Journal of Heredity Advance Access originally published online on July 1, 2005
Journal of Heredity 2005 96(5):576-581; doi:10.1093/jhered/esi076
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brief Communication |
Inferences on the Role of Insertion in a Mutation Accumulation Experiment with Drosophila melanogaster Using RAPDs
From the Departamento de Producción Animal, Facultad de Veterinaria, Universidad Complutense, 28040 Madrid (Salgado and Nieto); and Departamento de Mejora Genética Animal, INIA, 28040 Madrid (Toro); and Departamento de Genética, Facultad de Biología, Universidad Complutense, 28040 Madrid (López-Fanjul and García-Dorado)
Address correspondence to A. García-Dorado at the address above, or e-mail: augardo{at}bio.ucm.es.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
The genetic variability for RAPDs band pattern was studied in a set of 157 mutation accumulation (MA) lines of Drosophila melanogaster. These MA lines were derived from the same isogenic base population and subsequently maintained by full-sib mating during 132 generations. The ancestral pattern of the original isogenic base can be unambiguously established as the consensus pattern of the MA lines and, because these lines are expected to be homozygous, dominance for band pattern is not a concern. Only repeatable changes in band pattern were considered. The number of ancestral bands detected implies that nine-nucleotide targets are enough for repeatable PCR amplification. Compared with the ancestral pattern, one MA line lost one band and two MA lines gained a new one. These results can be accounted for by the insertion of transposable elements occurring at a rate 0.07 < i < 0.21 per whole haploid genome and generation. This range is typical for Drosophila and consistent with the previously observed mobility for the roo family, supporting the generality of previous estimates of spontaneous mutation rates for morphological and fitness traits based on these MA lines. The sequence of one of the new bands suggests that the Idefix family is also active in the lines.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
The rate of occurrence and the nature of spontaneous mutations are of great evolutionary relevance, and an important effort has being devoted to its study from different methodological approaches. Pooling different Drosophila estimates, Keightley and Eyre-Walker (1999) inferred a 7.5 x 10–9 rate for electrophoretic band-morph mutation in enzymatic loci with an average length of 400 codons. Taking into account that about 30% of all the amino acid changes were detectable and that only about 2/3 of the base pair substitutions result in an amino acid change, gives a per nucleotide mutation rate u = 7.5 x 10–7/(2 x 400 x 0.3) = 3 x 10–9. Furthermore, in Drosophila melanogaster, insertion of transposable elements have been estimated to occur at a rate between 0.1 and 0.2 per haploid genome and generation (Maside et al. 2000; Nuzhdin and Mackay 1995).
Random amplification of polymorphic DNA (RAPD) by the polymerase chain reaction (PCR), using a single short oligonucleotide primer, provides a simple method to detect polymorphisms (Welsh and McClelland, 1990; Williams et al. 1990). Short DNA fragments are size-fractionated by agarose gel electrophoresis, and in our case, the ancestral band pattern of the original isogenic line is unambiguously established as the consensus pattern of the analyzed mutation accumulation (MA) lines. Thus, the technique allows the detection of mutations that induce repeatable modifications in band pattern (presence of a new band or absence of an ancestral one), either by a change in the target that is recognized by the primer or by rearrangement of sequences involving targets.
