acids Res. 20:5859-5860[Free Full Text].

ROBERTSON, H. M., C. R. PRESTON, R. W. PHILLIS, D. M. JOHNSON-SCHLITZ, and W. K. BENZ et al., 1988  A stable genomic source of P element transposase in Drosophila melanogaster.. Genetics 118:461-470[Abstract/Free Full Text].

ROSS, J., 1996  Control of messenger RNA stability in higher eukaryotes. Trends Genet. 12:171-175[Medline].

RUBIN, G. M. and A. C. SPRADLING, 1982  Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353[Abstract/Free Full Text].

SHORESH, M., S. ORGAD, O. SHMUELI, R. WERCZBERGER, and D. GELBAUM et al., 1998  Overexpression Beadex mutations and loss-of-function heldup-a mutations in Drosophila affect the 3' regulatory and coding components, respectively, of the Dlmo gene. Genetics 150:283-299[Abstract/Free Full Text].

SIEGAL, M. L. and D. L. HARTL, 1999  An experimental test for lineage-specific position effects on alcohol dehydrogenase (Adh) genes in Drosophila.. Proc. Natl. Acad. Sci. USA 95:15513-15518[Abstract/Free Full Text].

SPRADLING, A. C. and G. M. RUBIN, 1982  Transposition of cloned P-elements into Drosophila germ line chromosomes. Science 218:341-347[Abstract/Free Full Text].

ST. JOHNSTON, D., 1995  The intracellular localization of messenger RNAs. Cell 81:161-170[Medline].

STAM, L. F. and C. C. LAURIE, 1996  Molecular dissection of a major gene effect on a quantitative trait: the level of alcohol dehydrogenase expression in Drosophila melanogaster.. Genetics 144:1559-1564[Abstract].

SULLIVAN, D. T., P. W. ATKINSON, and W. T. STARMER, 1990  Molecular evolution of the alcohol dehydrogenase genes in the genus Drosophila. Evol. Biol. 24:107-147.

WILSON, A. C., 1977  Biochemical evolution. Annu. Rev. Biochem. 46:573-639[Medline].

This article has been cited by other articles:

				H.-J. Shih and C. D. Jones Patterns of Amino Acid Evolution in the Drosophila ananassae Chimeric Gene, siren, Parallel Those of Other Adh-Derived Chimeras Genetics, October 1, 2008; 180(2): 1261 - 1263. [Abstract] [Full Text] [PDF]

				C. D. Jones and D. J. Begun Parallel evolution of chimeric fusion genes PNAS, August 9, 2005; 102(32): 11373 - 11378. [Abstract] [Full Text] [PDF]

				U. H. Frey, H. Alakus, J. Wohlschlaeger, K. J. Schmitz, G. Winde, H. G. van Calker, K.-H. Jockel, W. Siffert, and K. W. Schmid GNAS1 T393C Polymorphism and Survival in Patients with Sporadic Colorectal Cancer Clin. Cancer Res., July 15, 2005; 11(14): 5071 - 5077. [Abstract] [Full Text] [PDF]

				H. Araki, S. Yoshizumi, N. Inomata, and T. Yamazaki Genetic Coadaptation of the Amylase Gene System in Drosophila melanogaster: Evidence for the Selective Advantage of the Lowest AMY Activity and of Its Epistatic Genetic Background J. Hered., July 1, 2005; 96(4): 388 - 395. [Abstract] [Full Text] [PDF]

				J. Parsch Functional Analysis of Drosophila melanogaster Gene Regulatory Sequences by Transgene Coplacement Genetics, September 1, 2004; 168(1): 559 - 561. [Abstract] [Full Text] [PDF]

				J. F. Baines, J. Parsch, and W. Stephan Pleiotropic Effect of Disrupting a Conserved Sequence Involved in a Long-Range Compensatory Interaction in the Drosophila Adh Gene Genetics, January 1, 2004; 166(1): 237 - 242. [Abstract] [Full Text] [PDF]

				J. Duan, M. S. Wainwright, J. M. Comeron, N. Saitou, A. R. Sanders, J. Gelernter, and P. V. Gejman Synonymous mutations in the human dopamine receptor D2 (DRD2) affect mRNA stability and synthesis of the receptor Hum. Mol. Genet., February 1, 2003; 12(3): 205 - 216. [Abstract] [Full Text] [PDF]