In this article, we report an analysis of the rate of mutation for RAPD markers in a set of D. melanogaster lines after 132 generations of MA. In previous studies, these lines have shown typical mutational variances for several morphological traits (see García-Dorado et al. 1999 and references therein), but the rate of mutational viability decline was substantially lower than that originally reported by Mukai and co-workers (Mukai et al. 1972; Chavarrías et al. 2001, García-Dorado and Caballero 2002). The nature of such difference is still a mater of debate, and it has been suggested that it could be due to different transposition rates (Keightley and Eyre-Walker 1999). However, insertions of TE of the roo family have occurred at a large per generation rate ie = 0.029 in the euchromatin of our MA lines (Masside et al. 2001), suggesting normal transposition rates in these lines. If, as the latter authors conclude, 
55% of the TE genome is located in the heterochromatine, the above value implies a rate i 
0.065 for the whole haploid genome. Here we provide complementary information based on mutational changes on RAPD band patterns. This will contribute to the understanding of the molecular basis of spontaneous mutation and of the morphological and fitness genetic variability accumulated in our MA lines.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
MA Lines
Two hundred full-sib mating lines were derived from a D. melanogaster line isogenic for all chromosomes and carrying the recessive eye-color marker sepia (se) as an indicator against exogenous contamination (Caballero et al. 1991; Chavarrías et al. 2001). After 132 generations of spontaneous MA, 157 lines survived and were screened for RAPD mutations.
DNA Extraction and PCR Amplification
Homogenates were prepared by grinding a single female fly per line in 160 µl extraction buffer (10 mM Tris, 60 mM NaCl, 5% sucrose, 10 mM EDTA, pH 7.8). A second buffer (300 mM Tris, 1.25% SDS, 5% sucrose, 10 mM EDTA, pH 8) was added, and the mixture was incubated at 65°C with agitation for 30 min. After this, 60 µl of potassium acetate was added, the samples were kept for 20 min at –20°C and, subsequently, they were spun at 12,000 x g. The purified DNA was resuspended in 100 µl H2O (2 µl used in each reaction).
Total reaction volumes of 25 µl were used, containing 10 mM Tris, pH 8.3, 50 mM KCl, 2 mM MgCl2, 0.001% gelatin, 100 µM each dNTP (dATP, dCTP, dGTP, and dTTP), 5 pmol of a single 10-base primer, 25 ng genomic DNA, and 0.4 U of Taq-DNA polimerase (Dynazyme F-500L). This procedure was indicated by Williams et al. (1990).
The primers used in this study were Kit C and Kit F Operon oligonucleotides (except OPF7, OPF11, OPF12, OPF18, and OPF20). PCR reactions were conducted on a MJ Thermal Cycler (MJ Research). Amplification was performed according to the following program: 6 min at 94°C followed by 45 cycles of 1 min at 94°C, 1 min at 36°C, and 2 min at 72°C, with a final extension step of 6 min at 72°C. A fragment is generated by PCR when one double-strand DNA has the sequence matching the primer in both opposite strands (each in the appropriate matching orientation) within a distance that is readily amplifiable by PCR, say below about m = 2,000 pb (Erlich 1989). PCR products were visualized under UV light after electrophoresis on 1.5% agarose gels and staining with ethidium bromide. Ready-Load 100 bp DNA Ladder (Life Technologies) was used as a molecular size standard.
Drosophila RAPD fragments reproducibility in replicate reactions per individual has been tested by Pérez et al. (1998). They found a moderate repeatability of banding patterns and of homologous bands, which renders this technique suboptimal for measuring variability in segregating populations. However, our lines have a common isogenic origin, and the ancestral band pattern can be unambiguously established, so that permanent changes in the band pattern of a line caused by fixation of a new mutation can be reliably detected through repeated assays. Thus, we have only considered those band differences consistently present in two replicate assays. One of the two new bands detected corresponded to a line from which no material was available for further analysis. The other band was again obtained using the same RAPD technique, cloned using a TA cloning vector pCR2.1 and sequenced using an Applied Biosystems Automated DNA Sequencer.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
A total of 35 (10-mer) oligonucleotide primers were used for genomic DNA amplification. On the whole, 190 ancestral bands were obtained. They were between 300 and 2,050 bp in size, with average length 929 bp, as expected for a technique detecting random fragments with length below a maximum m = 2,000 bp. An approximate expected length l = 1,000 will be used thereafter.
As shown in the Appendix, the observed number of bands is larger than expected from targets exactly matching the primer. This is in agreement with other empirical observations (Pérez et al. 1998). Thus we estimate the probability (p) of a matching sequence from the observed number of ancestral bands, which gives p = 3.99 x 10–6 in close agreement with that expected if one mismatch is allowed in a precise position (p = 1/49 = 3.81 x 10–6).
Out of 157 lines screened, three showed a band pattern that differed from the rest (Table 1). An extra band was present in lines 42 and 72 for the oligonucleotide OPC-04, and the absence of a band for oligonucleotide OPC-14 was observed in line 100.
|
Mutation Rate Estimates
Unfortunately, there is no way to obtain separate estimates for the rate of TE insertions (the prevailing transposition activity) and that of point mutation from the observed rates of band loss and gain. Thus, we will consider both processes separately.
Expectations from Point Mutation.
Because point mutation is assumed to be random and does not change genome size, it induces the same rates of band loss (du) and band gain (au) per gamete and generation, that is, au = du = 2rxu (see Appendix), where r is the number of relevant nucleotides in a target sequence, x is the number of ancestral bands detected, and u is the per nucleotide mutation rate. As explained, we will assume r = 9. Thus, for a mutation rate u = 3 x 10–9 (see Introduction) the rate is au = 10–5, so that we expect 0.21 bands lost and the same number of new bands in the set of 157 lines after 132 generations of mutation accumulation. Assuming that band gains and losses are independent and that the total number of changes is Poisson distributed, the probability of obtaining three or more changes through point mutations is 0.009.
Inferences on Transposition Activity from Band Loss.
The expected rate of band loss due to insertions occurring inside ancestral bands is
 |
< 0.05) a total insertion rate i 
0.21 (ie 0.09 for the euchromatin). Furthermore, using the i value observed for the roo family (ie = 0.029), the expected number of bands lost in the 157 lines after 132 generations is 157 x 132 x di = 1.51. Therefore, our observed number of bands lost (one) could be fully explained by TE insertions.
Inferences on Transposition Activity from New Bands.
The more direct way a new band can arise through transposition is the insertion close to a target of a TE including the complementary target. The probability of this event has been derived in the Appendix assuming that the presence of targets in the TE copies is random. This allows to compute an a priori expectation for this probability in the absence of precise information about the presence of targets for the 35 primers used in the extant TE copies. Thus, the average rate of accumulation of new bands per generation and primer is
 |
 |
The sequence of the new band detected for line 72 using oligo OPC-04 was found to be a 569-bp fragment of the pol-protein of the Idefix LTR-retroelement (95% coincidence with the canonical element). In the canonical sequence of Idefix (AJ009736 [GenBank] ), the sequence matching the band was flanked by sets of nucleotides that were very similar to the nine-nucleotide targets required for PCR amplification using oligo OPC-04 (nucleotides matching are in boldface, and mismatches are in italics):
- Sequence in Idefix: agcatcaac-band sequence-gtagatgcc
- Sequence required for amplification: cgcatctac-band sequence-gtagatgcg
Unfortunately, no flies or DNA were left to obtain the sequence for the new band in line 42. Interestingly, out of the 35 oligos used in this experiment, OPC-4 detected the two new bands found, both of them of similar estimated size according to their electrophoretic mobility. However, it is highly unlikely that both bands correspond to the same segment of Idefix if this requires the occurrence of the same point mutation in both lines.
Of course, the new band in line 42 could also be associated to the activity of other elements. We have found several transposable elements bearing the exact 9-nucleotide target for OPC-04 (mdg-1, s2, Cr1a, 1360, and mariner2 elements, see Release 3 transposons sequence in the BDGP Web site). In two cases (mdg-1 and 1360) these targets were placed near an extreme, so that a new band might be induced if they became close to another target after transposition, but mdg-1 seems to be inactive in these lines (Domínguez and Albornoz personal communication; Maside et al. 2001). On the other hand, the insertion rate reported for the roo family (Maside et al. 2001; ie = 0.029, i = 0.065) gives an a priori prediction of three expected new bands (number of new bands assumed to be Poisson distributed), and several roo sequences have targets matching the eight last 3' primer's positions (see transposon sequences in Release 3 in the DGBP Web site), so that it could be that some copy in our lines matched nine positions. Thus, both roo and 1360 TE families could be responsible for the gain of the band in line 42.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
Point mutation occurring at a rate 3 x 10–9 per nucleotide is expected to cause 0.42 band changes on the whole. Therefore, it is unlikely to be responsible for the between line differences that we have detected in RAPD band patterns. Short deletions and insertions occur at a rate of about 10% and 3% the point mutation rate, respectively (Blumenstiel et al. 2002) and are expected to contribute substantially less than point mutations to these differences. However, band loss and gain can be convincingly accounted for by TE insertions.
Our observation of one band lost suggests a total insertion rate i = 0.043 (or about ie = 0.019 in the euchromantine), and is consistent with i up to 0.21 (ie = 0.09). The new band found in line 42 a priori suggests i = 0.022 (ie = 0.010) and is consistent with i up to 0.10 (ie = 0.045). Both observations are consistent with the insertion activity of the roo family (ie = 0.029) described in our lines by Maside et al. (2001). However, it should be noted that inferences from the rate of new band acquisition are subject to large sampling error, depending on the presence of appropriate targets in specific terminal fragments of the TE families that are active in the lines considered. Thus, we should be more confident on the information from band loss.
The precise origin of the new band in line 42 cannot be ascertained because the genetic material to obtain its sequence became unavailable. Elements of the 1360 TE family, whose mobility has not been previously studied in our lines, have been found to bear exact nine-nucleotide targets for OPC-04 in appropriate position to generate new bands after insertion, whereas several others, including roo, bear eight-nucleotide targets.
The new band in line 72 was sequenced and found to correspond to an inner fragment of the LTR retroelement Idefix, which was included in none of the studies already cited. It seems to be caused by mutation producing appropriate targets rather than to rearrangement of targets produced by insertion, but it is most likely associated with transposition activity that will increase the mutation rate within the TE sequence. Furthermore, it has been suggested that amplification could also depend on structural features (Pérez et al. 1998). Thus, even if nine matches are required on the average, fewer matches could occasionally be enough for amplification depending on the genomic location. Therefore, transposition could also induce a new band due to the insertion of a potential band into a location more accessible for amplification. In any case, the new band in line 72 suggests transposition activity involving the Idefix family.
Therefore, our results suggest an euchromatic haploid insertion rate about ie = 0.04 per haploid genome and generation, although it is consistent with rates up to 0.09. Considering that Maside et al. (2001) have described in the same lines a large insertion rate in the euchromatin due to roo activity (ie 0.029), and that at least the Idefix element seems also to be active, it seems reasonable to conclude that the true euchromatic insertion rate should be 0.03 < ie < 0.09 (0.07 < i < 0.21 for the whole genome).
In our MA lines, TE activity has been studied for 14 families (Domínguez and Albornoz 1996; Maside et al. 2001), but mobility was only found for the roo family (ie 0.029). By extrapolating the mobility of roo to the whole TE set, Maside et al. inferred a total insertion rate ie 0.09 in the euchromatine, although this extrapolation is subject to very important sampling error due to the dramatic between-family variability for TE activity. This inferred transposition rate (ie 0.09) is similar to the one reported by Maside et al. (2000) for chromosome II copies that had been maintained in the heterozygous condition through a Mukai-like design (ie = 0.10), and is somewhat smaller than the transposition rate reported by Nuzhdin and Mackay (1995, ie = 0.20). Our low rate of band loss suggests that those estimates must be taken as an upper limit.
Our MA lines have shown typical mutational variance for morphological traits (García-Dorado et al. 1999) but a small mutational viability decline, both in competitive and noncompetitive conditions (Chavarrías et al. 2001). In contrast, moderate viability declines have been observed in Fry et al. (1999) and Ohnishi (1977) MA experiments and large ones in the two classical MA experiments carried out by Mukai (1964; Mukai et al.1972). As a consequence, the classical Bateman-Mukai method (Mukai et al. 1972) gives a low rate of mildly deleterious mutation for our MA lines but higher ones for Mukai's experiments. The causes of these discrepancies are likely to remain unknown, but it has been argued that they could be spurious to some extent (García-Dorado et al. 1999), that they could be due to different transposition rates in the different genetic backgrounds (Keightley and Eyre-Walker 1999), and that the large decline in the former Mukai's experiment could have been due to accelerated transposition (García-Dorado and Caballero 2002). Our results corroborate that our MA lines show typical transposition rate, giving no support to concerns about the generality of their small mutational viability decline.
|
Appendix |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
Assume that amplification of a fragment by PCR requires that a DNA strand includes the target for a primer in the appropriate orientation, and that the complementary strand has another target within a distance smaller than m = 2000 bp in the appropriate direction. This means that a particular single DNA strand must have the target and, at a distance smaller than 2,000 bp in direction 3'
5', it must also have the sequence complementary to that target (i.e., the properly oriented primer sequence if no mismatch is allowed). This holds symmetrically in the complementary strand in the opposite direction. Thus, to determine fragment formation only one strand must be considered.
The expected number of targets for a primer in a whole haploid genome (considering a single DNA strand) with a length of g bp is n = p x g, where p is the probability that a sequence randomly sampled in that genome is a target. For any given target, a band is obtained when the distance y to the next complementary target is below m = 2,000 nucleotides (for a related approach, see Clark and Lanigan 1993) and the distance z to the next direct target is smaller than y, both y and z being random variables exponentially distributed with mean 1/p. Thus, the expected number of bands per primer is
 |
y)). Thus,
 |
If the sequence is random with respect to the distribution of targets, a 10-nucleotide match is expect with probability 1/410 = 9.54 x 10–7. Considering g = 1.7 x 108 (Drake et al. 1998), this gives an expected total number of ancestral bands for the 35 primers analyzed x = 10.8, which is inconsistent with the empirically observed number (190). Thus, we will estimate p from the observed average number of ancestral bands per primer (190/35 = 5.43) as
 |
Rates of Band Loss and Gain from TE Insertions
When a TE is inserted into an ancestral band, the fragment between the targets becomes too large to be amplified, and that band is lost. Thus, the rate of band loss due to TE insertion is
 |
Assume that elements of a TE family have the target for a primer at a distance of y nucleotides (y < m) of the appropriate extreme for inducing a new band after insertion. If one of these TE is inserted, the probability that the next complementary target is at a distance y < m is
 |
 |
Therefore, assuming that the presence of targets in the whole set of elements for all TE families is random, the expected rate of new bands per primer due to TE insertions is
 |
 |
Rates of Band Loss and Gain from Point Mutation
A band will be lost when point mutation occurs in any nucleotide relevant to the recognition between the primer and the two targets flanking the fragment. Thus, the rate of band loss from point mutation per RAPD is
 |
We have used r = 9 by the reasons already given. Assuming r = 10 gives an even smaller contribution of point mutation to the observed results.
|
Acknowledgments |
|---|
We thank A. Domínguez by helpful discussion and C. López-Fernández and R. Linacero for their help in cloning a new band. This work was supported by grant BMC2002-00476 (Ministerio de Ciencia y Tecnología).
|
Footnotes |
|---|
Corresponding Editor: Ross MacIntyre
Received January 15, 2005
Accepted March 17, 2005
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendixReferences |
|---|
-
Alonso-González L, Domínguez A, and Albornoz J, 2003. Structural heterogeneity and genomic distribution of Drososphila melanogaster LTR-retrotransposons. Mol Biol Evol 20:401–409.
Blumenstiel JP, Hartl DL, and Lozovsky ER, 2002. Patterns of insertion and deletion in contrasting chromatin domains. Mol Biol Evol 19:2211–2225.
Caballero A, Toro MA, and López-Fanjul C, 1991. The response to artificial selection from new mutation in Drosophila melanogaster. Genetics 127:89–102.
Chavarrías D, López-Fanjul C, and García-Dorado A, 2001. The rate of mutation and the homozygous and heterozygous mutational effects for competitive viability: a long-term experiment with Drosophila melanogaster. Genetics 158:681–693.
Clark A, and Lanigan CMS, 1993. Prospects for estimating nucleotide divergence with RAPDs. Mol Biol Evol 10:1096–1111.[Abstract]
Dominguez A, and Albornoz J, 1996. Rates of movement of transposable elements in Drosophila melanogaster. Mol Gen Genet 251:130–138.[Web of Science][Medline]
Domínguez A, and Albornoz J, 2000. Structural instability of 297 element in Drosophila melanogaster. Genetica 105:239–248.
Drake JW, Charlesworth B, Charlesworth D, and Crow JF, 1998. Rates of spontaneous mutation. Genetics 148:1667–1686.
Erlich HA, 1989. PCR technology, principles and applications for DNA amplification Stockton Press.
Fry JD, Keightley PD, Heinsohn SL, and Nuzhdin SV, 1999. New estimates of the rates and effects of mildly deleterious mutation in Drosophila melanogaster. Proc Natl Acad Sci USA 96:574–579.
García-Dorado A, López-Fanjul C, and Caballero A, 1999. Properties of spontaneous mutations affecting quantitative traits. Genet Res 74:341–350.[CrossRef][Web of Science][Medline]
García-Dorado A, and Caballero A, 2002. The mutational rate of Drosophila viability decline: tinkering with old data. Genet Res 80:99–105.[CrossRef][Web of Science][Medline]
Keightley PD, and Eyre-Walker A, 1999. Terumi Mukai and the riddle of deleterious mutation rates. Genetics 153:515–523.
Maside X, Assimacopoulos S, and Charlesworth B, 2000. Rates of movement of transposable elements on the second chromosome of Drosophila melanogaster. Genet Res 75:275–284.[CrossRef][Web of Science][Medline]
Maside J, Bartolomé C, Assimacopoulos S, and Charlesworth B, 2001. Rates of movement and distribution of transposable elements in Drosophila melanogaster: in situ hybridization vs Southern blotting data. Genet Res 78:121–136.[Web of Science][Medline]
Mukai T, 1964. The genetic structure of natural populations of Drosophila melanogaster. I. Spontaneous mutation rate of polygenes controlling viability. Genetics 50:1–19.
Mukai T, Chigusa SI, Mettler LE, and Crow JF, 1972. Mutation rate and dominance of genes affecting viability in Drosophila melanogaster. Genetics 72:333–355.
Nuzhdin S, and Mackay T, 1995. The genomic rate of transposable element movement in Drosophila melanogaster. Mol Biol Evol 12:180–181.[Web of Science][Medline]
Ohnishi O, 1977. Spontaneous and ethyl methane-sulfonate induced mutations controlling viability in Drosophila melanogaster. II Homozygous effect of polygenic mutations. Genetics 87:529–545.
Pérez T, Albornoz J, and Domínguez A, 1998. An evaluation of RAPD fragment reproducibility and nature. Mol Ecol 7:1347–1357.[CrossRef][Medline]
Welsh J, and McClelland M, 1990. Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res 18:7213–7218.
Williams JGW, Kubelik AR, Livak KJ, Rafalski JA, and Tingey SV, 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531–6535.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Papaceit, V. Avila, M. Aguade, and A. Garcia-Dorado The Dynamics of the roo Transposable Element In Mutation-Accumulation Lines and Segregating Populations of Drosophila melanogaster Genetics, September 1, 2007; 177(1): 511 - 522. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||